osmocom
/
wireshark
11
0
Fork 1
Code Issues Pull Requests Projects Releases Wiki Activity
wireshark/epan/dissectors/packet-diameter.c

2667 lines
90 KiB
C
Raw Normal View History

David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
/* packet-diameter.c
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
* Routines for Diameter packet disassembly
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
*
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
* Copyright (c) 2001 by David Frascone <dave@frascone.com>
luis.ontanon@gmail.com => luis@ontanon.org svn path=/trunk/; revision=25937
2008-08-05 21:03:46 +00:00
* Copyright (c) 2007 by Luis E. Garcia Ontanon <luis@ontanon.org>
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
*
Move proto_reg_handoff... to the end of the file. Whitespace cleanup: remove trailing spaces from a few lines. svn path=/trunk/; revision=32569
2010-04-26 22:19:34 +00:00
* Support for Request-Answer tracking and Tapping
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
* introduced by Abhik Sarkar
*
change a whole bunch of ethereal into wireshark svn path=/trunk/; revision=18196
2006-05-21 04:49:01 +00:00
* Wireshark - Network traffic analyzer
* By Gerald Combs <gerald@wireshark.org>
Fix the copyright notice to reflect the fact that Gerald holds the copyright on Ethereal as a whole. svn path=/trunk/; revision=3473
2001-05-27 21:38:46 +00:00
* Copyright 1998 Gerald Combs
*
dissectors: use SPDX identifiers. Change-Id: I92c94448e6641716d03158a5f332c8b53709423a Reviewed-on: https://code.wireshark.org/review/25756 Petri-Dish: Dario Lombardo <lomato@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-02-12 11:23:27 +00:00
* SPDX-License-Identifier: GPL-2.0-or-later
Update RFC and I-D references. Give draft-frascone-xml-dictionary for the dictionary syntax. Give RFC numbers and I-D names, rather than URLs to the plain text versions of them; that way, you can more easily go to the HTML versions, which give more information. The "more information" for I-Ds may include later drafts and the final RFCs, and for RFCs may give newer RFCs obsoleting the older ones; add the RFC numbers for the I-Ds and the RFC number for the current Diameter RFC. Change-Id: If01ea341af3ae892755a243bd1dd66acfdfd5062 Reviewed-on: https://code.wireshark.org/review/23086 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2017-08-15 18:31:05 +00:00
*
From Anders Broman: fix the value strings for decoding of AVP values, add the T flag, make the result code and session ID named fields, and add some more field types. svn path=/trunk/; revision=10430
2004-03-21 23:10:31 +00:00
* References:
Update RFC and I-D references. Give draft-frascone-xml-dictionary for the dictionary syntax. Give RFC numbers and I-D names, rather than URLs to the plain text versions of them; that way, you can more easily go to the HTML versions, which give more information. The "more information" for I-Ds may include later drafts and the final RFCs, and for RFCs may give newer RFCs obsoleting the older ones; add the RFC numbers for the I-Ds and the RFC number for the current Diameter RFC. Change-Id: If01ea341af3ae892755a243bd1dd66acfdfd5062 Reviewed-on: https://code.wireshark.org/review/23086 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2017-08-15 18:31:05 +00:00
*
* RFC 3588, "Diameter Base Protocol" (now RFC 6733)
* draft-ietf-aaa-diameter-mobileip-16, "Diameter Mobile IPv4 Application"
* (now RFC 4004)
* draft-ietf-aaa-diameter-nasreq-14, "Diameter Network Access Server
* Application" (now RFC 4005)
* drafts/draft-ietf-aaa-diameter-cc-03, "Diameter Credit-Control
* Application" (now RFC 4006)
* draft-ietf-aaa-diameter-sip-app-01, "Diameter Session Initiation
* Protocol (SIP) Application" (now RFC 4740)
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
* RFC 5779, "Diameter Proxy Mobile IPv6: Mobile Access Gateway and
* Local Mobility Anchor Interaction with Diameter Server"
* 3GPP TS 29.273, V15.2.0
From Anders Broman: fix the value strings for decoding of AVP values, add the T flag, make the result code and session ID named fields, and add some more field types. svn path=/trunk/; revision=10430
2004-03-21 23:10:31 +00:00
* http://www.ietf.org/html.charters/aaa-charter.html
Update RFC and I-D references. Give draft-frascone-xml-dictionary for the dictionary syntax. Give RFC numbers and I-D names, rather than URLs to the plain text versions of them; that way, you can more easily go to the HTML versions, which give more information. The "more information" for I-Ds may include later drafts and the final RFCs, and for RFCs may give newer RFCs obsoleting the older ones; add the RFC numbers for the I-Ds and the RFC number for the current Diameter RFC. Change-Id: If01ea341af3ae892755a243bd1dd66acfdfd5062 Reviewed-on: https://code.wireshark.org/review/23086 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2017-08-15 18:31:05 +00:00
* http://www.iana.org/assignments/radius-types
From Olivier Jacques: - Add a configuration option for Diameter version. Currently, the choice is between everything before draft-v16 and RFC3588. - Fix diameter-ip-address parsing depending on the Diameter version (and showing decoding errors accordingly). - Change registration of Diameter from TCP and SCTP port 1812 to TCP and SCTP port 3868 (this is according to RFC3588, section 11.5). svn path=/trunk/; revision=12121
2004-09-28 17:52:36 +00:00
* http://www.iana.org/assignments/address-family-numbers
* http://www.iana.org/assignments/enterprise-numbers
From: "Jacques, Olivier : - Added new RFC 3588 header format support (vendor-Id is replaced by application-Id) and command code has its own separated name space. The Draft v16 behaviour is maintained (through Preferences menu). - Added 3GPP IMS Cx/Dx interface decoding (new imscxdx.xml) - Fixed some issues in dictionary.xml - Refreshed Diameter applications in embedded AVPs (that do not rely on XML dictionary) svn path=/trunk/; revision=12235
2004-10-08 09:35:56 +00:00
* http://www.iana.org/assignments/aaa-parameters
From Olivier Jacques: - Add a configuration option for Diameter version. Currently, the choice is between everything before draft-v16 and RFC3588. - Fix diameter-ip-address parsing depending on the Diameter version (and showing decoding errors accordingly). - Change registration of Diameter from TCP and SCTP port 1812 to TCP and SCTP port 3868 (this is according to RFC3588, section 11.5). svn path=/trunk/; revision=12121
2004-09-28 17:52:36 +00:00
*/
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
#include "config.h"
Include files from the "epan" directory and subdirectories thereof with "epan/..." pathnames, so as to avoid collisions with header files in any of the directories in which we look (e.g., "proto.h", as some other package has its own "proto.h" file which it installs in the top-level include directory). Don't add "-I" flags to search "epan", as that's no longer necessary (and we want includes of "epan" headers to fail if the "epan/" is left out, so that we don't re-introduce includes lacking "epan/"). svn path=/trunk/; revision=4586
2002-01-21 07:37:49 +00:00
#include <epan/packet.h>
Add missing includes in order to remove exceptions.h from proto.h (next commit). svn path=/trunk/; revision=53230
2013-11-10 15:59:37 +00:00
#include <epan/exceptions.h>
Move prefs.c and prefs.h into the epan subdirectory. svn path=/trunk/; revision=12115
2004-09-27 22:55:15 +00:00
#include <epan/prefs.h>
Move the #defines for SMI Network Management Private Enterprise Codes for organizations to an <epan/sminmpec.h> header, and add in the ones that were used (without #defines) in the Diameter dissector. Merge the value_string tables for them from the Diameter and Radius dissectors into epan/sminmpec.c and merge them. Use that value_string table in the Diameter and Radius dissectors. Constify some stuff in the Diameter dissector. svn path=/trunk/; revision=12159
2004-09-30 20:27:47 +00:00
#include <epan/sminmpec.h>
Parse enterprise-numbers at run time "enterprise-numbers" is converted to tab-separated values and renamed "enterprises". Unused fields are stripped. PENs are stored in a hash table loaded at run-time. User "enterprises" file is loaded from the personal config dir. Misc make-sminmpec.pl improvements and fixes. Note: names of type "Entity (formerly ...)" have the formerly part commented out for a cleaner output. Change-Id: I60c533afbe3e399077fbf432088064471ad3e1e2 Reviewed-on: https://code.wireshark.org/review/22246 Petri-Dish: João Valverde <j@v6e.pt> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Reviewed-by: João Valverde <j@v6e.pt>
2017-06-24 13:25:41 +00:00
#include <epan/addr_resolv.h>
From Martin Mathieson: packet-diameter.c -------------------------- I completely reindented dissect_avps() before I made any changes, but when ignoring white space (in tkdiff, -w plus checking 'Ignore blanks when diffing'), its easy to see the small changes I've made: - when fail to find AVP info, show code in tree parent in decimal (as specs do) - add an expert info (undecoded, note) to indicate unknown AVP codes diameter/imscxdx.xml ------------------------------- - added 'Associated-Identities' svn path=/trunk/; revision=18641
2006-07-03 21:18:36 +00:00
#include <epan/expert.h>
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
#include <epan/tap.h>
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
#include <epan/srt_table.h>
Add export_pdu_tap to packet-diameter.c svn path=/trunk/; revision=49448
2013-05-20 18:30:32 +00:00
#include <epan/exported_pdu.h>
Sanitize epan includes svn path=/trunk/; revision=29499
2009-08-21 20:25:24 +00:00
#include <epan/diam_dict.h>
Register for Diameter's SCTP PPID. svn path=/trunk/; revision=44501
2012-08-14 20:44:58 +00:00
#include <epan/sctpppids.h>
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
#include <epan/show_exception.h>
Make the avp_str available to sub dissectors and Let the subdissector have precedence filling in the avp_item string. Use that in a couple of places. Change-Id: I1af7a1ca4c14fb56ddeaab336202e6c2a18e556b Reviewed-on: https://code.wireshark.org/review/699 Reviewed-by: Michael Mann <mmann78@netscape.net> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-16 17:57:18 +00:00
#include <epan/to_str.h>
Move wmem string utility functions to wsutil
2021-12-18 23:22:27 +00:00
#include <epan/strutil.h>
Diameter: Handle (IP)address type E164 as string The Diameter type Address hase a two byte address type family field previously only IPv4 and IPv6 was handled. Add handling of E.164 when encoded as a string.
2020-10-12 08:34:49 +00:00
#include <epan/afn.h>
Fix some dissectors doing termio (fprintf(stderr,..), g_warning()). - Use report_...failure() (in most cases). - Also: Do some misc fixes in certain disectors - re-arrange order of #includes - Fixup preferences help text Change-Id: I385f6f97257f365f53ce611df02f57f9257dc5f9 Reviewed-on: https://code.wireshark.org/review/6039 Petri-Dish: Bill Meier <wmeier@newsguy.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-12-24 15:44:03 +00:00
#include <wsutil/filesystem.h>
Clean up handling of enabled/disabled protocols/heuristic dissectors. Add a "report a warning message" routine to the "report_err" code in libwsutil, and rename files and routines appropriately, as they don't only handle errors any more. Have a routine read_enabled_and_disabled_protos() that reads all the files that enable or disable protocols or heuristic dissectors, enables and disables them based on the contents of those files, and reports errors itself (as warnings) using the new "report a warning message" routine. Fix that error reporting to report separately on the disabled protocols, enabled protocols, and heuristic dissectors files. Have a routine to set up the enabled and disabled protocols and heuristic dissectors from the command-line arguments, so it's done the same way in all programs. If we try to enable or disable an unknown heuristic dissector via a command-line argument, report an error. Update a bunch of comments. Update the name of disabled_protos_cleanup(), as it cleans up information for disabled *and* enabled protocols and for heuristic dissectors. Support the command-line flags to enable and disable protocols and heuristic dissectors in tfshark. Change-Id: I9b8bd29947cccdf6dc34a0540b5509ef941391df Reviewed-on: https://code.wireshark.org/review/20966 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2017-04-08 19:45:19 +00:00
#include <wsutil/report_message.h>
Make both the Diameter-over-TCP and Diameter-over-SCTP dissectors "new-style" dissectors, and have them reject packets that don't start with a version number of 1, in the hopes that it'll be less likely to try to dissect non-Diameter packets to/from port 1812 as Diameter and thus less likely to spew out a bunch of warnings that cause users to ask the Ethereal mailing lists what the problem is. Use "tcp_dissect_pdus()" to handle Diameter-over-TCP. svn path=/trunk/; revision=10165
2004-02-21 10:29:52 +00:00
#include "packet-tcp.h"
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
#include "packet-diameter.h"
diameter: Register the [D]TLS port Port 5868 is IANA registered for Diameter over TLS/TCP and DTLS/SCTP. Register the diameter TCP handle by name so that it can be registered to tls.
2022-07-02 01:03:05 +00:00
#include "packet-tls.h"
#include "packet-dtls.h"
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
#include "packet-e212.h"
Add a function to dissect an E.164 (MSISDN) number in UTF8 format and use it in the Diameter dissector. This new API adds a filter for the MSISDN as well as a subtree and filter for the Country Code. Change-Id: Ibcbf4b5f72178b7e4af63efa7496188d608a9de7 Reviewed-on: https://code.wireshark.org/review/3760 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-21 02:47:22 +00:00
#include "packet-e164.h"
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
Fix some trivial warnings. svn path=/trunk/; revision=54085
2013-12-14 10:29:26 +00:00
void proto_register_diameter(void);
void proto_reg_handoff_diameter(void);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
/* Diameter Header Flags */
Use port range for SCTP too. svn path=/trunk/; revision=34557
2010-10-18 15:56:06 +00:00
/* RPETrrrrCCCCCCCCCCCCCCCCCCCCCCCC */
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
#define DIAM_FLAGS_R 0x80
#define DIAM_FLAGS_P 0x40
#define DIAM_FLAGS_E 0x20
#define DIAM_FLAGS_T 0x10
#define DIAM_FLAGS_RESERVED4 0x08
#define DIAM_FLAGS_RESERVED5 0x04
#define DIAM_FLAGS_RESERVED6 0x02
#define DIAM_FLAGS_RESERVED7 0x01
#define DIAM_FLAGS_RESERVED 0x0f
#if 0 some unused #defines. svn path=/trunk/; revision=54306
2013-12-20 16:03:24 +00:00
#if 0
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
#define DIAM_LENGTH_MASK 0x00ffffffl
#define DIAM_COMMAND_MASK DIAM_LENGTH_MASK
#define DIAM_GET_FLAGS(dh) ((dh.flagsCmdCode & ~DIAM_COMMAND_MASK) >> 24)
#define DIAM_GET_VERSION(dh) ((dh.versionLength & (~DIAM_LENGTH_MASK)) >> 24)
#define DIAM_GET_COMMAND(dh) (dh.flagsCmdCode & DIAM_COMMAND_MASK)
#define DIAM_GET_LENGTH(dh) (dh.versionLength & DIAM_LENGTH_MASK)
#if 0 some unused #defines. svn path=/trunk/; revision=54306
2013-12-20 16:03:24 +00:00
#endif
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
/* Diameter AVP Flags */
#define AVP_FLAGS_P 0x20
#define AVP_FLAGS_V 0x80
#define AVP_FLAGS_M 0x40
#define AVP_FLAGS_RESERVED3 0x10
#define AVP_FLAGS_RESERVED4 0x08
#define AVP_FLAGS_RESERVED5 0x04
#define AVP_FLAGS_RESERVED6 0x02
#define AVP_FLAGS_RESERVED7 0x01
#define AVP_FLAGS_RESERVED 0x1f /* 00011111 -- V M P X X X X X */
#define DIAMETER_RFC 1
Add export_pdu_tap to packet-diameter.c svn path=/trunk/; revision=49448
2013-05-20 18:30:32 +00:00
static gint exported_pdu_tap = -1;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
/* Conversation Info */
typedef struct _diameter_conv_info_t {
Rename a variable for clarity. It's confusing to have 'pdus_tree' mean both the map of pdu_trees and the pdu trees themselves. Change-Id: Ie875798eb140b60a1309ddc0c0bf885b48c0407c Reviewed-on: https://code.wireshark.org/review/14413 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2016-03-11 02:03:52 +00:00
wmem_map_t *pdu_trees;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
} diameter_conv_info_t;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
typedef struct _diam_ctx_t {
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_tree *tree;
packet_info *pinfo;
Use epan-scoped memory for building more of the diameter table stuff. svn path=/trunk/; revision=51754
2013-09-04 21:06:56 +00:00
wmem_tree_t *avps;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
} diam_ctx_t;
typedef struct _diam_avp_t diam_avp_t;
typedef struct _avp_type_t avp_type_t;
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
typedef const char *(*diam_avp_dissector_t)(diam_ctx_t *, diam_avp_t *, tvbuff_t *, diam_sub_dis_t *);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
typedef struct _diam_vnd_t {
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
guint32 code;
diameter: change GArray into wmem_array. This change fixes a leak in packet-diameter that loads a dictionary but doesn't free all the data. Found by valgrind. ==30481== 36,656 (960 direct, 35,696 indirect) bytes in 24 blocks are definitely lost in loss record 3,417 of 3,421 ==30481== at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==30481== by 0xA7FE610: g_malloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA81422D: g_slice_alloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA7CDC44: g_array_sized_new (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0x6863743: dictionary_load (packet-diameter.c:1980) ==30481== by 0x6863743: proto_register_diameter (packet-diameter.c:2344) ==30481== by 0x71C4BA4: register_all_protocols (register.c:323) ==30481== by 0x65EEFA7: proto_init (proto.c:521) ==30481== by 0x65CD621: epan_init (epan.c:126) ==30481== by 0x115330: main (tshark.c:1220) Change-Id: I3c0d19e1accab415355aa0f50c598f0c83356985 Reviewed-on: https://code.wireshark.org/review/13821 Petri-Dish: Dario Lombardo <lomato@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-02-18 10:30:41 +00:00
wmem_array_t *vs_avps;
Make value_string_ext const-correct. There is still some const-incorrect usage of them but those can be ironed out after this change has been made. Change-Id: Iba0631c804bdab34d7c0232b49967130e3370488 Reviewed-on: https://code.wireshark.org/review/3199 Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-07-25 16:29:04 +00:00
value_string_ext *vs_avps_ext;
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
} diam_vnd_t;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
struct _diam_avp_t {
guint32 code;
Don't make promises of constness that we can't honor. svn path=/trunk/; revision=54757
2014-01-14 03:24:37 +00:00
diam_vnd_t *vendor;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
diam_avp_dissector_t dissector_rfc;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
gint ett;
int hf_value;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
void *type_data;
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
};
diameter: change GArray into wmem_array. This change fixes a leak in packet-diameter that loads a dictionary but doesn't free all the data. Found by valgrind. ==30481== 36,656 (960 direct, 35,696 indirect) bytes in 24 blocks are definitely lost in loss record 3,417 of 3,421 ==30481== at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==30481== by 0xA7FE610: g_malloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA81422D: g_slice_alloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA7CDC44: g_array_sized_new (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0x6863743: dictionary_load (packet-diameter.c:1980) ==30481== by 0x6863743: proto_register_diameter (packet-diameter.c:2344) ==30481== by 0x71C4BA4: register_all_protocols (register.c:323) ==30481== by 0x65EEFA7: proto_init (proto.c:521) ==30481== by 0x65CD621: epan_init (epan.c:126) ==30481== by 0x115330: main (tshark.c:1220) Change-Id: I3c0d19e1accab415355aa0f50c598f0c83356985 Reviewed-on: https://code.wireshark.org/review/13821 Petri-Dish: Dario Lombardo <lomato@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-02-18 10:30:41 +00:00
#define VND_AVP_VS(v) ((value_string *)(void *)(wmem_array_get_raw((v)->vs_avps)))
#define VND_AVP_VS_LEN(v) (wmem_array_get_count((v)->vs_avps))
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
typedef struct _diam_dictionary_t {
Use epan-scoped memory for building more of the diameter table stuff. svn path=/trunk/; revision=51754
2013-09-04 21:06:56 +00:00
wmem_tree_t *avps;
wmem_tree_t *vnds;
Add special handling of the (derived) AppId AVP type so that we can maintain one (and only one) Application ID list. This means we don't have to list all the Application IDs several times in the XML file. Collapse those lists into one (much more complete, now) list while also fixing URIs for several of the specifications (and making it clear for others that we don't have access to the specification--by declaring their URI to be "none"). Add a bunch more entries to the list of application IDs. Change-Id: Ia5c96b1f6f1fe3a9521b3d70142889e3881fae5e Reviewed-on: https://code.wireshark.org/review/8147 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-04-20 22:14:35 +00:00
value_string_ext *applications;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
value_string *commands;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
} diam_dictionary_t;
Don't make promises of constness that we can't honor. svn path=/trunk/; revision=54757
2014-01-14 03:24:37 +00:00
typedef diam_avp_t *(*avp_constructor_t)(const avp_type_t *, guint32, diam_vnd_t *, const char *, const value_string *, void *);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
struct _avp_type_t {
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
const char *name;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
diam_avp_dissector_t rfc;
enum ftenum ft;
Change enum type in struct to an int to allow different enum types to be used for the field display type in the struct. svn path=/trunk/; revision=36818
2011-04-22 17:25:25 +00:00
int base;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
avp_constructor_t build;
};
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
struct _build_dict {
Build diameter's hf array in a wmem array instead of a garray. Fixes ~270KB of leak when running epan_cleanup(). svn path=/trunk/; revision=50405
2013-07-06 12:45:33 +00:00
wmem_array_t *hf;
GPtrArray *ett;
GHashTable *types;
GHashTable *avps;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
};
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
typedef struct _address_avp_t {
gint ett;
int hf_address_type;
int hf_ipv4;
int hf_ipv6;
Diameter: Handle (IP)address type E164 as string The Diameter type Address hase a two byte address type family field previously only IPv4 and IPv6 was handled. Add handling of E.164 when encoded as a string.
2020-10-12 08:34:49 +00:00
int hf_e164_str;
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
int hf_other;
} address_avp_t;
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
typedef enum {
REASEMBLE_NEVER = 0,
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
REASEMBLE_AT_END,
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
REASEMBLE_BY_LENGTH
} avp_reassemble_mode_t;
typedef struct _proto_avp_t {
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
char *name;
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
dissector_handle_t handle;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
avp_reassemble_mode_t reassemble_mode;
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
} proto_avp_t;
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
static const char *simple_avp(diam_ctx_t *, diam_avp_t *, tvbuff_t *, diam_sub_dis_t *);
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
static diam_vnd_t unknown_vendor = { 0xffffffff, NULL, NULL };
static diam_vnd_t no_vnd = { 0, NULL, NULL };
static diam_avp_t unknown_avp = {0, &unknown_vendor, simple_avp, -1, -1, NULL };
Diameter: Allow diameter command code text to be added to custom column Create the command value string when loading the dictionary and assign it to the command code field when registering it. This makes it behave like any other value_string for adding to a custom column (allowing resolved and unresolved), etc. It could be changed to use a wmem_array instead of a GArray, as is currently done with the AVPs. Fix #16833. (Note that solving the analogous issue for the AVP code text is more complicated, since the interpretation depends on the vendor ID.)
2022-07-19 11:53:25 +00:00
static const value_string *cmd_vs;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
static diam_dictionary_t dictionary = { NULL, NULL, NULL, NULL };
static struct _build_dict build_dict;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const value_string *vnd_short_vs;
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
static dissector_handle_t data_handle;
From Sebastien Decugis: Add Missing Diameter EAP support. https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3600 svn path=/trunk/; revision=28849
2009-06-26 05:48:03 +00:00
static dissector_handle_t eap_handle;
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
static const value_string diameter_avp_data_addrfamily_vals[]= {
{1,"IPv4"},
{2,"IPv6"},
{3,"NSAP"},
{4,"HDLC"},
{5,"BBN"},
{6,"IEEE-802"},
{7,"E-163"},
{8,"E-164"},
{9,"F-69"},
{10,"X-121"},
{11,"IPX"},
{12,"Appletalk"},
{13,"Decnet4"},
{14,"Vines"},
{15,"E-164-NSAP"},
{16,"DNS"},
{17,"DistinguishedName"},
{18,"AS"},
{19,"XTPoIPv4"},
{20,"XTPoIPv6"},
{21,"XTPNative"},
{22,"FibrePortName"},
{23,"FibreNodeName"},
{24,"GWID"},
{0,NULL}
};
Make some symbols static. svn path=/trunk/; revision=35140
2010-12-06 23:20:21 +00:00
static value_string_ext diameter_avp_data_addrfamily_vals_ext = VALUE_STRING_EXT_INIT(diameter_avp_data_addrfamily_vals);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
static int proto_diameter = -1;
static int hf_diameter_length = -1;
static int hf_diameter_code = -1;
Updates from David Frascone. svn path=/trunk/; revision=3287
2001-04-10 21:49:23 +00:00
static int hf_diameter_hopbyhopid =-1;
static int hf_diameter_endtoendid =-1;
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
static int hf_diameter_version = -1;
static int hf_diameter_vendor_id = -1;
From: "Jacques, Olivier : - Added new RFC 3588 header format support (vendor-Id is replaced by application-Id) and command code has its own separated name space. The Draft v16 behaviour is maintained (through Preferences menu). - Added 3GPP IMS Cx/Dx interface decoding (new imscxdx.xml) - Fixed some issues in dictionary.xml - Refreshed Diameter applications in embedded AVPs (that do not rely on XML dictionary) svn path=/trunk/; revision=12235
2004-10-08 09:35:56 +00:00
static int hf_diameter_application_id = -1;
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
static int hf_diameter_flags = -1;
static int hf_diameter_flags_request = -1;
static int hf_diameter_flags_proxyable = -1;
static int hf_diameter_flags_error = -1;
From Anders Broman: fix the value strings for decoding of AVP values, add the T flag, make the result code and session ID named fields, and add some more field types. svn path=/trunk/; revision=10430
2004-03-21 23:10:31 +00:00
static int hf_diameter_flags_T = -1;
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
static int hf_diameter_flags_reserved4 = -1;
static int hf_diameter_flags_reserved5 = -1;
static int hf_diameter_flags_reserved6 = -1;
static int hf_diameter_flags_reserved7 = -1;
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
Allow filtering of whole AVP svn path=/trunk/; revision=22284
2007-07-10 17:45:11 +00:00
static int hf_diameter_avp = -1;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
static int hf_diameter_avp_len = -1;
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
static int hf_diameter_avp_code = -1;
static int hf_diameter_avp_flags = -1;
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
static int hf_diameter_avp_flags_vendor_specific = -1;
static int hf_diameter_avp_flags_mandatory = -1;
static int hf_diameter_avp_flags_protected = -1;
static int hf_diameter_avp_flags_reserved3 = -1;
static int hf_diameter_avp_flags_reserved4 = -1;
static int hf_diameter_avp_flags_reserved5 = -1;
static int hf_diameter_avp_flags_reserved6 = -1;
static int hf_diameter_avp_flags_reserved7 = -1;
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
static int hf_diameter_avp_vendor_id = -1;
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
static int hf_diameter_avp_data_wrong_length = -1;
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
static int hf_diameter_avp_pad = -1;
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
static int hf_diameter_answer_in = -1;
static int hf_diameter_answer_to = -1;
static int hf_diameter_answer_time = -1;
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
/* AVPs with special/extra decoding */
static int hf_framed_ipv6_prefix_reserved = -1;
static int hf_framed_ipv6_prefix_length = -1;
static int hf_framed_ipv6_prefix_bytes = -1;
static int hf_framed_ipv6_prefix_ipv6 = -1;
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
static int hf_diameter_3gpp2_exp_res = -1;
Show the Experimental-Result-Code if we don't know have a subdissector for that vendor ID. Otherwise the value is simply not shown to the user. Adding support for a vendor ID's Experimental Result Codes isn't as easy as modifying the XML so don't add an expert info about it. Change-Id: I65f2cb13853cc7141fb242fa03c6e474a6c02cb9 Reviewed-on: https://code.wireshark.org/review/11294 Reviewed-by: Michael Mann <mmann78@netscape.net> Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-10-26 20:44:18 +00:00
static int hf_diameter_other_vendor_exp_res = -1;
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
static int hf_diameter_mip6_feature_vector = -1;
static int hf_diameter_mip6_feature_vector_mip6_integrated = -1;
static int hf_diameter_mip6_feature_vector_local_home_agent_assignment = -1;
static int hf_diameter_mip6_feature_vector_pmip6_supported = -1;
static int hf_diameter_mip6_feature_vector_ip4_hoa_supported = -1;
static int hf_diameter_mip6_feature_vector_local_mag_routing_supported = -1;
static int hf_diameter_3gpp_mip6_feature_vector = -1;
static int hf_diameter_3gpp_mip6_feature_vector_assign_local_ip = -1;
static int hf_diameter_3gpp_mip6_feature_vector_mip4_supported = -1;
static int hf_diameter_3gpp_mip6_feature_vector_optimized_idle_mode_mobility = -1;
static int hf_diameter_3gpp_mip6_feature_vector_gtpv2_supported = -1;
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
static int hf_diameter_user_equipment_info_imeisv = -1;
static int hf_diameter_user_equipment_info_mac = -1;
static int hf_diameter_user_equipment_info_eui64 = -1;
static int hf_diameter_user_equipment_info_modified_eui64 = -1;
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
static gint ett_diameter = -1;
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
static gint ett_diameter_flags = -1;
static gint ett_diameter_avp_flags = -1;
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
static gint ett_diameter_avpinfo = -1;
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
static gint ett_unknown = -1;
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
static gint ett_diameter_mip6_feature_vector = -1;
static gint ett_diameter_3gpp_mip6_feature_vector = -1;
Convert remaining proto_item_set_expert_flags in dissector directory. svn path=/trunk/; revision=51142
2013-08-05 01:54:15 +00:00
static expert_field ei_diameter_reserved_bit_set = EI_INIT;
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
static expert_field ei_diameter_avp_len = EI_INIT;
static expert_field ei_diameter_avp_no_data = EI_INIT;
static expert_field ei_diameter_application_id = EI_INIT;
static expert_field ei_diameter_version = EI_INIT;
static expert_field ei_diameter_avp_pad = EI_INIT;
Diameter: Add MS-CHAP/MS-CHAPv2 AVP support for EAP-TLS Created the Microsoft Diameter file based on MS-CHAP-* AVPs listed at https://www.iana.org/assignments/eap-numbers/eap-numbers.xhtml#eap-numbers-10 Many values are displayed as bytes for simplicit. The MS-CHAP2-Success attribute could for example be dissected further as 1 byte followed by a string, but that requires more effort. Allow padding to be missing since the eap-ttls-mschapv2.pcapng capture would throw a Malformed Packet exception otherwise. Bug: 15603 Change-Id: I9efc322a86802e78bb6cd4bc3df1c1282a45fe9e Reviewed-on: https://code.wireshark.org/review/34291 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 23:19:33 +00:00
static expert_field ei_diameter_avp_pad_missing = EI_INIT;
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
static expert_field ei_diameter_code = EI_INIT;
static expert_field ei_diameter_avp_code = EI_INIT;
static expert_field ei_diameter_avp_vendor_id = EI_INIT;
Diameter: check IPv6 prefix length before copying it in e_in6_addr structure Bug: 11792 Change-Id: I37a07044d40f10e9a1a90025d90753fdb3db2278 Reviewed-on: https://code.wireshark.org/review/12248 Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
2015-11-28 10:45:24 +00:00
static expert_field ei_diameter_invalid_ipv6_prefix_len = EI_INIT;
[Diameter] Improve dissection of malformed packets by continnuing dissection and display the problem more prominetly. Change-Id: Ia1a32667a18e1e5b60b5c167da9b6dd945ba3dfc Reviewed-on: https://code.wireshark.org/review/17385 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-08-30 12:55:54 +00:00
static expert_field ei_diameter_invalid_avp_len = EI_INIT;
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
static expert_field ei_diameter_invalid_user_equipment_info_value_len = EI_INIT;
Diameter: improve IMEI/IMEISV dissection in User-Equipment-Info AVP 3GPP.User-Equipment-Info AVP should contain the IMEISV (if type is 0). However some implementations put the IMEI there: decode it but display a protocol error indication too.
2022-03-19 12:44:56 +00:00
static expert_field ei_diameter_unexpected_imei_as_user_equipment_info = EI_INIT;
Convert remaining proto_item_set_expert_flags in dissector directory. svn path=/trunk/; revision=51142
2013-08-05 01:54:15 +00:00
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
/* Tap for Diameter */
static int diameter_tap = -1;
/* For conversations */
On request from Tommy Wu: Enable Diameter over UDP dissection as there is an application using it. svn path=/trunk/; revision=51473
2013-08-22 17:21:20 +00:00
static dissector_handle_t diameter_udp_handle;
Add port ranges. svn path=/trunk/; revision=23265
2007-10-24 15:04:01 +00:00
static dissector_handle_t diameter_tcp_handle;
Use port range for SCTP too. svn path=/trunk/; revision=34557
2010-10-18 15:56:06 +00:00
static dissector_handle_t diameter_sctp_handle;
diameter: Register the [D]TLS port Port 5868 is IANA registered for Diameter over TLS/TCP and DTLS/SCTP. Register the diameter TCP handle by name so that it can be registered to tls.
2022-07-02 01:03:05 +00:00
/* This is IANA registered for TCP and SCTP (and reserved for UDP) */
Minor cleanup mostly related to proto_reg_handoff svn path=/trunk/; revision=26236
2008-09-19 17:39:44 +00:00
#define DEFAULT_DIAMETER_PORT_RANGE "3868"
diameter: Register the [D]TLS port Port 5868 is IANA registered for Diameter over TLS/TCP and DTLS/SCTP. Register the diameter TCP handle by name so that it can be registered to tls.
2022-07-02 01:03:05 +00:00
/* This is IANA registered for TLS/TCP and DTLS/SCTP (and reserved for UDP) */
#define DEFAULT_DIAMETER_TLS_PORT 5868
Add port ranges. svn path=/trunk/; revision=23265
2007-10-24 15:04:01 +00:00
Support for desegmentation of Diameter-over-TCP, from David Frascone. svn path=/trunk/; revision=4141
2001-11-03 12:19:56 +00:00
/* desegmentation of Diameter over TCP */
static gboolean gbl_diameter_desegment = TRUE;
Make it possible to do subdissection of Diameter AVP:s and add dissection of some 3GPP AVP:s. svn path=/trunk/; revision=24211
2008-01-27 21:06:33 +00:00
/* Dissector tables */
Dissect Vendor id and Auth-application Id. svn path=/trunk/; revision=24216
2008-01-28 19:55:20 +00:00
static dissector_table_t diameter_dissector_table;
Make it possible to do subdissection of Diameter AVP:s and add dissection of some 3GPP AVP:s. svn path=/trunk/; revision=24211
2008-01-27 21:06:33 +00:00
static dissector_table_t diameter_3gpp_avp_dissector_table;
Register a dissector table for dissecting Ericsson AVP:s. svn path=/trunk/; revision=32712
2010-05-07 11:08:56 +00:00
static dissector_table_t diameter_ericsson_avp_dissector_table;
[Diameter] Add a table for subdissectiong AVPs for VENDOR_VERIZON Change-Id: I37397e590f294d209b89fe868b365cafb5343464 Reviewed-on: https://code.wireshark.org/review/21363 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2017-04-27 11:03:42 +00:00
static dissector_table_t diameter_verizon_avp_dissector_table;
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
static dissector_table_t diameter_expr_result_vnd_table;
Make it possible to do subdissection of Diameter AVP:s and add dissection of some 3GPP AVP:s. svn path=/trunk/; revision=24211
2008-01-27 21:06:33 +00:00
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *avpflags_str[] = {
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
"---",
"--P",
"-M-",
"-MP",
"V--",
"V-P",
"VM-",
"VMP",
};
Removed trailing whitespaces from .h and .c files using the winapi_cleanup tool written by Patrik Stridvall for the wine project. svn path=/trunk/; revision=6117
2002-08-28 21:04:11 +00:00
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
#define SUBSCRIPTION_ID_TYPE_E164 0
#define SUBSCRIPTION_ID_TYPE_IMSI 1
#define SUBSCRIPTION_ID_TYPE_SIP_URI 2
#define SUBSCRIPTION_ID_TYPE_NAI 3
#define SUBSCRIPTION_ID_TYPE_PRIVATE 4
Decode User-Name as an IMSI in S13/S13' and S7a/S7d. Also move the application ID macros into the diameter header file. Change-Id: Iaca5707c8476d81f50ecdb3aab76be293b5ccfe7 Reviewed-on: https://code.wireshark.org/review/3786 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-21 21:40:04 +00:00
#define SUBSCRIPTION_ID_TYPE_UNKNOWN (guint32)-1
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
#define USER_EQUIPMENT_INFO_TYPE_IMEISV 0
#define USER_EQUIPMENT_INFO_TYPE_MAC 1
#define USER_EQUIPMENT_INFO_TYPE_EUI64 2
#define USER_EQUIPMENT_INFO_TYPE_MODIFIED_EUI64 3
#define USER_EQUIPMENT_INFO_TYPE_UNKNOWN (guint32)-1
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
Add export_pdu_tap to packet-diameter.c svn path=/trunk/; revision=49448
2013-05-20 18:30:32 +00:00
static void
export_diameter_pdu(packet_info *pinfo, tvbuff_t *tvb)
{
Dissector names are not protocol names. A given protocol's packet format may depend, for example, on which lower-level protocol is transporting the protocol in question. For example, protocols that run atop both byte-stream protocols such as TCP and TLS, and packet-oriented protocols such as UDP or DTLS, might begin the packet with a length when running atop a byte-stream protocol, to indicate where this packet ends and the next packet begins in the byte stream, but not do so when running atop a packet-oriented protocol. Dissectors can handle this in various ways: For example, the dissector could attempt to determine the protocol over which the packet was transported. Unfortunately, many of those mechanisms do so by fetching data from the packet_info structure, and many items in that structure act as global variables, so that, for example, if there are two two PDUs for protocol A inside a TCP segment, and the first protocol for PDU A contains a PDU for protocol B, and protocol B's dissector, or a dissector it calls, modifies the information in the packet_info structure so that it no longer indicates that the parent protocol is TCP, the second PDU for protocol A might not be correctly dissected. Another such mechanism is to query the previous element in the layers structure of the packet_info structure, which is a list of protocol IDs. Unfortunately, that is not a list of earlier protocols in the protocol stack, it's a list of earlier protocols in the dissection, which means that, in the above example, when the second PDU for protocol A is dissected, the list is {...,TCP,A,B,...,A}, which means that the previous element in the list is not TCP, so, again, the second PDU for protocol A will not be correctly dissected. An alternative is to have multiple dissectors for the same protocol, with the part of the protocol that's independent of the protocol transporting the PDU being dissected by common code. Protocol B might have an "over a byte-stream transport" dissector and an "over a packet transport" dissector, with the first dissector being registered for use over TCP and TLS and the other dissector being registered for use over packet protocols. This mechanism, unlike the other mechanisms, is not dependent on information in the packet_info structure that might be affected by dissectors other than the one for the protocol that transports protocol B. Furthermore, in a LINKTYPE_WIRESHARK_UPPER_PDU pcap or pcapng packet for protocol B, there might not be any information to indicate the protocol that transports protocol B, so there would have to be separate dissectors for protocol B, with separate names, so that a tag giving the protocol name would differ for B-over-byte-stream and B-over-packets. So: We rename EXP_PDU_TAG_PROTO_NAME and EXP_PDU_TAG_HEUR_PROTO_NAME to EXP_PDU_TAG_DISSECTOR_NAME and EXP_PDU_TAG_HEUR_DISSECTOR_NAME, to emphasize that they are *not* protocol names, they are dissector names (which has always been the case - if there's a protocol with that name, but no dissector with that name, Wireshark will not be able to handle the packet, as it will try to look up a dissector given that name and fail). We fix that exported PDU dissector to refer to those tags as dissector names, not protocol names. We update documentation to refer to them as DISSECTOR_NAME tags, not PROTO_NAME tags. (If there is any documentation for this outside the Wireshark source, it should be updated as well.) We add comments for calls to dissector_handle_get_dissector_name() where the dissector name is shown to the user, to indicate that it might be that the protocol name should be used. We update the TLS and DTLS dissectors to show the encapsulated protocol as the string returned by dissector_handle_get_long_name(); as the default is "Application Data", it appeaers that a descriptive name, rather than a short API name, should be used. (We continue to use the dissector name in debugging messages, to indicate which dissector was called.)
2022-09-11 05:37:11 +00:00
exp_pdu_data_t *exp_pdu_data = export_pdu_create_common_tags(pinfo, "diameter", EXP_PDU_TAG_DISSECTOR_NAME);
Add export_pdu_tap to packet-diameter.c svn path=/trunk/; revision=49448
2013-05-20 18:30:32 +00:00
Exported PDU: add support for data length on the wire Change-Id: Iff14ec4d0297ec85f3597b33871fb0ab5256a597 Reviewed-on: https://code.wireshark.org/review/728 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-18 22:01:22 +00:00
exp_pdu_data->tvb_captured_length = tvb_captured_length(tvb);
exp_pdu_data->tvb_reported_length = tvb_reported_length(tvb);
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
exp_pdu_data->pdu_tvb = tvb;
Add export_pdu_tap to packet-diameter.c svn path=/trunk/; revision=49448
2013-05-20 18:30:32 +00:00
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
tap_queue_packet(exported_pdu_tap, pinfo, exp_pdu_data);
Add export_pdu_tap to packet-diameter.c svn path=/trunk/; revision=49448
2013-05-20 18:30:32 +00:00
}
Build dynamic diameter value_string arrays with wmem arrays not garrays. Fixes another ~60KB of leaks. svn path=/trunk/; revision=50412
2013-07-06 18:01:57 +00:00
static int
compare_avps(const void *a, const void *b)
Use value_string_ext for AVP:s svn path=/trunk/; revision=32932
2010-05-24 14:20:53 +00:00
{
Build dynamic diameter value_string arrays with wmem arrays not garrays. Fixes another ~60KB of leaks. svn path=/trunk/; revision=50412
2013-07-06 18:01:57 +00:00
const value_string *vsa = (const value_string *)a;
const value_string *vsb = (const value_string *)b;
Use value_string_ext for AVP:s svn path=/trunk/; revision=32932
2010-05-24 14:20:53 +00:00
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
if (vsa->value > vsb->value)
Use value_string_ext for AVP:s svn path=/trunk/; revision=32932
2010-05-24 14:20:53 +00:00
return 1;
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
if (vsa->value < vsb->value)
Use value_string_ext for AVP:s svn path=/trunk/; revision=32932
2010-05-24 14:20:53 +00:00
return -1;
return 0;
}
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
static GHashTable* diameterstat_cmd_str_hash = NULL;
#define DIAMETER_NUM_PROCEDURES 1
static void
Get rid of some GTK+-only stuff. Change-Id: I841a1cdb637f37673466e4d065474c0f7e2dfc56 Reviewed-on: https://code.wireshark.org/review/27512 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-05-13 19:45:16 +00:00
diameterstat_init(struct register_srt* srt _U_, GArray* srt_array)
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
{
srt_stat_table *diameter_srt_table;
int* idx;
Fix a lot of typos and misspellings Change-Id: I8512cfa1d424f82a873a0e0e1d22c7b075fdd7f3 Reviewed-on: https://code.wireshark.org/review/13069 Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-01-06 00:58:42 +00:00
/* XXX - This is a hack/workaround support so resetting/freeing parameters at the dissector
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
level doesn't need to be supported. */
if (diameterstat_cmd_str_hash != NULL)
{
g_hash_table_destroy(diameterstat_cmd_str_hash);
}
Replace instances of wmem_alloc with wmem_new This commit replaces instances of (myobj *)wmem_alloc(wmem_file_scope(), sizeof(myobj)) and replaces them with: wmem_new(wmem_file_scope(), myobj) to improve the readability of Wireshark's code. The replacement locations were identified with grep and replaced with the Python script below. grep command: egrep "wmem_alloc0?\(wmem_file_scope\(\), sizeof\([a-z_]+\)\)" . -R -l python script: import re import sys import fileinput pattern = r'\(([^\s]+) ?\*\) ?wmem_alloc(0?)\((wmem_[a-z]+_scope\(\)), sizeof\(\1\)\)' replacewith = r'wmem_new\2(\3, \1)' fname = sys.argv[1] for line in fileinput.input(fname, inplace=1, mode='rb'): output = re.sub(pattern, replacewith, line) sys.stdout.write(output) Change-Id: Ieac246c104bf01e32cbc6e11e53e81c7f639d870 Reviewed-on: https://code.wireshark.org/review/37158 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2020-05-07 12:35:45 +00:00
idx = wmem_new0(wmem_epan_scope(), int);
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
diameterstat_cmd_str_hash = g_hash_table_new(g_str_hash,g_str_equal);
Remove -Wwrite-strings compiler flag The "-Wwrite-strings" flag produces nuisance warnings. These warnings are not useful, they're impossible to fix in a sane way and therefore are being handled with casts of static strings to (char *). This just moves the warning to [-Wcast-qual] and a compiler pragma is in turn required (and used) to squelch that warning. Remove the Wwrite-strings warning. Let that responsibility fall on the programmer (as is done by casting). Change-Id: I5a44dfd9decd6d80797a521a3373593074962fb5 Reviewed-on: https://code.wireshark.org/review/12162 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-11-21 23:48:27 +00:00
g_hash_table_insert(diameterstat_cmd_str_hash, "Unknown", idx);
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
/** @todo the filter to use in stead of NULL is "diameter.cmd.code"
Fix a lot of typos and misspellings Change-Id: I8512cfa1d424f82a873a0e0e1d22c7b075fdd7f3 Reviewed-on: https://code.wireshark.org/review/13069 Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-01-06 00:58:42 +00:00
* to enable the filter popup in the service response time dialogue
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
* Note to make it work the command code must be stored rather than the
* index.
*/
Get rid of some GTK+-only stuff. Change-Id: I841a1cdb637f37673466e4d065474c0f7e2dfc56 Reviewed-on: https://code.wireshark.org/review/27512 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-05-13 19:45:16 +00:00
diameter_srt_table = init_srt_table("Diameter Requests", NULL, srt_array, DIAMETER_NUM_PROCEDURES, NULL, NULL, NULL);
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
init_srt_table_row(diameter_srt_table, 0, "Unknown");
}
Add a "failed" return for tap packet routines. This allows taps that can fail to report an error and fail; a failed tap's packet routine won't be called again, so they don't have to keep track of whether they've failed themselves. We make the return value from the packet routine an enum. Don't have a separate type for the per-packet routine for "follow" taps; they're expected to act like tap packet routines, so just use the type for tap packet routines. One tap packet routine returned -1; that's not a valid return value, and wasn't one before this change (the return value was a boolean), so presume the intent was "don't redraw". Another tap routine's early return, without doing any work, returned TRUE; this is presumably an error (no work done, no need to redraw), so presumably it should be "don't redraw". Clean up some white space while we're at it. Change-Id: Ia7d2b717b2cace4b13c2b886e699aa4d79cc82c8 Reviewed-on: https://code.wireshark.org/review/31283 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-01-01 03:36:12 +00:00
static tap_packet_status
tap: Adding flags for tap_packet This allows flags to be passed by the registering listener to the collection of information
2022-06-09 15:47:35 +00:00
diameterstat_packet(void *pss, packet_info *pinfo, epan_dissect_t *edt _U_, const void *prv, tap_flags_t flags _U_)
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
{
guint i = 0;
srt_stat_table *diameter_srt_table;
srt_data_t *data = (srt_data_t *)pss;
const diameter_req_ans_pair_t *diameter=(const diameter_req_ans_pair_t *)prv;
int* idx = NULL;
/* Process only answers where corresponding request is found.
Fix a lot of typos and misspellings Change-Id: I8512cfa1d424f82a873a0e0e1d22c7b075fdd7f3 Reviewed-on: https://code.wireshark.org/review/13069 Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-01-06 00:58:42 +00:00
* Unpaired diameter messages are currently not supported by statistics.
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
* Return 0, since redraw is not needed. */
if(!diameter || diameter->processing_request || !diameter->req_frame)
Add a "failed" return for tap packet routines. This allows taps that can fail to report an error and fail; a failed tap's packet routine won't be called again, so they don't have to keep track of whether they've failed themselves. We make the return value from the packet routine an enum. Don't have a separate type for the per-packet routine for "follow" taps; they're expected to act like tap packet routines, so just use the type for tap packet routines. One tap packet routine returned -1; that's not a valid return value, and wasn't one before this change (the return value was a boolean), so presume the intent was "don't redraw". Another tap routine's early return, without doing any work, returned TRUE; this is presumably an error (no work done, no need to redraw), so presumably it should be "don't redraw". Clean up some white space while we're at it. Change-Id: Ia7d2b717b2cace4b13c2b886e699aa4d79cc82c8 Reviewed-on: https://code.wireshark.org/review/31283 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-01-01 03:36:12 +00:00
return TAP_PACKET_DONT_REDRAW;
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
diameter_srt_table = g_array_index(data->srt_array, srt_stat_table*, i);
idx = (int*) g_hash_table_lookup(diameterstat_cmd_str_hash, diameter->cmd_str);
if (idx == NULL) {
Replace instances of wmem_alloc with wmem_new This commit replaces instances of (myobj *)wmem_alloc(wmem_file_scope(), sizeof(myobj)) and replaces them with: wmem_new(wmem_file_scope(), myobj) to improve the readability of Wireshark's code. The replacement locations were identified with grep and replaced with the Python script below. grep command: egrep "wmem_alloc0?\(wmem_file_scope\(\), sizeof\([a-z_]+\)\)" . -R -l python script: import re import sys import fileinput pattern = r'\(([^\s]+) ?\*\) ?wmem_alloc(0?)\((wmem_[a-z]+_scope\(\)), sizeof\(\1\)\)' replacewith = r'wmem_new\2(\3, \1)' fname = sys.argv[1] for line in fileinput.input(fname, inplace=1, mode='rb'): output = re.sub(pattern, replacewith, line) sys.stdout.write(output) Change-Id: Ieac246c104bf01e32cbc6e11e53e81c7f639d870 Reviewed-on: https://code.wireshark.org/review/37158 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2020-05-07 12:35:45 +00:00
idx = wmem_new(wmem_epan_scope(), int);
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
*idx = (int) g_hash_table_size(diameterstat_cmd_str_hash);
g_hash_table_insert(diameterstat_cmd_str_hash, (gchar*) diameter->cmd_str, idx);
init_srt_table_row(diameter_srt_table, *idx, (const char*) diameter->cmd_str);
}
add_srt_table_data(diameter_srt_table, *idx, &diameter->req_time, pinfo);
Add a "failed" return for tap packet routines. This allows taps that can fail to report an error and fail; a failed tap's packet routine won't be called again, so they don't have to keep track of whether they've failed themselves. We make the return value from the packet routine an enum. Don't have a separate type for the per-packet routine for "follow" taps; they're expected to act like tap packet routines, so just use the type for tap packet routines. One tap packet routine returned -1; that's not a valid return value, and wasn't one before this change (the return value was a boolean), so presume the intent was "don't redraw". Another tap routine's early return, without doing any work, returned TRUE; this is presumably an error (no work done, no need to redraw), so presumably it should be "don't redraw". Clean up some white space while we're at it. Change-Id: Ia7d2b717b2cace4b13c2b886e699aa4d79cc82c8 Reviewed-on: https://code.wireshark.org/review/31283 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-01-01 03:36:12 +00:00
return TAP_PACKET_REDRAW;
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
}
Include padding length in the AVP's length. svn path=/trunk/; revision=48702
2013-04-02 18:15:42 +00:00
/* Special decoding of some AVPs */
Dissect Vendor id and Auth-application Id. svn path=/trunk/; revision=24216
2008-01-28 19:55:20 +00:00
static int
Initial commit to support yet another method of passing data between dissectors. Add new parameter 'data' to heur_dissector_t and new_dissector_t, for now it's always NULL svn path=/trunk/; revision=44860
2012-09-10 21:40:21 +00:00
dissect_diameter_vendor_id(tvbuff_t *tvb, packet_info *pinfo _U_, proto_tree *tree, void *data _U_)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Dissect Vendor id and Auth-application Id. svn path=/trunk/; revision=24216
2008-01-28 19:55:20 +00:00
int offset = 0;
Convert 'encoding' parameter of certain proto_tree_add_item() calls in non-autogenerated epan/dissectors: Specifically: Replace FALSE|0 and TRUE|1 by ENC_BIG_ENDIAN|ENC_LITTLE_ENDIAN as the encoding parameter for proto_tree_add_item() calls which directly reference an item in hf[] which has a type of: FT_UINT8 FT_UINT16 FT_UINT24 FT_UINT32 FT_UINT64 FT_INT8 FT_INT16 FT_INT24 FT_INT32 FT_INT64 FT_FLOAT FT_DOUBLE svn path=/trunk/; revision=39288
2011-10-06 03:35:44 +00:00
proto_tree_add_item(tree, hf_diameter_vendor_id, tvb, 0, 4, ENC_BIG_ENDIAN);
Dissect Vendor id and Auth-application Id. svn path=/trunk/; revision=24216
2008-01-28 19:55:20 +00:00
offset++;
return offset;
}
From Sebastien Decugis: Add Missing Diameter EAP support. https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3600 svn path=/trunk/; revision=28849
2009-06-26 05:48:03 +00:00
static int
Initial commit to support yet another method of passing data between dissectors. Add new parameter 'data' to heur_dissector_t and new_dissector_t, for now it's always NULL svn path=/trunk/; revision=44860
2012-09-10 21:40:21 +00:00
dissect_diameter_eap_payload(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data _U_)
Use port range for SCTP too. svn path=/trunk/; revision=34557
2010-10-18 15:56:06 +00:00
{
From Sebastien Decugis: Add Missing Diameter EAP support. https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3600 svn path=/trunk/; revision=28849
2009-06-26 05:48:03 +00:00
gboolean save_writable;
/* Ensure the packet is displayed as Diameter, not EAP */
Allow control of individual columns to be (un)writable. Most protocols just want to limit COL_INFO or COL_PROTOCOL so give that level of granularity. Bug: 12144 Bug: 5117 Bug: 11144 Change-Id: I8de9b7d2c69e90d3fbfc0a52c2bd78c3de58e2f8 Reviewed-on: https://code.wireshark.org/review/15894 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com> Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-06-14 15:55:08 +00:00
save_writable = col_get_writable(pinfo->cinfo, COL_PROTOCOL);
col_set_writable(pinfo->cinfo, COL_PROTOCOL, FALSE);
From Sebastien Decugis: Add Missing Diameter EAP support. https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3600 svn path=/trunk/; revision=28849
2009-06-26 05:48:03 +00:00
call_dissector(eap_handle, tvb, pinfo, tree);
Allow control of individual columns to be (un)writable. Most protocols just want to limit COL_INFO or COL_PROTOCOL so give that level of granularity. Bug: 12144 Bug: 5117 Bug: 11144 Change-Id: I8de9b7d2c69e90d3fbfc0a52c2bd78c3de58e2f8 Reviewed-on: https://code.wireshark.org/review/15894 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com> Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-06-14 15:55:08 +00:00
col_set_writable(pinfo->cinfo, COL_PROTOCOL, save_writable);
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
return tvb_reported_length(tvb);
From Sebastien Decugis: Add Missing Diameter EAP support. https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3600 svn path=/trunk/; revision=28849
2009-06-26 05:48:03 +00:00
}
Dissect Vendor id and Auth-application Id. svn path=/trunk/; revision=24216
2008-01-28 19:55:20 +00:00
Fix up some broken links to specs from dissectors. check_dissector_urls.py was written and used to find URLs within epan/dissectors/*.c and try to fetch them using 'requests'. Will be commmitted separately. Most of the changes were to adapt to reorganisation of IETF or 3gpp2 links, but many of the broken links are for websites or companies that no longer exist. Change-Id: Ie9afdb95099218402a61626a0cd5193c6f781b96 Reviewed-on: https://code.wireshark.org/review/36769 Reviewed-by: Martin Mathieson <martin.r.mathieson@googlemail.com>
2020-04-10 11:04:46 +00:00
/* https://www.3gpp2.org/Public_html/X/VSA-VSE.cfm */
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
static const value_string diameter_3gpp2_exp_res_vals[]= {
{ 5001, "Diameter_Error_User_No_WLAN_Subscription"},
{ 5002, "Diameter_Error_Roaming_Not_Allowed(Obsoleted)"},
{ 5003, "Diameter_Error_User_No_FAP_Subscription"},
{0,NULL}
};
static int
Show the Experimental-Result-Code if we don't know have a subdissector for that vendor ID. Otherwise the value is simply not shown to the user. Adding support for a vendor ID's Experimental Result Codes isn't as easy as modifying the XML so don't add an expert info about it. Change-Id: I65f2cb13853cc7141fb242fa03c6e474a6c02cb9 Reviewed-on: https://code.wireshark.org/review/11294 Reviewed-by: Michael Mann <mmann78@netscape.net> Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-10-26 20:44:18 +00:00
dissect_diameter_3gpp2_exp_res(tvbuff_t *tvb, packet_info *pinfo _U_, proto_tree *tree, void *data)
{
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
proto_item *pi;
Reject the packet if data is NULL. svn path=/trunk/; revision=53898
2013-12-10 03:25:04 +00:00
diam_sub_dis_t *diam_sub_dis;
/* Reject the packet if data is NULL */
if (data == NULL)
return 0;
diam_sub_dis = (diam_sub_dis_t*)data;
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (tree) {
pi = proto_tree_add_item(tree, hf_diameter_3gpp2_exp_res, tvb, 0, 4, ENC_BIG_ENDIAN);
diam_sub_dis->avp_str = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), diam_sub_dis->avp_str);
diam_sub_dis->avp_str = strstr(diam_sub_dis->avp_str,": ")+2;
}
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
return 4;
}
Show the Experimental-Result-Code if we don't know have a subdissector for that vendor ID. Otherwise the value is simply not shown to the user. Adding support for a vendor ID's Experimental Result Codes isn't as easy as modifying the XML so don't add an expert info about it. Change-Id: I65f2cb13853cc7141fb242fa03c6e474a6c02cb9 Reviewed-on: https://code.wireshark.org/review/11294 Reviewed-by: Michael Mann <mmann78@netscape.net> Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-10-26 20:44:18 +00:00
static void
dissect_diameter_other_vendor_exp_res(tvbuff_t *tvb, proto_tree *tree, diam_sub_dis_t *diam_sub_dis)
{
proto_item *pi;
if (tree) {
pi = proto_tree_add_item(tree, hf_diameter_other_vendor_exp_res, tvb, 0, 4, ENC_BIG_ENDIAN);
diam_sub_dis->avp_str = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), diam_sub_dis->avp_str);
diam_sub_dis->avp_str = strstr(diam_sub_dis->avp_str,": ")+2;
}
}
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
/* From RFC 3162 section 2.3 */
static int
diameter: fix 'pinfo/vs' was marked unused but was used [-Werror,-Wused-but-marked-unused] Change-Id: Ib7f1682a3cdeb8571a410db07261ff68bac03432 Reviewed-on: https://code.wireshark.org/review/14977 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-04-15 17:39:06 +00:00
dissect_diameter_base_framed_ipv6_prefix(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
{
Make the avp_str available to sub dissectors and Let the subdissector have precedence filling in the avp_item string. Use that in a couple of places. Change-Id: I1af7a1ca4c14fb56ddeaab336202e6c2a18e556b Reviewed-on: https://code.wireshark.org/review/699 Reviewed-by: Michael Mann <mmann78@netscape.net> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-16 17:57:18 +00:00
diam_sub_dis_t *diam_sub_dis = (diam_sub_dis_t*)data;
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
guint32 prefix_len, prefix_len_bytes;
Diameter: check IPv6 prefix length before copying it in e_in6_addr structure Bug: 11792 Change-Id: I37a07044d40f10e9a1a90025d90753fdb3db2278 Reviewed-on: https://code.wireshark.org/review/12248 Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
2015-11-28 10:45:24 +00:00
proto_item *pi;
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
Whitespace cleanup: Indentation, trailing, ... svn path=/trunk/; revision=42266
2012-04-26 16:05:05 +00:00
proto_tree_add_item(tree, hf_framed_ipv6_prefix_reserved, tvb, 0, 1, ENC_BIG_ENDIAN);
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
pi = proto_tree_add_item_ret_uint(tree, hf_framed_ipv6_prefix_length, tvb, 1, 1, ENC_BIG_ENDIAN, &prefix_len);
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
Diameter: check IPv6 prefix length before copying it in e_in6_addr structure Bug: 11792 Change-Id: I37a07044d40f10e9a1a90025d90753fdb3db2278 Reviewed-on: https://code.wireshark.org/review/12248 Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
2015-11-28 10:45:24 +00:00
if (prefix_len > 128) {
expert_add_info(pinfo, pi, &ei_diameter_invalid_ipv6_prefix_len);
}
Whitespace cleanup: Indentation, trailing, ... svn path=/trunk/; revision=42266
2012-04-26 16:05:05 +00:00
prefix_len_bytes = prefix_len / 8;
if (prefix_len % 8)
prefix_len_bytes++;
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
Whitespace cleanup: Indentation, trailing, ... svn path=/trunk/; revision=42266
2012-04-26 16:05:05 +00:00
proto_tree_add_item(tree, hf_framed_ipv6_prefix_bytes, tvb, 2, prefix_len_bytes, ENC_NA);
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
Whitespace cleanup: Indentation, trailing, ... svn path=/trunk/; revision=42266
2012-04-26 16:05:05 +00:00
/* If we have a fully IPv6 address, display it as such */
Don't mark a used parameter as unused. Also fix up some whitespace. Change-Id: I7cd78740199ce7b2682902a5687c4f05c2c963b2 Reviewed-on: https://code.wireshark.org/review/3716 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-19 13:01:43 +00:00
if (prefix_len_bytes == 16) {
Whitespace cleanup: Indentation, trailing, ... svn path=/trunk/; revision=42266
2012-04-26 16:05:05 +00:00
proto_tree_add_item(tree, hf_framed_ipv6_prefix_ipv6, tvb, 2, prefix_len_bytes, ENC_NA);
Diameter: check IPv6 prefix length before copying it in e_in6_addr structure Bug: 11792 Change-Id: I37a07044d40f10e9a1a90025d90753fdb3db2278 Reviewed-on: https://code.wireshark.org/review/12248 Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
2015-11-28 10:45:24 +00:00
} else if (prefix_len_bytes < 16) {
Add a ws_in6_addr typedef for struct e_in6_addr. That allows a parallel typedef of ws_in4_addr for guint32. Change-Id: I03b230247065e0e3840eb87635315a8e523ef562 Reviewed-on: https://code.wireshark.org/review/24073 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2017-10-26 08:50:00 +00:00
ws_in6_addr value;
Replace ip6_to_str and ip6_guint8_to_str with address_to_str and tvb_ip_to_str. Change-Id: I1d258923a7a63539ec8456d3e306bca5016a1e4b Reviewed-on: https://code.wireshark.org/review/6060 Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-12-21 17:38:13 +00:00
address addr;
Display framed_ipv6_prefix as an IPv6 address even when it's not the full 16 bits. Change-Id: I6cef08e2da9467cb54b5d08526c77c87e7afe1fa Reviewed-on: https://code.wireshark.org/review/645 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-14 12:55:34 +00:00
memset(&value.bytes, 0, sizeof(value));
tvb_memcpy(tvb, (guint8 *)&value.bytes, 2, prefix_len_bytes);
value.bytes[prefix_len_bytes] = value.bytes[prefix_len_bytes] & (0xff<<(prefix_len % 8));
Change proto_tree_add_ipv6() to take a struct e_in6_addr pointer tvb_get_ipv6() takes a struct e_in6_addr *, use that here too. Change-Id: Id8b368daa05c151a61d4bc01dc88c00da13e9c88 Reviewed-on: https://code.wireshark.org/review/10953 Reviewed-by: Michael Mann <mmann78@netscape.net> Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Balint Reczey <balint@balintreczey.hu> Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
2015-10-05 21:43:46 +00:00
proto_tree_add_ipv6(tree, hf_framed_ipv6_prefix_ipv6, tvb, 2, prefix_len_bytes, &value);
Use address functions instead of ADDRESS macros in asn1 and epan Replace CMP_ADDRESS, COPY_ADDRESS, et al with their lower-case equivalents in the asn1 and epan directories. Change-Id: I4043b0931d4353d60cffbd829e30269eb8d08cf4 Reviewed-on: https://code.wireshark.org/review/11200 Petri-Dish: Michal Labedzki <michal.labedzki@tieto.com> Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-10-21 19:04:16 +00:00
set_address(&addr, AT_IPv6, 16, value.bytes);
Replace ip6_to_str and ip6_guint8_to_str with address_to_str and tvb_ip_to_str. Change-Id: I1d258923a7a63539ec8456d3e306bca5016a1e4b Reviewed-on: https://code.wireshark.org/review/6060 Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-12-21 17:38:13 +00:00
diam_sub_dis->avp_str = wmem_strdup_printf(wmem_packet_scope(), "%s/%u", address_to_str(wmem_packet_scope(), &addr), prefix_len);
Display framed_ipv6_prefix as an IPv6 address even when it's not the full 16 bits. Change-Id: I6cef08e2da9467cb54b5d08526c77c87e7afe1fa Reviewed-on: https://code.wireshark.org/review/645 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-14 12:55:34 +00:00
}
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
Whitespace cleanup: Indentation, trailing, ... svn path=/trunk/; revision=42266
2012-04-26 16:05:05 +00:00
return(prefix_len_bytes+2);
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
}
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
/* AVP Code: 1 User-Name */
/* Do special decoding of the User-Name depending on the interface */
static int
dissect_diameter_user_name(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
{
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
diam_sub_dis_t *diam_sub_dis = (diam_sub_dis_t*)data;
guint32 application_id = 0, str_len;
if (diam_sub_dis) {
application_id = diam_sub_dis->application_id;
}
switch (application_id) {
case DIAM_APPID_3GPP_S6A_S6D:
case DIAM_APPID_3GPP_SLH:
case DIAM_APPID_3GPP_S7A:
str_len = tvb_reported_length(tvb);
dissect_e212_utf8_imsi(tvb, pinfo, tree, 0, str_len);
return str_len;
}
return 0;
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
}
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
/* AVP Code: 124 MIP6-Feature-Vector */
/* RFC 5447, 5779 */
/* 3GPP TS 29.273, V15.2.0 */
static int
dissect_diameter_mip6_feature_vector(tvbuff_t *tvb, packet_info *pinfo _U_, proto_tree *tree, void *data)
{
Fix the type of arrays of pointers to hf_ values for bitfield routines. The static arrays are supposed to be arrays of const pointers to int, not arrays of non-const pointers to const int. Fixing that means some bugs (scribbling on what's *supposed* to be a const array) will be caught (see packet-ieee80211-radiotap.c for examples, the first of which inspired this change and the second of which was discovered while testing compiles with this change), and removes the need for some annoying casts. Also make some of those arrays static while we're at it. Update documentation and dissector-generator tools. Change-Id: I789da5fc60aadc15797cefecfd9a9fbe9a130ccc Reviewed-on: https://code.wireshark.org/review/37517 Petri-Dish: Guy Harris <gharris@sonic.net> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-06-19 01:14:46 +00:00
static int * const flags_rfc[] = {
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
&hf_diameter_mip6_feature_vector_mip6_integrated,
&hf_diameter_mip6_feature_vector_local_home_agent_assignment,
&hf_diameter_mip6_feature_vector_pmip6_supported,
&hf_diameter_mip6_feature_vector_ip4_hoa_supported,
&hf_diameter_mip6_feature_vector_local_mag_routing_supported,
NULL
};
Fix the type of arrays of pointers to hf_ values for bitfield routines. The static arrays are supposed to be arrays of const pointers to int, not arrays of non-const pointers to const int. Fixing that means some bugs (scribbling on what's *supposed* to be a const array) will be caught (see packet-ieee80211-radiotap.c for examples, the first of which inspired this change and the second of which was discovered while testing compiles with this change), and removes the need for some annoying casts. Also make some of those arrays static while we're at it. Update documentation and dissector-generator tools. Change-Id: I789da5fc60aadc15797cefecfd9a9fbe9a130ccc Reviewed-on: https://code.wireshark.org/review/37517 Petri-Dish: Guy Harris <gharris@sonic.net> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-06-19 01:14:46 +00:00
static int * const flags_3gpp[] = {
&hf_diameter_3gpp_mip6_feature_vector_assign_local_ip,
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
&hf_diameter_3gpp_mip6_feature_vector_mip4_supported,
&hf_diameter_3gpp_mip6_feature_vector_optimized_idle_mode_mobility,
&hf_diameter_3gpp_mip6_feature_vector_gtpv2_supported,
NULL
};
guint32 application_id = 0;
diam_sub_dis_t *diam_sub_dis_inf = (diam_sub_dis_t*)data;
Remove unnecessary null check in diameter dissector dissect_diameter_mip6_feature_vector is checking whether the data argument (assigned to diam_sub_dis_inf) is null, but later dereferences it outside the conditional, so if it was null it would crash anyway. It doesn't seem possible for the data argument to actually be null, so this commit removes the redundant check. I'm also adding an assert to document the non-null assumption. Bug found by clang static analyzer. Fixes #17427.
2021-06-05 23:54:29 +00:00
DISSECTOR_ASSERT(diam_sub_dis_inf);
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
Remove unnecessary null check in diameter dissector dissect_diameter_mip6_feature_vector is checking whether the data argument (assigned to diam_sub_dis_inf) is null, but later dereferences it outside the conditional, so if it was null it would crash anyway. It doesn't seem possible for the data argument to actually be null, so this commit removes the redundant check. I'm also adding an assert to document the non-null assumption. Bug found by clang static analyzer. Fixes #17427.
2021-06-05 23:54:29 +00:00
application_id = diam_sub_dis_inf->application_id;
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
/* Hide the item created in packet-diameter.c and only show the one created here */
proto_item_set_hidden(diam_sub_dis_inf->item);
/* Dissect values defined in RFC 5447, 5779 */
proto_tree_add_bitmask(tree, tvb, 0, hf_diameter_mip6_feature_vector, ett_diameter_mip6_feature_vector, flags_rfc, ENC_BIG_ENDIAN);
switch (application_id) {
case DIAM_APPID_3GPP_STA:
case DIAM_APPID_3GPP_SWM:
case DIAM_APPID_3GPP_SWX:
case DIAM_APPID_3GPP_S6B:
/* Dissect values defined in TGPP TS 29.273, V15.2.0 */
proto_tree_add_bitmask(tree, tvb, 0, hf_diameter_3gpp_mip6_feature_vector, ett_diameter_3gpp_mip6_feature_vector, flags_3gpp, ENC_BIG_ENDIAN);
break;
}
return 8;
}
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
/* AVP Code: 443 Subscription-Id */
static int
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
dissect_diameter_subscription_id(tvbuff_t *tvb _U_, packet_info *pinfo _U_, proto_tree *tree _U_, void *data)
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
{
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
/* Just reset our global subscription-id-type variable */
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
diam_sub_dis_t *diam_sub_dis_inf = (diam_sub_dis_t*)data;
diam_sub_dis_inf->subscription_id_type = SUBSCRIPTION_ID_TYPE_UNKNOWN;
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
return 0;
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
}
/* AVP Code: 450 Subscription-Id-Type */
static int
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
dissect_diameter_subscription_id_type(tvbuff_t *tvb, packet_info *pinfo _U_, proto_tree *tree _U_, void *data)
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
{
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
diam_sub_dis_t *diam_sub_dis_inf = (diam_sub_dis_t*)data;
diam_sub_dis_inf->subscription_id_type = tvb_get_ntohl(tvb, 0);
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
return 0;
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
}
/* AVP Code: 444 Subscription-Id-Data */
static int
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
dissect_diameter_subscription_id_data(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
{
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
guint32 str_len;
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
diam_sub_dis_t *diam_sub_dis_inf = (diam_sub_dis_t*)data;
guint32 subscription_id_type = diam_sub_dis_inf->subscription_id_type;
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
switch (subscription_id_type) {
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
case SUBSCRIPTION_ID_TYPE_IMSI:
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
str_len = tvb_reported_length(tvb);
dissect_e212_utf8_imsi(tvb, pinfo, tree, 0, str_len);
return str_len;
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
case SUBSCRIPTION_ID_TYPE_E164:
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
str_len = tvb_reported_length(tvb);
dissect_e164_msisdn(tvb, tree, 0, str_len, E164_ENC_UTF8);
return str_len;
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
}
return 0;
}
/* AVP Code: 458 User-Equipment-Info */
static int
dissect_diameter_user_equipment_info(tvbuff_t *tvb _U_, packet_info *pinfo _U_, proto_tree *tree _U_, void *data)
{
/* Just reset our global subscription-id-type variable */
diam_sub_dis_t *diam_sub_dis_inf = (diam_sub_dis_t*)data;
diam_sub_dis_inf->user_equipment_info_type = USER_EQUIPMENT_INFO_TYPE_UNKNOWN;
return 0;
}
/* AVP Code: 459 User-Equipment-Info-Type */
/* RFC 8506 section 8.50 */
static int
dissect_diameter_user_equipment_info_type(tvbuff_t *tvb, packet_info *pinfo _U_, proto_tree *tree _U_, void *data)
{
diam_sub_dis_t *diam_sub_dis_inf = (diam_sub_dis_t*)data;
diam_sub_dis_inf->user_equipment_info_type = tvb_get_ntohl(tvb, 0);
return 0;
}
/* AVP Code: 460 User-Equipment-Info-Value */
/* RFC 8506 section 8.51 */
static int
dissect_diameter_user_equipment_info_value(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
{
guint32 len;
diam_sub_dis_t *diam_sub_dis_inf = (diam_sub_dis_t*)data;
guint32 user_equipment_info_type = diam_sub_dis_inf->user_equipment_info_type;
switch (user_equipment_info_type) {
case USER_EQUIPMENT_INFO_TYPE_IMEISV:
/* RFC 8506 section 8.53, 3GPP TS 23.003 */
len = tvb_reported_length(tvb);
Diameter: improve IMEI/IMEISV dissection in User-Equipment-Info AVP 3GPP.User-Equipment-Info AVP should contain the IMEISV (if type is 0). However some implementations put the IMEI there: decode it but display a protocol error indication too.
2022-03-19 12:44:56 +00:00
/* IMEISV is 16 digits, but often transmitted BCD coded in 8 octets.
Some implementations use IMEI (15 digits) instead of IMEISV */
DIAMETER: Process BCD encoded IMEISV in User-Equipment-Info Some example captures of DIAMETER show the IMEISV in User-Equipment-Info being BCD encoded (as commonly seen in other protocols.) If the number of octets used is 8, assume it is BCD encoded. If the number of octets used is 16, assume it is ASCII. Otherwise, set an expert info for the wrong length like the other types. Use the defined constants to verify the lengths of the MAC, EUI64, and MODIFIED_EUI64 types. Correct the length of the Modified EUI64 type, as it is also 8 bytes, the same as EUI64 (there must have been confusion with IPv6 addresses, which contain a Modified EUI64 in their 8 least significant bytes.)
2021-03-31 21:58:30 +00:00
if (len == 8) {
proto_tree_add_item(tree, hf_diameter_user_equipment_info_imeisv, tvb, 0, len, ENC_BCD_DIGITS_0_9|ENC_NA);
return len;
} else if (len == 16) {
Tools: Fix fix-encoding-args.pl ASCII string validation Do not require a useless ENC_NA parameter for string encodings. FT_STRING and FT_STRINGZ types don't have any ndianness. Follow-up to 6ec429622c9258eefd388caf21ce92ab5b9f54b4.
2022-02-15 02:45:20 +00:00
proto_tree_add_item(tree, hf_diameter_user_equipment_info_imeisv, tvb, 0, len, ENC_ASCII);
DIAMETER: Process BCD encoded IMEISV in User-Equipment-Info Some example captures of DIAMETER show the IMEISV in User-Equipment-Info being BCD encoded (as commonly seen in other protocols.) If the number of octets used is 8, assume it is BCD encoded. If the number of octets used is 16, assume it is ASCII. Otherwise, set an expert info for the wrong length like the other types. Use the defined constants to verify the lengths of the MAC, EUI64, and MODIFIED_EUI64 types. Correct the length of the Modified EUI64 type, as it is also 8 bytes, the same as EUI64 (there must have been confusion with IPv6 addresses, which contain a Modified EUI64 in their 8 least significant bytes.)
2021-03-31 21:58:30 +00:00
return len;
Diameter: improve IMEI/IMEISV dissection in User-Equipment-Info AVP 3GPP.User-Equipment-Info AVP should contain the IMEISV (if type is 0). However some implementations put the IMEI there: decode it but display a protocol error indication too.
2022-03-19 12:44:56 +00:00
} else if (len == 15) {
proto_tree_add_item(tree, hf_diameter_user_equipment_info_imeisv, tvb, 0, len, ENC_ASCII);
proto_tree_add_expert(tree, pinfo, &ei_diameter_unexpected_imei_as_user_equipment_info, tvb, 0, len);
return len;
DIAMETER: Process BCD encoded IMEISV in User-Equipment-Info Some example captures of DIAMETER show the IMEISV in User-Equipment-Info being BCD encoded (as commonly seen in other protocols.) If the number of octets used is 8, assume it is BCD encoded. If the number of octets used is 16, assume it is ASCII. Otherwise, set an expert info for the wrong length like the other types. Use the defined constants to verify the lengths of the MAC, EUI64, and MODIFIED_EUI64 types. Correct the length of the Modified EUI64 type, as it is also 8 bytes, the same as EUI64 (there must have been confusion with IPv6 addresses, which contain a Modified EUI64 in their 8 least significant bytes.)
2021-03-31 21:58:30 +00:00
}
proto_tree_add_expert(tree, pinfo, &ei_diameter_invalid_user_equipment_info_value_len, tvb, 0, len);
break;
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
case USER_EQUIPMENT_INFO_TYPE_MAC:
/* RFC 8506 section 8.54, RFC 5777 section 4.1.7.8 */
len = tvb_reported_length(tvb);
DIAMETER: Process BCD encoded IMEISV in User-Equipment-Info Some example captures of DIAMETER show the IMEISV in User-Equipment-Info being BCD encoded (as commonly seen in other protocols.) If the number of octets used is 8, assume it is BCD encoded. If the number of octets used is 16, assume it is ASCII. Otherwise, set an expert info for the wrong length like the other types. Use the defined constants to verify the lengths of the MAC, EUI64, and MODIFIED_EUI64 types. Correct the length of the Modified EUI64 type, as it is also 8 bytes, the same as EUI64 (there must have been confusion with IPv6 addresses, which contain a Modified EUI64 in their 8 least significant bytes.)
2021-03-31 21:58:30 +00:00
if (len == FT_ETHER_LEN) {
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
proto_tree_add_item(tree, hf_diameter_user_equipment_info_mac, tvb, 0, len, ENC_NA);
return len;
}
proto_tree_add_expert(tree, pinfo, &ei_diameter_invalid_user_equipment_info_value_len, tvb, 0, len);
break;
case USER_EQUIPMENT_INFO_TYPE_EUI64:
/* RFC 8506 section 8.55 */
len = tvb_reported_length(tvb);
DIAMETER: Process BCD encoded IMEISV in User-Equipment-Info Some example captures of DIAMETER show the IMEISV in User-Equipment-Info being BCD encoded (as commonly seen in other protocols.) If the number of octets used is 8, assume it is BCD encoded. If the number of octets used is 16, assume it is ASCII. Otherwise, set an expert info for the wrong length like the other types. Use the defined constants to verify the lengths of the MAC, EUI64, and MODIFIED_EUI64 types. Correct the length of the Modified EUI64 type, as it is also 8 bytes, the same as EUI64 (there must have been confusion with IPv6 addresses, which contain a Modified EUI64 in their 8 least significant bytes.)
2021-03-31 21:58:30 +00:00
if (len == FT_EUI64_LEN) {
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
proto_tree_add_item(tree, hf_diameter_user_equipment_info_eui64, tvb, 0, len, ENC_BIG_ENDIAN);
return len;
}
proto_tree_add_expert(tree, pinfo, &ei_diameter_invalid_user_equipment_info_value_len, tvb, 0, len);
break;
case USER_EQUIPMENT_INFO_TYPE_MODIFIED_EUI64:
/* RFC 8506 section 8.56, RFC 4291 */
len = tvb_reported_length(tvb);
DIAMETER: Process BCD encoded IMEISV in User-Equipment-Info Some example captures of DIAMETER show the IMEISV in User-Equipment-Info being BCD encoded (as commonly seen in other protocols.) If the number of octets used is 8, assume it is BCD encoded. If the number of octets used is 16, assume it is ASCII. Otherwise, set an expert info for the wrong length like the other types. Use the defined constants to verify the lengths of the MAC, EUI64, and MODIFIED_EUI64 types. Correct the length of the Modified EUI64 type, as it is also 8 bytes, the same as EUI64 (there must have been confusion with IPv6 addresses, which contain a Modified EUI64 in their 8 least significant bytes.)
2021-03-31 21:58:30 +00:00
if (len == FT_EUI64_LEN) {
proto_tree_add_item(tree, hf_diameter_user_equipment_info_modified_eui64, tvb, 0, len, ENC_BIG_ENDIAN);
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
return len;
}
proto_tree_add_expert(tree, pinfo, &ei_diameter_invalid_user_equipment_info_value_len, tvb, 0, len);
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
break;
}
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
return 0;
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
}
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
/* Call subdissectors for AVPs.
* This is a separate function to avoid having any local variables that might
* get clobbered by the exception longjmp() (without having to declare the
* variables as volatile and deal with casting them).
*/
static void
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
call_avp_subdissector(guint32 vendorid, guint32 code, tvbuff_t *subtvb, packet_info *pinfo, proto_tree *avp_tree, diam_sub_dis_t *diam_sub_dis_inf)
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
{
TRY {
switch (vendorid) {
case 0:
Use dissector_try_uint_new() and include diam_sub_dis_inf when calling all AVP sub dissectors. Change-Id: I6dcc362ea755794d3e18d8d27ed2128b46bd887b Reviewed-on: https://code.wireshark.org/review/642 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-13 20:08:57 +00:00
dissector_try_uint_new(diameter_dissector_table, code, subtvb, pinfo, avp_tree, FALSE, diam_sub_dis_inf);
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
break;
case VENDOR_ERICSSON:
Use dissector_try_uint_new() and include diam_sub_dis_inf when calling all AVP sub dissectors. Change-Id: I6dcc362ea755794d3e18d8d27ed2128b46bd887b Reviewed-on: https://code.wireshark.org/review/642 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-13 20:08:57 +00:00
dissector_try_uint_new(diameter_ericsson_avp_dissector_table, code, subtvb, pinfo, avp_tree, FALSE, diam_sub_dis_inf);
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
break;
[Diameter] Add a table for subdissectiong AVPs for VENDOR_VERIZON Change-Id: I37397e590f294d209b89fe868b365cafb5343464 Reviewed-on: https://code.wireshark.org/review/21363 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2017-04-27 11:03:42 +00:00
case VENDOR_VERIZON:
dissector_try_uint_new(diameter_verizon_avp_dissector_table, code, subtvb, pinfo, avp_tree, FALSE, diam_sub_dis_inf);
break;
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
case VENDOR_THE3GPP:
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
dissector_try_uint_new(diameter_3gpp_avp_dissector_table, code, subtvb, pinfo, avp_tree, FALSE, diam_sub_dis_inf);
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
break;
default:
break;
}
/* Debug
Eliminate proto_tree_add_text from some dissectors. Other minor cleanup while in the area. Change-Id: Id8d957d3d68a2e3dd5089f490bd59d773e1be967 Reviewed-on: https://code.wireshark.org/review/3427 Reviewed-by: Michael Mann <mmann78@netscape.net> Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-08-04 01:10:24 +00:00
proto_tree_add_subtree(avp_tree, subtvb, 0, -1, "AVP %u data, Vendor Id %u ",code,vendorid);
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
*/
}
CATCH_NONFATAL_ERRORS {
show_exception(subtvb, pinfo, avp_tree, EXCEPT_CODE, GET_MESSAGE);
}
ENDTRY;
}
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
/* Dissect an AVP at offset */
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
static int
EAP, Diameter: add EAP-TTLS and EAP-Message dissection support Add support for dissecting the decrypted TLS payload as Diameter. Add support for dissecting the EAP-Message attribute as EAP. Disable retransmission detection when EAP-Message is detected (EAP in TLS in EAP) since this results in false positives. Tested with captures from Bug 15603: * eap-ttls-pap.pcapng - ok, User-Name and User-Password AVPs. * eap-ttls-eap-gtc.pcapng, eap-ttls-eap-md5.pcapng - EAP-Message AVP. * eap-ttls-mschapv2.pcapng - partially supported, does not conform to Diameter AVP requirements as it is not padded. Microsoft vendor types are also not yet supported. To be fixed later. * eapttls-diameter-avp.pcapng (Bug 12880) - EAP-Message AVP. Bug: 12880 Bug: 15603 Change-Id: Ie7ea282d05c1d3ff8463c34bf259107562714440 Reviewed-on: https://code.wireshark.org/review/34281 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 01:47:09 +00:00
dissect_diameter_avp(diam_ctx_t *c, tvbuff_t *tvb, int offset, diam_sub_dis_t *diam_sub_dis_inf, gboolean update_col_info)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Rework "extended value strings": - Allow direct access when a range of values begins with a value other than 0; - Provide value_string_ext_new() for creating extended value strings at runtime; - Do access to value_string_ext members via a macro (all but value_string.c); - Update documentation. svn path=/trunk/; revision=34514
2010-10-14 17:50:35 +00:00
guint32 code = tvb_get_ntohl(tvb,offset);
guint32 len = tvb_get_ntohl(tvb,offset+4);
guint32 vendor_flag = len & 0x80000000;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
guint32 flags_bits_idx = (len & 0xE0000000) >> 29;
Rework "extended value strings": - Allow direct access when a range of values begins with a value other than 0; - Provide value_string_ext_new() for creating extended value strings at runtime; - Do access to value_string_ext members via a macro (all but value_string.c); - Update documentation. svn path=/trunk/; revision=34514
2010-10-14 17:50:35 +00:00
guint32 flags_bits = (len & 0xFF000000) >> 24;
guint32 vendorid = vendor_flag ? tvb_get_ntohl(tvb,offset+8) : 0 ;
Fix a couple of warning: initializer element is not computable at load time svn path=/trunk/; revision=54820
2014-01-15 19:29:03 +00:00
wmem_tree_key_t k[3];
Fix a bug intrduced in http://anonsvn.wireshark.org/viewvc/viewvc.cgi?view=rev&revision=54820 svn path=/trunk/; revision=54831
2014-01-17 18:55:49 +00:00
diam_avp_t *a;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
proto_item *pi, *avp_item;
proto_tree *avp_tree, *save_tree;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
tvbuff_t *subtvb;
diam_vnd_t *vendor;
const char *code_str;
Fix [ 5%] Building C object epan/CMakeFiles/epan.dir/dissectors/packet-diameter.c.o /home/jmayer/work/wireshark/svn/trunk/epan/dissectors/packet-diameter.c: In function ‘dissect_diameter_avp’: /home/jmayer/work/wireshark/svn/trunk/epan/dissectors/packet-diameter.c:621:37: error: ‘avp_str’ may be used uninitialized in this function [-Werror=maybe-uninitialized] if (avp_str) proto_item_append_text(avp_item," val=%s", avp_str); ^ cc1: all warnings being treated as errors make[2]: *** [epan/CMakeFiles/epan.dir/dissectors/packet-diameter.c.o] Error 1 svn path=/trunk/; revision=53787
2013-12-04 21:32:16 +00:00
const char *avp_str = NULL;
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
guint8 pad_len;
Fix a couple of warning: initializer element is not computable at load time svn path=/trunk/; revision=54820
2014-01-15 19:29:03 +00:00
k[0].length = 1;
k[0].key = &code;
k[1].length = 1;
k[1].key = &vendorid;
k[2].length = 0;
k[2].key = NULL;
Fix a bug intrduced in http://anonsvn.wireshark.org/viewvc/viewvc.cgi?view=rev&revision=54820 svn path=/trunk/; revision=54831
2014-01-17 18:55:49 +00:00
a = (diam_avp_t *)wmem_tree_lookup32_array(dictionary.avps,k);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
len &= 0x00ffffff;
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
pad_len = (len % 4) ? 4 - (len % 4) : 0 ;
Removed trailing whitespaces from .h and .c files using the winapi_cleanup tool written by Patrik Stridvall for the wine project. svn path=/trunk/; revision=6117
2002-08-28 21:04:11 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
if (!a) {
a = &unknown_avp;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
if (vendor_flag) {
Use epan-scoped memory for building more of the diameter table stuff. svn path=/trunk/; revision=51754
2013-09-04 21:06:56 +00:00
if (! (vendor = (diam_vnd_t *)wmem_tree_lookup32(dictionary.vnds,vendorid) ))
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
vendor = &unknown_vendor;
} else {
vendor = &no_vnd;
}
} else {
Rework "extended value strings": - Allow direct access when a range of values begins with a value other than 0; - Provide value_string_ext_new() for creating extended value strings at runtime; - Do access to value_string_ext members via a macro (all but value_string.c); - Update documentation. svn path=/trunk/; revision=34514
2010-10-14 17:50:35 +00:00
vendor = (diam_vnd_t *)a->vendor;
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
if (vendor->vs_avps_ext == NULL) {
diameter: change GArray into wmem_array. This change fixes a leak in packet-diameter that loads a dictionary but doesn't free all the data. Found by valgrind. ==30481== 36,656 (960 direct, 35,696 indirect) bytes in 24 blocks are definitely lost in loss record 3,417 of 3,421 ==30481== at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==30481== by 0xA7FE610: g_malloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA81422D: g_slice_alloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA7CDC44: g_array_sized_new (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0x6863743: dictionary_load (packet-diameter.c:1980) ==30481== by 0x6863743: proto_register_diameter (packet-diameter.c:2344) ==30481== by 0x71C4BA4: register_all_protocols (register.c:323) ==30481== by 0x65EEFA7: proto_init (proto.c:521) ==30481== by 0x65CD621: epan_init (epan.c:126) ==30481== by 0x115330: main (tshark.c:1220) Change-Id: I3c0d19e1accab415355aa0f50c598f0c83356985 Reviewed-on: https://code.wireshark.org/review/13821 Petri-Dish: Dario Lombardo <lomato@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-02-18 10:30:41 +00:00
wmem_array_sort(vendor->vs_avps, compare_avps);
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
vendor->vs_avps_ext = value_string_ext_new(VND_AVP_VS(vendor),
VND_AVP_VS_LEN(vendor)+1,
Revert "Revert "diameter: fix 400kb leaked memory on exit"" This reverts commit 5fea2b5f4198f1a36f313ef38532ddffd02ac5b1. I.e., it puts back the change; the reverted version passed the tests on which the versions with this change crashed. Change-Id: Idcc0eb11588cf14e2fe666de1905ee63917b0fcf Reviewed-on: https://code.wireshark.org/review/17413 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-08-31 08:16:24 +00:00
wmem_strdup_printf(wmem_epan_scope(), "diameter_vendor_%s",
Parse enterprise-numbers at run time "enterprise-numbers" is converted to tab-separated values and renamed "enterprises". Unused fields are stripped. PENs are stored in a hash table loaded at run-time. User "enterprises" file is loaded from the personal config dir. Misc make-sminmpec.pl improvements and fixes. Note: names of type "Entity (formerly ...)" have the formerly part commented out for a cleaner output. Change-Id: I60c533afbe3e399077fbf432088064471ad3e1e2 Reviewed-on: https://code.wireshark.org/review/22246 Petri-Dish: João Valverde <j@v6e.pt> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Reviewed-by: João Valverde <j@v6e.pt>
2017-06-24 13:25:41 +00:00
enterprises_lookup(vendorid, "Unknown")));
Use value_string_ext for AVP:s svn path=/trunk/; revision=32932
2010-05-24 14:20:53 +00:00
#if 0
Rework "extended value strings": - Allow direct access when a range of values begins with a value other than 0; - Provide value_string_ext_new() for creating extended value strings at runtime; - Do access to value_string_ext members via a macro (all but value_string.c); - Update documentation. svn path=/trunk/; revision=34514
2010-10-14 17:50:35 +00:00
{ /* Debug code */
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
value_string *vendor_avp_vs = VALUE_STRING_EXT_VS_P(vendor->vs_avps_ext);
Rework "extended value strings": - Allow direct access when a range of values begins with a value other than 0; - Provide value_string_ext_new() for creating extended value strings at runtime; - Do access to value_string_ext members via a macro (all but value_string.c); - Update documentation. svn path=/trunk/; revision=34514
2010-10-14 17:50:35 +00:00
gint i = 0;
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
while (vendor_avp_vs[i].strptr != NULL) {
dissectors: Replace g_log() with ws_log()
2021-06-15 15:03:55 +00:00
ws_warning("%u %s", vendor_avp_vs[i].value, vendor_avp_vs[i].strptr);
Rework "extended value strings": - Allow direct access when a range of values begins with a value other than 0; - Provide value_string_ext_new() for creating extended value strings at runtime; - Do access to value_string_ext members via a macro (all but value_string.c); - Update documentation. svn path=/trunk/; revision=34514
2010-10-14 17:50:35 +00:00
i++;
}
Use value_string_ext for AVP:s svn path=/trunk/; revision=32932
2010-05-24 14:20:53 +00:00
}
#endif
}
[Diameter] Improve dissection of malformed packets by continnuing dissection and display the problem more prominetly. Change-Id: Ia1a32667a18e1e5b60b5c167da9b6dd945ba3dfc Reviewed-on: https://code.wireshark.org/review/17385 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-08-30 12:55:54 +00:00
/* Check if the length is sane */
if (len > (guint32)tvb_reported_length_remaining(tvb, offset)) {
proto_tree_add_expert_format(c->tree, c->pinfo, &ei_diameter_invalid_avp_len, tvb, offset + 4, 4,
"Wrong AVP(%u) length %u",
code,
len);
return tvb_reported_length(tvb);
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Diameter: Add MS-CHAP/MS-CHAPv2 AVP support for EAP-TLS Created the Microsoft Diameter file based on MS-CHAP-* AVPs listed at https://www.iana.org/assignments/eap-numbers/eap-numbers.xhtml#eap-numbers-10 Many values are displayed as bytes for simplicit. The MS-CHAP2-Success attribute could for example be dissected further as 1 byte followed by a string, but that requires more effort. Allow padding to be missing since the eap-ttls-mschapv2.pcapng capture would throw a Malformed Packet exception otherwise. Bug: 15603 Change-Id: I9efc322a86802e78bb6cd4bc3df1c1282a45fe9e Reviewed-on: https://code.wireshark.org/review/34291 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 23:19:33 +00:00
/*
* Workaround for a MS-CHAPv2 capture from Bug 15603 that lacks padding.
*/
if (tvb_reported_length_remaining(tvb, offset + len) < pad_len) {
pad_len = (guint32)tvb_reported_length_remaining(tvb, offset + len);
}
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
/* Add root of tree for this AVP */
Include padding length in the AVP's length. svn path=/trunk/; revision=48702
2013-04-02 18:15:42 +00:00
avp_item = proto_tree_add_item(c->tree, hf_diameter_avp, tvb, offset, len + pad_len, ENC_NA);
avp_tree = proto_item_add_subtree(avp_item, a->ett);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Convert 'encoding' parameter of certain proto_tree_add_item() calls in non-autogenerated epan/dissectors: Specifically: Replace FALSE|0 and TRUE|1 by ENC_BIG_ENDIAN|ENC_LITTLE_ENDIAN as the encoding parameter for proto_tree_add_item() calls which directly reference an item in hf[] which has a type of: FT_UINT8 FT_UINT16 FT_UINT24 FT_UINT32 FT_UINT64 FT_INT8 FT_INT16 FT_INT24 FT_INT32 FT_INT64 FT_FLOAT FT_DOUBLE svn path=/trunk/; revision=39288
2011-10-06 03:35:44 +00:00
pi = proto_tree_add_item(avp_tree,hf_diameter_avp_code,tvb,offset,4,ENC_BIG_ENDIAN);
Rework "extended value strings": - Allow direct access when a range of values begins with a value other than 0; - Provide value_string_ext_new() for creating extended value strings at runtime; - Do access to value_string_ext members via a macro (all but value_string.c); - Update documentation. svn path=/trunk/; revision=34514
2010-10-14 17:50:35 +00:00
code_str = val_to_str_ext_const(code, vendor->vs_avps_ext, "Unknown");
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
proto_item_append_text(pi," %s", code_str);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
/* Code */
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
if (a == &unknown_avp) {
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_tree *tu = proto_item_add_subtree(pi,ett_unknown);
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
proto_tree_add_expert_format(tu, c->pinfo, &ei_diameter_avp_code, tvb, offset, 4,
"Unknown AVP %u (vendor=%s), if you know what this is you can add it to dictionary.xml", code,
Parse enterprise-numbers at run time "enterprise-numbers" is converted to tab-separated values and renamed "enterprises". Unused fields are stripped. PENs are stored in a hash table loaded at run-time. User "enterprises" file is loaded from the personal config dir. Misc make-sminmpec.pl improvements and fixes. Note: names of type "Entity (formerly ...)" have the formerly part commented out for a cleaner output. Change-Id: I60c533afbe3e399077fbf432088064471ad3e1e2 Reviewed-on: https://code.wireshark.org/review/22246 Petri-Dish: João Valverde <j@v6e.pt> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Reviewed-by: João Valverde <j@v6e.pt>
2017-06-24 13:25:41 +00:00
enterprises_lookup(vendorid, "Unknown"));
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
}
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
offset += 4;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
proto_item_set_text(avp_item,"AVP: %s(%u) l=%u f=%s", code_str, code, len, avpflags_str[flags_bits_idx]);
EAP, Diameter: add EAP-TTLS and EAP-Message dissection support Add support for dissecting the decrypted TLS payload as Diameter. Add support for dissecting the EAP-Message attribute as EAP. Disable retransmission detection when EAP-Message is detected (EAP in TLS in EAP) since this results in false positives. Tested with captures from Bug 15603: * eap-ttls-pap.pcapng - ok, User-Name and User-Password AVPs. * eap-ttls-eap-gtc.pcapng, eap-ttls-eap-md5.pcapng - EAP-Message AVP. * eap-ttls-mschapv2.pcapng - partially supported, does not conform to Diameter AVP requirements as it is not padded. Microsoft vendor types are also not yet supported. To be fixed later. * eapttls-diameter-avp.pcapng (Bug 12880) - EAP-Message AVP. Bug: 12880 Bug: 15603 Change-Id: Ie7ea282d05c1d3ff8463c34bf259107562714440 Reviewed-on: https://code.wireshark.org/review/34281 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 01:47:09 +00:00
if (update_col_info) {
col_append_fstr(c->pinfo->cinfo, COL_INFO, " %s", code_str);
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
/* Flags */
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{
Fix the type of arrays of pointers to hf_ values for bitfield routines. The static arrays are supposed to be arrays of const pointers to int, not arrays of non-const pointers to const int. Fixing that means some bugs (scribbling on what's *supposed* to be a const array) will be caught (see packet-ieee80211-radiotap.c for examples, the first of which inspired this change and the second of which was discovered while testing compiles with this change), and removes the need for some annoying casts. Also make some of those arrays static while we're at it. Update documentation and dissector-generator tools. Change-Id: I789da5fc60aadc15797cefecfd9a9fbe9a130ccc Reviewed-on: https://code.wireshark.org/review/37517 Petri-Dish: Guy Harris <gharris@sonic.net> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-06-19 01:14:46 +00:00
static int * const diameter_avp_flags[] = {
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
&hf_diameter_avp_flags_vendor_specific,
&hf_diameter_avp_flags_mandatory,
&hf_diameter_avp_flags_protected,
&hf_diameter_avp_flags_reserved3,
&hf_diameter_avp_flags_reserved4,
&hf_diameter_avp_flags_reserved5,
&hf_diameter_avp_flags_reserved6,
&hf_diameter_avp_flags_reserved7,
NULL
};
pi = proto_tree_add_bitmask_with_flags(avp_tree, tvb, offset, hf_diameter_avp_flags,
ett_diameter_avp_flags, diameter_avp_flags, ENC_BIG_ENDIAN, BMT_NO_FALSE | BMT_NO_INT);
if (flags_bits & 0x1f) expert_add_info(c->pinfo, pi, &ei_diameter_reserved_bit_set);
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
offset += 1;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
/* Length */
Convert 'encoding' parameter of certain proto_tree_add_item() calls in non-autogenerated epan/dissectors: Specifically: Replace FALSE|0 and TRUE|1 by ENC_BIG_ENDIAN|ENC_LITTLE_ENDIAN as the encoding parameter for proto_tree_add_item() calls which directly reference an item in hf[] which has a type of: FT_UINT8 FT_UINT16 FT_UINT24 FT_UINT32 FT_UINT64 FT_INT8 FT_INT16 FT_INT24 FT_INT32 FT_INT64 FT_FLOAT FT_DOUBLE svn path=/trunk/; revision=39288
2011-10-06 03:35:44 +00:00
proto_tree_add_item(avp_tree,hf_diameter_avp_len,tvb,offset,3,ENC_BIG_ENDIAN);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
offset += 3;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
/* Vendor flag */
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
if (vendor_flag) {
proto_item_append_text(avp_item," vnd=%s", val_to_str(vendorid, vnd_short_vs, "%d"));
Convert 'encoding' parameter of certain proto_tree_add_item() calls in non-autogenerated epan/dissectors: Specifically: Replace FALSE|0 and TRUE|1 by ENC_BIG_ENDIAN|ENC_LITTLE_ENDIAN as the encoding parameter for proto_tree_add_item() calls which directly reference an item in hf[] which has a type of: FT_UINT8 FT_UINT16 FT_UINT24 FT_UINT32 FT_UINT64 FT_INT8 FT_INT16 FT_INT24 FT_INT32 FT_INT64 FT_FLOAT FT_DOUBLE svn path=/trunk/; revision=39288
2011-10-06 03:35:44 +00:00
pi = proto_tree_add_item(avp_tree,hf_diameter_avp_vendor_id,tvb,offset,4,ENC_BIG_ENDIAN);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
if (vendor == &unknown_vendor) {
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_tree *tu = proto_item_add_subtree(pi,ett_unknown);
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
proto_tree_add_expert(tu, c->pinfo, &ei_diameter_avp_vendor_id, tvb, offset, 4);
Minor changes to expert items svn path=/trunk/; revision=22350
2007-07-18 13:18:41 +00:00
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
offset += 4;
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
if ( len == (guint32)(vendor_flag ? 12 : 8) ) {
/* Data is empty so return now */
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
proto_tree_add_expert(avp_tree, c->pinfo, &ei_diameter_avp_no_data, tvb, offset, 0);
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
/* pad_len is always 0 in this case, but kept here for consistency */
return len+pad_len;
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
}
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
/* If we are dissecting a grouped AVP and find a Vendor Id AVP(266), save it */
Add/clarify some comments about how the Experimental-Result-Code AVP is handled. Specifically we put Experimental-Result-Codes specified by 3GPP in the XML file and let other vendors' codes be handled through a dissector table. Change-Id: I2c3977fb959ad84faa5cb90aeb3d191c8b465ede Reviewed-on: https://code.wireshark.org/review/8319 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-06 20:03:15 +00:00
if ((diam_sub_dis_inf->dis_gouped) && (!vendor_flag) && (code==266)) {
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
diam_sub_dis_inf->vendor_id = tvb_get_ntohl(tvb,offset);
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
tvb_new_subset -> tvb_new_subset_length when length parameters are equal. tvb_new_subset -> tvb_new_subset_remaining it appears that's what the intention is. Change-Id: I2334bbf3f10475b3c22391392fc8b6864454de2d Reviewed-on: https://code.wireshark.org/review/1999 Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
2014-06-06 14:35:50 +00:00
subtvb = tvb_new_subset_length(tvb,offset,len-(8+(vendor_flag?4:0)));
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
offset += len-(8+(vendor_flag?4:0));
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
save_tree = c->tree;
c->tree = avp_tree;
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
Add/clarify some comments about how the Experimental-Result-Code AVP is handled. Specifically we put Experimental-Result-Codes specified by 3GPP in the XML file and let other vendors' codes be handled through a dissector table. Change-Id: I2c3977fb959ad84faa5cb90aeb3d191c8b465ede Reviewed-on: https://code.wireshark.org/review/8319 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-06 20:03:15 +00:00
/* The Experimental-Result-Code AVP (298) comes inside the Experimental-Result
* grouped AVP (297). The Vendor-ID AVP in the Experimental-Result specifies the
* name space of the Experimental-Result-Code. Unfortunately we don't have a way
* to specify, in XML, different Experimental-Result-Code enum values for different
* Vendor-IDs so we choose a Vendor-ID whose values get to go in XML (we chose
* 3GPP) and handle other Vendor-IDs through the "diameter.vnd_exp_res" dissector
* table.
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
*/
Add/clarify some comments about how the Experimental-Result-Code AVP is handled. Specifically we put Experimental-Result-Codes specified by 3GPP in the XML file and let other vendors' codes be handled through a dissector table. Change-Id: I2c3977fb959ad84faa5cb90aeb3d191c8b465ede Reviewed-on: https://code.wireshark.org/review/8319 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-06 20:03:15 +00:00
if ((diam_sub_dis_inf->dis_gouped)
&& (!vendor_flag)
&& (code==298)
&& (diam_sub_dis_inf->vendor_id != 0)
&& (diam_sub_dis_inf->vendor_id != VENDOR_THE3GPP))
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
{
/* call subdissector */
Show the Experimental-Result-Code if we don't know have a subdissector for that vendor ID. Otherwise the value is simply not shown to the user. Adding support for a vendor ID's Experimental Result Codes isn't as easy as modifying the XML so don't add an expert info about it. Change-Id: I65f2cb13853cc7141fb242fa03c6e474a6c02cb9 Reviewed-on: https://code.wireshark.org/review/11294 Reviewed-by: Michael Mann <mmann78@netscape.net> Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-10-26 20:44:18 +00:00
if (!dissector_try_uint_new(diameter_expr_result_vnd_table, diam_sub_dis_inf->vendor_id,
subtvb, c->pinfo, avp_tree, FALSE, diam_sub_dis_inf)) {
/* No subdissector for this vendor ID, use the generic one */
dissect_diameter_other_vendor_exp_res(subtvb, avp_tree, diam_sub_dis_inf);
}
Add/clarify some comments about how the Experimental-Result-Code AVP is handled. Specifically we put Experimental-Result-Codes specified by 3GPP in the XML file and let other vendors' codes be handled through a dissector table. Change-Id: I2c3977fb959ad84faa5cb90aeb3d191c8b465ede Reviewed-on: https://code.wireshark.org/review/8319 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-06 20:03:15 +00:00
if (diam_sub_dis_inf->avp_str) {
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
proto_item_append_text(avp_item," val=%s", diam_sub_dis_inf->avp_str);
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
} else {
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
avp_str = a->dissector_rfc(c,a,subtvb, diam_sub_dis_inf);
Use VendorID from sminmpec. svn path=/trunk/; revision=16552
2005-11-19 15:59:30 +00:00
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
c->tree = save_tree;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Make the avp_str available to sub dissectors and Let the subdissector have precedence filling in the avp_item string. Use that in a couple of places. Change-Id: I1af7a1ca4c14fb56ddeaab336202e6c2a18e556b Reviewed-on: https://code.wireshark.org/review/699 Reviewed-by: Michael Mann <mmann78@netscape.net> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-16 17:57:18 +00:00
diam_sub_dis_inf->avp_str = NULL;
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
call_avp_subdissector(vendorid, code, subtvb, c->pinfo, avp_tree, diam_sub_dis_inf);
Make it possible to do subdissection of Diameter AVP:s and add dissection of some 3GPP AVP:s. svn path=/trunk/; revision=24211
2008-01-27 21:06:33 +00:00
Make the avp_str available to sub dissectors and Let the subdissector have precedence filling in the avp_item string. Use that in a couple of places. Change-Id: I1af7a1ca4c14fb56ddeaab336202e6c2a18e556b Reviewed-on: https://code.wireshark.org/review/699 Reviewed-by: Michael Mann <mmann78@netscape.net> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-16 17:57:18 +00:00
/* Let the subdissector have precedence filling in the avp_item string */
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (diam_sub_dis_inf->avp_str) {
Make the avp_str available to sub dissectors and Let the subdissector have precedence filling in the avp_item string. Use that in a couple of places. Change-Id: I1af7a1ca4c14fb56ddeaab336202e6c2a18e556b Reviewed-on: https://code.wireshark.org/review/699 Reviewed-by: Michael Mann <mmann78@netscape.net> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-16 17:57:18 +00:00
proto_item_append_text(avp_item," val=%s", diam_sub_dis_inf->avp_str);
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
} else if (avp_str) {
Make the avp_str available to sub dissectors and Let the subdissector have precedence filling in the avp_item string. Use that in a couple of places. Change-Id: I1af7a1ca4c14fb56ddeaab336202e6c2a18e556b Reviewed-on: https://code.wireshark.org/review/699 Reviewed-by: Michael Mann <mmann78@netscape.net> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-03-16 17:57:18 +00:00
proto_item_append_text(avp_item," val=%s", avp_str);
}
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
if (pad_len) {
guint8 i;
pi = proto_tree_add_item(avp_tree, hf_diameter_avp_pad, tvb, offset, pad_len, ENC_NA);
for (i=0; i < pad_len; i++) {
if (tvb_get_guint8(tvb, offset++) != 0) {
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
expert_add_info(c->pinfo, pi, &ei_diameter_avp_pad);
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
break;
}
}
}
Diameter: Add MS-CHAP/MS-CHAPv2 AVP support for EAP-TLS Created the Microsoft Diameter file based on MS-CHAP-* AVPs listed at https://www.iana.org/assignments/eap-numbers/eap-numbers.xhtml#eap-numbers-10 Many values are displayed as bytes for simplicit. The MS-CHAP2-Success attribute could for example be dissected further as 1 byte followed by a string, but that requires more effort. Allow padding to be missing since the eap-ttls-mschapv2.pcapng capture would throw a Malformed Packet exception otherwise. Bug: 15603 Change-Id: I9efc322a86802e78bb6cd4bc3df1c1282a45fe9e Reviewed-on: https://code.wireshark.org/review/34291 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 23:19:33 +00:00
if ((len + pad_len) % 4) {
proto_tree_add_expert(avp_tree, c->pinfo, &ei_diameter_avp_pad_missing, tvb, offset, pad_len);
}
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
return len+pad_len;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
Removed trailing whitespaces from .h and .c files using the winapi_cleanup tool written by Patrik Stridvall for the wine project. svn path=/trunk/; revision=6117
2002-08-28 21:04:11 +00:00
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
address_rfc_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
address_avp_t *t = (address_avp_t *)a->type_data;
Adjust indentation to match editor modelines; Do other minor whitespace changes. Change-Id: Ic020b2c92db5d14a2be9dc4d35aef4514b8b0353 Reviewed-on: https://code.wireshark.org/review/4502 Reviewed-by: Bill Meier <wmeier@newsguy.com>
2014-10-06 16:55:18 +00:00
gint len = tvb_reported_length(tvb);
Minor performance tweaks: - Only calculate tvb length once. - Use tvb_reported_length() instead of tvb_reported_length_remaining() as this is a subtvb offset is always 0. Change-Id: I03bd7a95061488d4576fa93f26e6b31d55f88738 Reviewed-on: https://code.wireshark.org/review/4060 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-09-10 13:13:46 +00:00
proto_item *pi = proto_tree_add_item(c->tree, a->hf_value, tvb, 0, len, ENC_BIG_ENDIAN);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_tree *pt = proto_item_add_subtree(pi,t->ett);
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
guint32 addr_type;
Minor performance tweaks: - Only calculate tvb length once. - Use tvb_reported_length() instead of tvb_reported_length_remaining() as this is a subtvb offset is always 0. Change-Id: I03bd7a95061488d4576fa93f26e6b31d55f88738 Reviewed-on: https://code.wireshark.org/review/4060 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-09-10 13:13:46 +00:00
len = len-2;
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
Diameter: Handle (IP)address type E164 as string The Diameter type Address hase a two byte address type family field previously only IPv4 and IPv6 was handled. Add handling of E.164 when encoded as a string.
2020-10-12 08:34:49 +00:00
proto_tree_add_item_ret_uint(pt, t->hf_address_type, tvb, 0, 2, ENC_NA, &addr_type);
/* See afn.h and https://www.iana.org/assignments/address-family-numbers/address-family-numbers.xhtml */
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
switch (addr_type ) {
Diameter: Handle (IP)address type E164 as string The Diameter type Address hase a two byte address type family field previously only IPv4 and IPv6 was handled. Add handling of E.164 when encoded as a string.
2020-10-12 08:34:49 +00:00
case AFNUM_INET:
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
if (len != 4) {
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
proto_tree_add_expert_format(pt, c->pinfo, &ei_diameter_avp_len, tvb, 2, len, "Wrong length for IPv4 Address: %d instead of 4", len);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
return "[Malformed]";
}
Amazingly, there *are* apparently protocols with IPv4 addresses in byte-swapped form - i.e., not network byte order, but the reverse of network byte order - as, a long time ago, somebody asked to have the "little_endian" flag affect the way proto_tree_add_item() fetched IPv4 addresses from the packet. Use ENC_BIG_ENDIAN for IPv4 addresses (ENC_NA and ENC_BIG_ENDIAN have the same value, but this makes it clearer that we, sadly, need to care). svn path=/trunk/; revision=39708
2011-11-02 01:53:37 +00:00
pi = proto_tree_add_item(pt,t->hf_ipv4,tvb,2,4,ENC_BIG_ENDIAN);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
break;
Diameter: Handle (IP)address type E164 as string The Diameter type Address hase a two byte address type family field previously only IPv4 and IPv6 was handled. Add handling of E.164 when encoded as a string.
2020-10-12 08:34:49 +00:00
case AFNUM_INET6:
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
if (len != 16) {
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
proto_tree_add_expert_format(pt, c->pinfo, &ei_diameter_avp_len, tvb, 2, len, "Wrong length for IPv6 Address: %d instead of 16", len);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
return "[Malformed]";
}
Set the ENC_ on the remaining proto_add_item(). svn path=/trunk/; revision=39683
2011-10-31 07:05:52 +00:00
pi = proto_tree_add_item(pt,t->hf_ipv6,tvb,2,16,ENC_NA);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
break;
Diameter: Handle (IP)address type E164 as string The Diameter type Address hase a two byte address type family field previously only IPv4 and IPv6 was handled. Add handling of E.164 when encoded as a string.
2020-10-12 08:34:49 +00:00
case AFNUM_E164:
/* It's unclear what format the e164 address would be encoded in but AVP 3GPP 2008 has
* ...value 8, E.164, and the address information is UTF8 encoded.
*/
if (tvb_ascii_isprint(tvb, 2, len)) {
pi = proto_tree_add_item(pt, t->hf_e164_str, tvb, 2, len, ENC_ASCII | ENC_NA);
} else {
pi = proto_tree_add_item(pt, t->hf_other, tvb, 2, -1, ENC_BIG_ENDIAN);
}
break;
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
default:
Set the ENC_ on the remaining proto_add_item(). svn path=/trunk/; revision=39683
2011-10-31 07:05:52 +00:00
pi = proto_tree_add_item(pt,t->hf_other,tvb,2,-1,ENC_BIG_ENDIAN);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
break;
}
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (c->tree) {
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
proto_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf)
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
{
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
proto_avp_t *t = (proto_avp_t *)a->type_data;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Allow control of individual columns to be (un)writable. Most protocols just want to limit COL_INFO or COL_PROTOCOL so give that level of granularity. Bug: 12144 Bug: 5117 Bug: 11144 Change-Id: I8de9b7d2c69e90d3fbfc0a52c2bd78c3de58e2f8 Reviewed-on: https://code.wireshark.org/review/15894 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com> Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-06-14 15:55:08 +00:00
col_set_writable(c->pinfo->cinfo, COL_PROTOCOL, FALSE);
col_set_writable(c->pinfo->cinfo, COL_INFO, FALSE);
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
if (!t->handle) {
t->handle = find_dissector(t->name);
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
if (!t->handle) t->handle = data_handle;
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
}
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
TRY {
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
call_dissector_with_data(t->handle, tvb, c->pinfo, c->tree, diam_sub_dis_inf);
Catch exceptions from AVP subdissectors in case the AVPs following the one that threw the exception are OK--this allows us to view as much of the message as possible even when Wireshark doesn't like whatever was in a particular AVP. Instead of using the 'volatile' keyword to avoid variable-clobbering warnings, put the exception-catching code in its own function with no local variables. svn path=/trunk/; revision=50728
2013-07-18 15:08:15 +00:00
}
CATCH_NONFATAL_ERRORS {
show_exception(tvb, c->pinfo, c->tree, EXCEPT_CODE, GET_MESSAGE);
}
ENDTRY;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
return "";
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
time_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
int len = tvb_reported_length(tvb);
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_item *pi;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
if ( len != 4 ) {
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
proto_tree_add_expert_format(c->tree, c->pinfo, &ei_diameter_avp_len, tvb, 0, 4,
"Bad Timestamp Length: %d instead of 4", len);
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
return "[Malformed]";
}
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (c->tree) {
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
Fix up time encodings. Add some new encodings for absolute time stamps, and use them as appropriate; this fixes some cases where the time stamps in question were being dissected incorrectly. For the encodings with seconds and 1/2^32s of a second, don't arbitrarily give only microsecond resolution; 2^32 is greater than 1 million, and, in fact, at least some NTP RFCs explicitly talk about time resolution greater than 1 microsecond. Update references in the RELOAD dissector to reflect the documents in question having been updated and published as RFCs. Change-Id: Icbe0b696d65eb622978eb71e99ddf699b84e4fca Reviewed-on: https://code.wireshark.org/review/20759 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2017-03-28 10:17:48 +00:00
pi = proto_tree_add_item(c->tree, (a->hf_value), tvb, 0, 4, ENC_TIME_SECS_NTP|ENC_BIG_ENDIAN);
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
address_radius_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
address_avp_t *t = (address_avp_t *)a->type_data;
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
proto_item *pi = proto_tree_add_item(c->tree,a->hf_value,tvb,0,tvb_reported_length(tvb),ENC_BIG_ENDIAN);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_tree *pt = proto_item_add_subtree(pi,t->ett);
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
guint32 len = tvb_reported_length(tvb);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
switch (len) {
case 4:
Amazingly, there *are* apparently protocols with IPv4 addresses in byte-swapped form - i.e., not network byte order, but the reverse of network byte order - as, a long time ago, somebody asked to have the "little_endian" flag affect the way proto_tree_add_item() fetched IPv4 addresses from the packet. Use ENC_BIG_ENDIAN for IPv4 addresses (ENC_NA and ENC_BIG_ENDIAN have the same value, but this makes it clearer that we, sadly, need to care). svn path=/trunk/; revision=39708
2011-11-02 01:53:37 +00:00
pi = proto_tree_add_item(pt,t->hf_ipv4,tvb,0,4,ENC_BIG_ENDIAN);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
break;
case 16:
Set the ENC_ on the remaining proto_add_item(). svn path=/trunk/; revision=39683
2011-10-31 07:05:52 +00:00
pi = proto_tree_add_item(pt,t->hf_ipv6,tvb,0,16,ENC_NA);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
break;
default:
Set the ENC_ on the remaining proto_add_item(). svn path=/trunk/; revision=39683
2011-10-31 07:05:52 +00:00
pi = proto_tree_add_item(pt,t->hf_other,tvb,0,len,ENC_BIG_ENDIAN);
expert_add_info_format_text -> expert_add_info_format svn path=/trunk/; revision=51852
2013-09-09 00:44:09 +00:00
expert_add_info_format(c->pinfo, pi, &ei_diameter_avp_len,
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
"Bad Address Length (%u)", len);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
break;
}
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (c->tree) {
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
simple_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
if (c->tree) {
proto_item *pi = proto_tree_add_item(c->tree,a->hf_value,tvb,0,tvb_reported_length(tvb),ENC_BIG_ENDIAN);
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Set the ENC_ on the remaining proto_add_item(). svn path=/trunk/; revision=39683
2011-10-31 07:05:52 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
utf8_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
Set the ENC_ on the remaining proto_add_item(). svn path=/trunk/; revision=39683
2011-10-31 07:05:52 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
if (c->tree) {
proto_item *pi = proto_tree_add_item(c->tree,a->hf_value,tvb,0,tvb_reported_length(tvb),ENC_UTF_8|ENC_BIG_ENDIAN);
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
return label;
}
from anders b diameter updates to add 3gpp vendor avp svn path=/trunk/; revision=14189
2005-04-26 08:15:45 +00:00
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
integer32_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_item *pi;
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
/* Verify length before adding */
Minor performance tweaks: - Only calculate tvb length once. - Use tvb_reported_length() instead of tvb_reported_length_remaining() as this is a subtvb offset is always 0. Change-Id: I03bd7a95061488d4576fa93f26e6b31d55f88738 Reviewed-on: https://code.wireshark.org/review/4060 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-09-10 13:13:46 +00:00
gint length = tvb_reported_length(tvb);
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
if (length == 4) {
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (c->tree) {
pi= proto_tree_add_item(c->tree, a->hf_value, tvb, 0, length, ENC_BIG_ENDIAN);
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
}
else {
pi = proto_tree_add_bytes_format(c->tree, hf_diameter_avp_data_wrong_length,
Modify proto_tree_add_bytes_format() and proto_tree_add_bytes_format_value() so that if the start_ptr is NULL the bytes are extracted from the given TVB using the given offset and length. Replace a bunch of: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, tvb_get_ptr(tvb, offset, length), [...]) with: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, NULL, [...]) svn path=/trunk/; revision=35896
2011-02-10 16:31:00 +00:00
tvb, 0, length, NULL,
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
"Error! Bad Integer32 Length");
expert_add_info_format_text -> expert_add_info_format svn path=/trunk/; revision=51852
2013-09-09 00:44:09 +00:00
expert_add_info_format(c->pinfo, pi, &ei_diameter_avp_len,
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
"Bad Integer32 Length (%u)", length);
epan: Convert our PROTO_ITEM_ macros to inline functions. Convert our various PROTO_ITEM_ macros to inline functions and document them. Change-Id: I070b15d4f70d2189217a177ee8ba2740be36327c Reviewed-on: https://code.wireshark.org/review/32706 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-03 21:32:30 +00:00
proto_item_set_generated(pi);
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
}
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
integer64_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_item *pi;
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
/* Verify length before adding */
Minor performance tweaks: - Only calculate tvb length once. - Use tvb_reported_length() instead of tvb_reported_length_remaining() as this is a subtvb offset is always 0. Change-Id: I03bd7a95061488d4576fa93f26e6b31d55f88738 Reviewed-on: https://code.wireshark.org/review/4060 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-09-10 13:13:46 +00:00
gint length = tvb_reported_length(tvb);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
if (length == 8) {
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (c->tree) {
pi= proto_tree_add_item(c->tree, a->hf_value, tvb, 0, length, ENC_BIG_ENDIAN);
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
else {
pi = proto_tree_add_bytes_format(c->tree, hf_diameter_avp_data_wrong_length,
Modify proto_tree_add_bytes_format() and proto_tree_add_bytes_format_value() so that if the start_ptr is NULL the bytes are extracted from the given TVB using the given offset and length. Replace a bunch of: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, tvb_get_ptr(tvb, offset, length), [...]) with: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, NULL, [...]) svn path=/trunk/; revision=35896
2011-02-10 16:31:00 +00:00
tvb, 0, length, NULL,
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
"Error! Bad Integer64 Length");
expert_add_info_format_text -> expert_add_info_format svn path=/trunk/; revision=51852
2013-09-09 00:44:09 +00:00
expert_add_info_format(c->pinfo, pi, &ei_diameter_avp_len,
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
"Bad Integer64 Length (%u)", length);
epan: Convert our PROTO_ITEM_ macros to inline functions. Convert our various PROTO_ITEM_ macros to inline functions and document them. Change-Id: I070b15d4f70d2189217a177ee8ba2740be36327c Reviewed-on: https://code.wireshark.org/review/32706 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-03 21:32:30 +00:00
proto_item_set_generated(pi);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Diameter: Prettify subdissection of bitfields. Change-Id: I818e00fc3e8e2ff3da50f2075f6e7f9ec085ca2d Reviewed-on: https://code.wireshark.org/review/32876 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-17 08:58:35 +00:00
unsigned32_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf)
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_item *pi;
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
/* Verify length before adding */
Minor performance tweaks: - Only calculate tvb length once. - Use tvb_reported_length() instead of tvb_reported_length_remaining() as this is a subtvb offset is always 0. Change-Id: I03bd7a95061488d4576fa93f26e6b31d55f88738 Reviewed-on: https://code.wireshark.org/review/4060 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-09-10 13:13:46 +00:00
gint length = tvb_reported_length(tvb);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
if (length == 4) {
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (c->tree) {
Diameter: Prettify subdissection of bitfields. Change-Id: I818e00fc3e8e2ff3da50f2075f6e7f9ec085ca2d Reviewed-on: https://code.wireshark.org/review/32876 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-17 08:58:35 +00:00
diam_sub_dis_inf->item = pi = proto_tree_add_item(c->tree, a->hf_value, tvb, 0, length, ENC_BIG_ENDIAN);
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
else {
pi = proto_tree_add_bytes_format(c->tree, hf_diameter_avp_data_wrong_length,
Modify proto_tree_add_bytes_format() and proto_tree_add_bytes_format_value() so that if the start_ptr is NULL the bytes are extracted from the given TVB using the given offset and length. Replace a bunch of: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, tvb_get_ptr(tvb, offset, length), [...]) with: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, NULL, [...]) svn path=/trunk/; revision=35896
2011-02-10 16:31:00 +00:00
tvb, 0, length, NULL,
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
"Error! Bad Unsigned32 Length");
expert_add_info_format_text -> expert_add_info_format svn path=/trunk/; revision=51852
2013-09-09 00:44:09 +00:00
expert_add_info_format(c->pinfo, pi, &ei_diameter_avp_len,
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
"Bad Unsigned32 Length (%u)", length);
epan: Convert our PROTO_ITEM_ macros to inline functions. Convert our various PROTO_ITEM_ macros to inline functions and document them. Change-Id: I070b15d4f70d2189217a177ee8ba2740be36327c Reviewed-on: https://code.wireshark.org/review/32706 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-03 21:32:30 +00:00
proto_item_set_generated(pi);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
unsigned64_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_item *pi;
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
/* Verify length before adding */
Minor performance tweaks: - Only calculate tvb length once. - Use tvb_reported_length() instead of tvb_reported_length_remaining() as this is a subtvb offset is always 0. Change-Id: I03bd7a95061488d4576fa93f26e6b31d55f88738 Reviewed-on: https://code.wireshark.org/review/4060 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-09-10 13:13:46 +00:00
gint length = tvb_reported_length(tvb);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
if (length == 8) {
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (c->tree) {
pi= proto_tree_add_item(c->tree, a->hf_value, tvb, 0, length, ENC_BIG_ENDIAN);
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
else {
pi = proto_tree_add_bytes_format(c->tree, hf_diameter_avp_data_wrong_length,
Modify proto_tree_add_bytes_format() and proto_tree_add_bytes_format_value() so that if the start_ptr is NULL the bytes are extracted from the given TVB using the given offset and length. Replace a bunch of: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, tvb_get_ptr(tvb, offset, length), [...]) with: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, NULL, [...]) svn path=/trunk/; revision=35896
2011-02-10 16:31:00 +00:00
tvb, 0, length, NULL,
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
"Error! Bad Unsigned64 Length");
expert_add_info_format_text -> expert_add_info_format svn path=/trunk/; revision=51852
2013-09-09 00:44:09 +00:00
expert_add_info_format(c->pinfo, pi, &ei_diameter_avp_len,
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
"Bad Unsigned64 Length (%u)", length);
epan: Convert our PROTO_ITEM_ macros to inline functions. Convert our various PROTO_ITEM_ macros to inline functions and document them. Change-Id: I070b15d4f70d2189217a177ee8ba2740be36327c Reviewed-on: https://code.wireshark.org/review/32706 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-03 21:32:30 +00:00
proto_item_set_generated(pi);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
float32_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_item *pi;
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
/* Verify length before adding */
Minor performance tweaks: - Only calculate tvb length once. - Use tvb_reported_length() instead of tvb_reported_length_remaining() as this is a subtvb offset is always 0. Change-Id: I03bd7a95061488d4576fa93f26e6b31d55f88738 Reviewed-on: https://code.wireshark.org/review/4060 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-09-10 13:13:46 +00:00
gint length = tvb_reported_length(tvb);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
if (length == 4) {
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (c->tree) {
pi= proto_tree_add_item(c->tree,a->hf_value, tvb, 0, length, ENC_BIG_ENDIAN);
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
else {
pi = proto_tree_add_bytes_format(c->tree, hf_diameter_avp_data_wrong_length,
Modify proto_tree_add_bytes_format() and proto_tree_add_bytes_format_value() so that if the start_ptr is NULL the bytes are extracted from the given TVB using the given offset and length. Replace a bunch of: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, tvb_get_ptr(tvb, offset, length), [...]) with: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, NULL, [...]) svn path=/trunk/; revision=35896
2011-02-10 16:31:00 +00:00
tvb, 0, length, NULL,
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
"Error! Bad Float32 Length");
expert_add_info_format_text -> expert_add_info_format svn path=/trunk/; revision=51852
2013-09-09 00:44:09 +00:00
expert_add_info_format(c->pinfo, pi, &ei_diameter_avp_len,
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
"Bad Float32 Length (%u)", length);
epan: Convert our PROTO_ITEM_ macros to inline functions. Convert our various PROTO_ITEM_ macros to inline functions and document them. Change-Id: I070b15d4f70d2189217a177ee8ba2740be36327c Reviewed-on: https://code.wireshark.org/review/32706 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-03 21:32:30 +00:00
proto_item_set_generated(pi);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
float64_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf _U_)
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
{
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
char *label = NULL;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_item *pi;
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
/* Verify length before adding */
Minor performance tweaks: - Only calculate tvb length once. - Use tvb_reported_length() instead of tvb_reported_length_remaining() as this is a subtvb offset is always 0. Change-Id: I03bd7a95061488d4576fa93f26e6b31d55f88738 Reviewed-on: https://code.wireshark.org/review/4060 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-09-10 13:13:46 +00:00
gint length = tvb_reported_length(tvb);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
if (length == 8) {
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
if (c->tree) {
pi= proto_tree_add_item(c->tree, a->hf_value, tvb, 0, length, ENC_BIG_ENDIAN);
label = (char *)wmem_alloc(wmem_packet_scope(), ITEM_LABEL_LENGTH+1);
proto_item_fill_label(PITEM_FINFO(pi), label);
label = strstr(label,": ")+2;
}
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
else {
pi = proto_tree_add_bytes_format(c->tree, hf_diameter_avp_data_wrong_length,
Modify proto_tree_add_bytes_format() and proto_tree_add_bytes_format_value() so that if the start_ptr is NULL the bytes are extracted from the given TVB using the given offset and length. Replace a bunch of: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, tvb_get_ptr(tvb, offset, length), [...]) with: proto_tree_add_bytes_format*(tree, hf, tvb, offset, length, NULL, [...]) svn path=/trunk/; revision=35896
2011-02-10 16:31:00 +00:00
tvb, 0, length, NULL,
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
"Error! Bad Float64 Length");
expert_add_info_format_text -> expert_add_info_format svn path=/trunk/; revision=51852
2013-09-09 00:44:09 +00:00
expert_add_info_format(c->pinfo, pi, &ei_diameter_avp_len,
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
"Bad Float64 Length (%u)", length);
epan: Convert our PROTO_ITEM_ macros to inline functions. Convert our various PROTO_ITEM_ macros to inline functions and document them. Change-Id: I070b15d4f70d2189217a177ee8ba2740be36327c Reviewed-on: https://code.wireshark.org/review/32706 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-03 21:32:30 +00:00
proto_item_set_generated(pi);
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
}
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
Do length checking for Integer32, Integer64, Unsigned64, Float32, and Float64, as well as for Unsigned32. svn path=/trunk/; revision=35332
2011-01-03 03:30:20 +00:00
return label;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
grouped_avp(diam_ctx_t *c, diam_avp_t *a, tvbuff_t *tvb, diam_sub_dis_t *diam_sub_dis_inf)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
int offset = 0;
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
int len = tvb_reported_length(tvb);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_item *pi = proto_tree_add_item(c->tree, a->hf_value, tvb , 0 , -1, ENC_BIG_ENDIAN);
proto_tree *pt = c->tree;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
c->tree = proto_item_add_subtree(pi,a->ett);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
/* Set the flag that we are dissecting a grouped AVP */
diam_sub_dis_inf->dis_gouped = TRUE;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
while (offset < len) {
EAP, Diameter: add EAP-TTLS and EAP-Message dissection support Add support for dissecting the decrypted TLS payload as Diameter. Add support for dissecting the EAP-Message attribute as EAP. Disable retransmission detection when EAP-Message is detected (EAP in TLS in EAP) since this results in false positives. Tested with captures from Bug 15603: * eap-ttls-pap.pcapng - ok, User-Name and User-Password AVPs. * eap-ttls-eap-gtc.pcapng, eap-ttls-eap-md5.pcapng - EAP-Message AVP. * eap-ttls-mschapv2.pcapng - partially supported, does not conform to Diameter AVP requirements as it is not padded. Microsoft vendor types are also not yet supported. To be fixed later. * eapttls-diameter-avp.pcapng (Bug 12880) - EAP-Message AVP. Bug: 12880 Bug: 15603 Change-Id: Ie7ea282d05c1d3ff8463c34bf259107562714440 Reviewed-on: https://code.wireshark.org/review/34281 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 01:47:09 +00:00
offset += dissect_diameter_avp(c, tvb, offset, diam_sub_dis_inf, FALSE);
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
}
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
/* Clear info collected in grouped AVP */
Reject the packet if data is NULL. svn path=/trunk/; revision=53898
2013-12-10 03:25:04 +00:00
diam_sub_dis_inf->vendor_id = 0;
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
diam_sub_dis_inf->dis_gouped = FALSE;
diam_sub_dis_inf->avp_str = NULL;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
c->tree = pt;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
return NULL;
}
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static const char *msgflags_str[] = {
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
"----", "---T", "--E-", "--ET",
"-P--", "-P-T", "-PE-", "-PET",
"R---", "R--T", "R-E-", "R-ET",
"RP--", "RP-T", "RPE-", "RPET"
};
Fix the type of arrays of pointers to hf_ values for bitfield routines. The static arrays are supposed to be arrays of const pointers to int, not arrays of non-const pointers to const int. Fixing that means some bugs (scribbling on what's *supposed* to be a const array) will be caught (see packet-ieee80211-radiotap.c for examples, the first of which inspired this change and the second of which was discovered while testing compiles with this change), and removes the need for some annoying casts. Also make some of those arrays static while we're at it. Update documentation and dissector-generator tools. Change-Id: I789da5fc60aadc15797cefecfd9a9fbe9a130ccc Reviewed-on: https://code.wireshark.org/review/37517 Petri-Dish: Guy Harris <gharris@sonic.net> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-06-19 01:14:46 +00:00
static int * const diameter_flags_fields[] = {
[Diameter] Use proto_tree_add_bitmask() to add the diameter flags. Change-Id: I9cb6e46551b0a25a458311a53365469552d5c195 Reviewed-on: https://code.wireshark.org/review/7177 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-02-16 14:13:25 +00:00
&hf_diameter_flags_request,
&hf_diameter_flags_proxyable,
&hf_diameter_flags_error,
&hf_diameter_flags_T,
&hf_diameter_flags_reserved4,
&hf_diameter_flags_reserved5,
&hf_diameter_flags_reserved6,
&hf_diameter_flags_reserved7,
NULL
};
Add data parameter to tcp_dissect_pdus() as well as convert it to using "new" style dissectors. Now that "bytes consumed" can be determined, should tcp_dissect_pdus() take advantage of that? Should tcp_dissect_pdus return length (bytes consumed)? There are many dissectors that just call tcp_dissect_pdus() then return tvb_length(tvb). Seems like that could all be rolled into one. svn path=/trunk/; revision=53198
2013-11-09 17:46:28 +00:00
static int
dissect_diameter_common(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data _U_)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
guint32 version;
guint64 flags_bits;
int packet_len;
- add an all_cmds array for looking up 'RFC' packets - improve error/expert info when see unmatched AVPs, commands, appids svn path=/trunk/; revision=22363
2007-07-20 11:39:09 +00:00
proto_item *pi, *cmd_item, *app_item, *version_item;
Don't call val_to_str() only to then compare the result with "Uknown": use match_strval() instead. Also, don't retrieve the application ID 3 more times; we've already got it in a variable... Point out a bug regarding how we're storing the application ID for use by subdissectors (from r34570). svn path=/trunk/; revision=43162
2012-06-08 19:59:33 +00:00
proto_tree *diam_tree;
Replace instances of wmem_alloc with wmem_new This commit replaces instances of (myobj *)wmem_alloc(wmem_file_scope(), sizeof(myobj)) and replaces them with: wmem_new(wmem_file_scope(), myobj) to improve the readability of Wireshark's code. The replacement locations were identified with grep and replaced with the Python script below. grep command: egrep "wmem_alloc0?\(wmem_file_scope\(\), sizeof\([a-z_]+\)\)" . -R -l python script: import re import sys import fileinput pattern = r'\(([^\s]+) ?\*\) ?wmem_alloc(0?)\((wmem_[a-z]+_scope\(\)), sizeof\(\1\)\)' replacewith = r'wmem_new\2(\3, \1)' fname = sys.argv[1] for line in fileinput.input(fname, inplace=1, mode='rb'): output = re.sub(pattern, replacewith, line) sys.stdout.write(output) Change-Id: Ieac246c104bf01e32cbc6e11e53e81c7f639d870 Reviewed-on: https://code.wireshark.org/review/37158 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2020-05-07 12:35:45 +00:00
diam_ctx_t *c = wmem_new0(wmem_packet_scope(), diam_ctx_t);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
int offset;
Don't call val_to_str() only to then compare the result with "Uknown": use match_strval() instead. Also, don't retrieve the application ID 3 more times; we've already got it in a variable... Point out a bug regarding how we're storing the application ID for use by subdissectors (from r34570). svn path=/trunk/; revision=43162
2012-06-08 19:59:33 +00:00
const char *cmd_str;
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
guint32 cmd;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
guint32 hop_by_hop_id, end_to_end_id;
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
conversation_t *conversation;
diameter_conv_info_t *diameter_conv_info;
Initialize a variable to avoid a (false) may-be-used-uninitialized warning turned error. svn path=/trunk/; revision=46555
2012-12-16 02:37:06 +00:00
diameter_req_ans_pair_t *diameter_pair = NULL;
More wmem. svn path=/trunk/; revision=51617
2013-08-31 15:58:15 +00:00
wmem_tree_t *pdus_tree;
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
proto_item *it;
nstime_t ns;
Convert some dissectors to wmem (leaving uat-like memory with emem for now) svn path=/trunk/; revision=52031
2013-09-14 10:53:29 +00:00
diam_sub_dis_t *diam_sub_dis_inf = wmem_new0(wmem_packet_scope(), diam_sub_dis_t);
Dissect S6 feature list. svn path=/trunk/; revision=50796
2013-07-22 20:44:01 +00:00
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
/* Set default value Subscription-Id-Type and User-Equipment-Info-Type as XXX_UNKNOWN */
diam_sub_dis_inf->subscription_id_type = SUBSCRIPTION_ID_TYPE_UNKNOWN;
diam_sub_dis_inf->user_equipment_info_type = USER_EQUIPMENT_INFO_TYPE_UNKNOWN;
Delay registration of Diameter fields until they're needed. ... Like the RADIUS and wimaxasncp dissectors do. Change-Id: Ifab019a0040d3938e52918a282a3beba9dfcfc70 Reviewed-on: https://code.wireshark.org/review/14900 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-04-13 03:01:18 +00:00
/* Load header fields if not already done */
if (hf_diameter_code == -1)
Change how dissectors do late-field-registration to avoid a double-registration assertion. If a dissector forces registration of fields during dissection it needs to do so in a way that clears the prefix registration. Otherwise epan will call the registration routine a 2nd time (which will cause us to assert out) if a user types a display filter (with the dissector's prefix) that doesn't exist. Update the proto_register_prefix() comments to reflect this. Change-Id: I3ce29243395fb55192bb5dfd950baa88410ac136 Reviewed-on: https://code.wireshark.org/review/15881 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jaap Keuter <jaap.keuter@xs4all.nl>
2016-06-13 19:28:16 +00:00
proto_registrar_get_byname("diameter.code");
Dissect S6 feature list. svn path=/trunk/; revision=50796
2013-07-22 20:44:01 +00:00
Don't guard col_set_str (COL_PROTOCOL) with col_check svn path=/trunk/; revision=29340
2009-08-09 06:26:46 +00:00
col_set_str(pinfo->cinfo, COL_PROTOCOL, "DIAMETER");
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Export packets before dissecting them This way even malformed packets are properly exported Change-Id: I923825459eea725d0a103810f3883906b95b3b21 Reviewed-on: https://code.wireshark.org/review/15259 Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com> Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-05-03 16:36:10 +00:00
if (have_tap_listener(exported_pdu_tap)){
export_diameter_pdu(pinfo,tvb);
}
For proto_tree_add_item(..., proto_xxx, ...)use ENC_NA as the encoding arg. Also: remove trailing whitespace for a number of files. svn path=/trunk/; revision=39503
2011-10-21 02:10:19 +00:00
pi = proto_tree_add_item(tree,proto_diameter,tvb,0,-1,ENC_NA);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
diam_tree = proto_item_add_subtree(pi,ett_diameter);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
c->tree = diam_tree;
c->pinfo = pinfo;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
version_item = proto_tree_add_item_ret_uint(diam_tree, hf_diameter_version, tvb, 0, 1, ENC_BIG_ENDIAN, &version);
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
if (version != DIAMETER_RFC) {
expert_add_info(c->pinfo, version_item, &ei_diameter_version);
}
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
proto_tree_add_item_ret_uint(diam_tree, hf_diameter_length, tvb, 1, 3, ENC_BIG_ENDIAN, &packet_len);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
pi = proto_tree_add_bitmask_ret_uint64(diam_tree, tvb, 4, hf_diameter_flags, ett_diameter_flags, diameter_flags_fields, ENC_BIG_ENDIAN, &flags_bits);
if (flags_bits & 0x0f) {
expert_add_info(c->pinfo, pi, &ei_diameter_reserved_bit_set);
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
cmd_item = proto_tree_add_item_ret_uint(diam_tree, hf_diameter_code, tvb, 5, 3, ENC_BIG_ENDIAN, &cmd);
Diameter-3gpp: Dissect the SMS inside AVP 3301 SM-RP-UI. Change-Id: I0f293ea529dce5147eef5bfb9d8d4b39640fd0aa Reviewed-on: https://code.wireshark.org/review/30554 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-11-09 13:02:24 +00:00
diam_sub_dis_inf->cmd_code = cmd;
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
app_item = proto_tree_add_item_ret_uint(diam_tree, hf_diameter_application_id, tvb, 8, 4,
ENC_BIG_ENDIAN, &diam_sub_dis_inf->application_id);
Don't call val_to_str() only to then compare the result with "Uknown": use match_strval() instead. Also, don't retrieve the application ID 3 more times; we've already got it in a variable... Point out a bug regarding how we're storing the application ID for use by subdissectors (from r34570). svn path=/trunk/; revision=43162
2012-06-08 19:59:33 +00:00
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
if (try_val_to_str_ext(diam_sub_dis_inf->application_id, dictionary.applications) == NULL) {
proto_tree *tu = proto_item_add_subtree(app_item,ett_unknown);
proto_tree_add_expert_format(tu, c->pinfo, &ei_diameter_application_id, tvb, 8, 4,
"Unknown Application Id (%u), if you know what this is you can add it to dictionary.xml", diam_sub_dis_inf->application_id);
Use VendorID from sminmpec. svn path=/trunk/; revision=16552
2005-11-19 15:59:30 +00:00
}
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
Use val_to_str_const() where appropriate; Also (for a few files): - create/use some extended value strings; - remove unneeded #include files; - remove unneeded variable initialization; - re-order fcns slightly so prefs_reg_handoff...() at end, etc svn path=/trunk/; revision=44438
2012-08-10 22:55:02 +00:00
cmd_str = val_to_str_const(cmd, cmd_vs, "Unknown");
- add an all_cmds array for looking up 'RFC' packets - improve error/expert info when see unmatched AVPs, commands, appids svn path=/trunk/; revision=22363
2007-07-20 11:39:09 +00:00
if (strcmp(cmd_str, "Unknown") == 0) {
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
expert_add_info(c->pinfo, cmd_item, &ei_diameter_code);
- add an all_cmds array for looking up 'RFC' packets - improve error/expert info when see unmatched AVPs, commands, appids svn path=/trunk/; revision=22363
2007-07-20 11:39:09 +00:00
}
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
Move proto_reg_handoff... to the end of the file. Whitespace cleanup: remove trailing spaces from a few lines. svn path=/trunk/; revision=32569
2010-04-26 22:19:34 +00:00
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
proto_tree_add_item_ret_uint(diam_tree, hf_diameter_hopbyhopid, tvb, 12, 4, ENC_BIG_ENDIAN, &hop_by_hop_id);
proto_tree_add_item_ret_uint(diam_tree, hf_diameter_endtoendid, tvb, 16, 4, ENC_BIG_ENDIAN, &end_to_end_id);
col_add_fstr(pinfo->cinfo, COL_INFO,
"cmd=%s%s(%d) flags=%s %s=%s(%d) h2h=%x e2e=%x",
cmd_str,
((flags_bits>>4)&0x08) ? " Request" : " Answer",
cmd,
msgflags_str[((flags_bits>>4)&0x0f)],
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
"appl",
val_to_str_ext_const(diam_sub_dis_inf->application_id, dictionary.applications, "Unknown"),
[Diameter] Use proto_tree_add_item_ret..() and proto_tree_add_bitmask_with_flags() Change-Id: I864a385283aa6975a075f8621d871b73356ecf57 Reviewed-on: https://code.wireshark.org/review/25139 Petri-Dish: Anders Broman <a.broman58@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-01-04 13:14:43 +00:00
diam_sub_dis_inf->application_id,
hop_by_hop_id,
end_to_end_id);
col_append_str(pinfo->cinfo, COL_INFO, " | ");
col_set_fence(pinfo->cinfo, COL_INFO);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
/* Conversation tracking stuff */
Move proto_reg_handoff... to the end of the file. Whitespace cleanup: remove trailing spaces from a few lines. svn path=/trunk/; revision=32569
2010-04-26 22:19:34 +00:00
/*
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
* FIXME: Looking at epan/conversation.c it seems unlikely that this will work properly in
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
* multi-homed SCTP connections. This will probably need to be fixed at some point.
*/
As suggested in http://www.wireshark.org/lists/wireshark-dev/200809/msg00075.html (as referenced in https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=2907 ) and https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=3411 : Write a new convenience routine for finding a conversation and, if it is not found, create it. The frame number and addresses are taken from pinfo (as is the common case). Use this function in a bunch of dissectors. svn path=/trunk/; revision=32790
2010-05-13 18:28:34 +00:00
conversation = find_or_create_conversation(pinfo);
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
diameter_conv_info = (diameter_conv_info_t *)conversation_get_proto_data(conversation, proto_diameter);
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
if (!diameter_conv_info) {
More wmem. svn path=/trunk/; revision=51617
2013-08-31 15:58:15 +00:00
diameter_conv_info = wmem_new(wmem_file_scope(), diameter_conv_info_t);
Rename a variable for clarity. It's confusing to have 'pdus_tree' mean both the map of pdu_trees and the pdu trees themselves. Change-Id: Ie875798eb140b60a1309ddc0c0bf885b48c0407c Reviewed-on: https://code.wireshark.org/review/14413 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2016-03-11 02:03:52 +00:00
diameter_conv_info->pdu_trees = wmem_map_new(wmem_file_scope(), g_direct_hash, g_direct_equal);
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
conversation_add_proto_data(conversation, proto_diameter, diameter_conv_info);
}
More wmem. svn path=/trunk/; revision=51617
2013-08-31 15:58:15 +00:00
/* pdus_tree is an wmem_tree keyed by frame number (in order to handle hop-by-hop collisions */
Rename a variable for clarity. It's confusing to have 'pdus_tree' mean both the map of pdu_trees and the pdu trees themselves. Change-Id: Ie875798eb140b60a1309ddc0c0bf885b48c0407c Reviewed-on: https://code.wireshark.org/review/14413 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2016-03-11 02:03:52 +00:00
pdus_tree = (wmem_tree_t *)wmem_map_lookup(diameter_conv_info->pdu_trees, GUINT_TO_POINTER(hop_by_hop_id));
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
if (pdus_tree == NULL && (flags_bits & DIAM_FLAGS_R)) {
/* This is the first request we've seen with this hop-by-hop id */
More wmem. svn path=/trunk/; revision=51617
2013-08-31 15:58:15 +00:00
pdus_tree = wmem_tree_new(wmem_file_scope());
Rename a variable for clarity. It's confusing to have 'pdus_tree' mean both the map of pdu_trees and the pdu trees themselves. Change-Id: Ie875798eb140b60a1309ddc0c0bf885b48c0407c Reviewed-on: https://code.wireshark.org/review/14413 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2016-03-11 02:03:52 +00:00
wmem_map_insert(diameter_conv_info->pdu_trees, GUINT_TO_POINTER(hop_by_hop_id), pdus_tree);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
}
if (pdus_tree) {
Try to squeeze some bytes out of the frame_data structure. Make the time stamp precision a 4-bit bitfield, so, when combined with the other bitfields, we have 32 bits. That means we put the flags at the same structure level as the time stamp precision, so they can be combined; that gets rid of an extra "flags." for references to the flags. Put the two pointers next to each other, and after a multiple of 8 bytes worth of other fields, so that there's no padding before or between them. It's still not down to 64 bytes, which is the next lower power of 2, so there's more work to do. Change-Id: I6f3e9d9f6f48137bbee8f100c152d2c42adb8fbe Reviewed-on: https://code.wireshark.org/review/31213 Petri-Dish: Guy Harris <guy@alum.mit.edu> Tested-by: Petri Dish Buildbot Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-12-27 02:26:24 +00:00
if (!pinfo->fd->visited) {
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
if (flags_bits & DIAM_FLAGS_R) {
/* This is a request */
More wmem. svn path=/trunk/; revision=51617
2013-08-31 15:58:15 +00:00
diameter_pair = wmem_new(wmem_file_scope(), diameter_req_ans_pair_t);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
diameter_pair->hop_by_hop_id = hop_by_hop_id;
diameter_pair->end_to_end_id = end_to_end_id;
diameter_pair->cmd_code = cmd;
diameter_pair->result_code = 0;
diameter_pair->cmd_str = cmd_str;
Add the packet number to the packet_info structure, and use it. That removes most of the uses of the frame number field in the frame_data structure. Change-Id: Ie22e4533e87f8360d7c0a61ca6ffb796cc233f22 Reviewed-on: https://code.wireshark.org/review/13509 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-01-24 03:40:51 +00:00
diameter_pair->req_frame = pinfo->num;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
diameter_pair->ans_frame = 0;
Add more fields to packet_info structure and use them. Add fields for the absolute time stamp (and another field for a presence flag for the absolute time stamp) and the packet encapsulation for the packet. This lets us remove the field for the packet encapsulation in the frame_data structure; do so. Change-Id: Ifb910a9a192414e2a53086f3f7b97f39ed36aa39 Reviewed-on: https://code.wireshark.org/review/13499 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-01-23 03:50:21 +00:00
diameter_pair->req_time = pinfo->abs_ts;
Add the packet number to the packet_info structure, and use it. That removes most of the uses of the frame number field in the frame_data structure. Change-Id: Ie22e4533e87f8360d7c0a61ca6ffb796cc233f22 Reviewed-on: https://code.wireshark.org/review/13509 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-01-24 03:40:51 +00:00
wmem_tree_insert32(pdus_tree, pinfo->num, (void *)diameter_pair);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
} else {
/* Look for a request which occurs earlier in the trace than this answer. */
Add the packet number to the packet_info structure, and use it. That removes most of the uses of the frame number field in the frame_data structure. Change-Id: Ie22e4533e87f8360d7c0a61ca6ffb796cc233f22 Reviewed-on: https://code.wireshark.org/review/13509 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-01-24 03:40:51 +00:00
diameter_pair = (diameter_req_ans_pair_t *)wmem_tree_lookup32_le(pdus_tree, pinfo->num);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
/* Verify the end-to-end-id matches before declaring a match */
if (diameter_pair && diameter_pair->end_to_end_id == end_to_end_id) {
Add the packet number to the packet_info structure, and use it. That removes most of the uses of the frame number field in the frame_data structure. Change-Id: Ie22e4533e87f8360d7c0a61ca6ffb796cc233f22 Reviewed-on: https://code.wireshark.org/review/13509 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-01-24 03:40:51 +00:00
diameter_pair->ans_frame = pinfo->num;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
}
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
} else {
/* Look for a request which occurs earlier in the trace than this answer. */
Add the packet number to the packet_info structure, and use it. That removes most of the uses of the frame number field in the frame_data structure. Change-Id: Ie22e4533e87f8360d7c0a61ca6ffb796cc233f22 Reviewed-on: https://code.wireshark.org/review/13509 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-01-24 03:40:51 +00:00
diameter_pair = (diameter_req_ans_pair_t *)wmem_tree_lookup32_le(pdus_tree, pinfo->num);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
/* If the end-to-end ID doesn't match then this is not the request we were
* looking for.
*/
if (diameter_pair && diameter_pair->end_to_end_id != end_to_end_id)
diameter_pair = NULL;
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
}
}
if (!diameter_pair) {
/* create a "fake" diameter_pair structure */
Replace instances of wmem_alloc with wmem_new This commit replaces instances of (myobj *)wmem_alloc(wmem_file_scope(), sizeof(myobj)) and replaces them with: wmem_new(wmem_file_scope(), myobj) to improve the readability of Wireshark's code. The replacement locations were identified with grep and replaced with the Python script below. grep command: egrep "wmem_alloc0?\(wmem_file_scope\(\), sizeof\([a-z_]+\)\)" . -R -l python script: import re import sys import fileinput pattern = r'\(([^\s]+) ?\*\) ?wmem_alloc(0?)\((wmem_[a-z]+_scope\(\)), sizeof\(\1\)\)' replacewith = r'wmem_new\2(\3, \1)' fname = sys.argv[1] for line in fileinput.input(fname, inplace=1, mode='rb'): output = re.sub(pattern, replacewith, line) sys.stdout.write(output) Change-Id: Ieac246c104bf01e32cbc6e11e53e81c7f639d870 Reviewed-on: https://code.wireshark.org/review/37158 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2020-05-07 12:35:45 +00:00
diameter_pair = wmem_new(wmem_packet_scope(), diameter_req_ans_pair_t);
From Andrej Kuehnal: New feature: extract specified diameter AVPs from large capture files. https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=4560 svn path=/trunk/; revision=32132
2010-03-06 20:54:58 +00:00
diameter_pair->hop_by_hop_id = hop_by_hop_id;
diameter_pair->cmd_code = cmd;
diameter_pair->result_code = 0;
diameter_pair->cmd_str = cmd_str;
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
diameter_pair->req_frame = 0;
diameter_pair->ans_frame = 0;
Add more fields to packet_info structure and use them. Add fields for the absolute time stamp (and another field for a presence flag for the absolute time stamp) and the packet encapsulation for the packet. This lets us remove the field for the packet encapsulation in the frame_data structure; do so. Change-Id: Ifb910a9a192414e2a53086f3f7b97f39ed36aa39 Reviewed-on: https://code.wireshark.org/review/13499 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-01-23 03:50:21 +00:00
diameter_pair->req_time = pinfo->abs_ts;
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
}
When checking if a message is Diameter also check for a minimum message length. This improves the detection of Diameter messages within a TCP bytestream (i.e., when the Diameter PDUs don't neatly align with frames). Bug: 10362 Change-Id: I49a6e8cf076a6ab8a14761493aab9f3b11e4756e Reviewed-on: https://code.wireshark.org/review/3557 Petri-Dish: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com>
2014-08-12 14:09:03 +00:00
diameter_pair->processing_request=(flags_bits & DIAM_FLAGS_R)!= 0;
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
/* print state tracking info in the tree */
if (flags_bits & DIAM_FLAGS_R) {
/* This is a request */
if (diameter_pair->ans_frame) {
it = proto_tree_add_uint(diam_tree, hf_diameter_answer_in,
tvb, 0, 0, diameter_pair->ans_frame);
epan: Convert our PROTO_ITEM_ macros to inline functions. Convert our various PROTO_ITEM_ macros to inline functions and document them. Change-Id: I070b15d4f70d2189217a177ee8ba2740be36327c Reviewed-on: https://code.wireshark.org/review/32706 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-03 21:32:30 +00:00
proto_item_set_generated(it);
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
}
} else {
/* This is an answer */
if (diameter_pair->req_frame) {
it = proto_tree_add_uint(diam_tree, hf_diameter_answer_to,
tvb, 0, 0, diameter_pair->req_frame);
epan: Convert our PROTO_ITEM_ macros to inline functions. Convert our various PROTO_ITEM_ macros to inline functions and document them. Change-Id: I070b15d4f70d2189217a177ee8ba2740be36327c Reviewed-on: https://code.wireshark.org/review/32706 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-03 21:32:30 +00:00
proto_item_set_generated(it);
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
Add more fields to packet_info structure and use them. Add fields for the absolute time stamp (and another field for a presence flag for the absolute time stamp) and the packet encapsulation for the packet. This lets us remove the field for the packet encapsulation in the frame_data structure; do so. Change-Id: Ifb910a9a192414e2a53086f3f7b97f39ed36aa39 Reviewed-on: https://code.wireshark.org/review/13499 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-01-23 03:50:21 +00:00
nstime_delta(&ns, &pinfo->abs_ts, &diameter_pair->req_time);
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
diameter_pair->srt_time = ns;
it = proto_tree_add_time(diam_tree, hf_diameter_answer_time, tvb, 0, 0, &ns);
epan: Convert our PROTO_ITEM_ macros to inline functions. Convert our various PROTO_ITEM_ macros to inline functions and document them. Change-Id: I070b15d4f70d2189217a177ee8ba2740be36327c Reviewed-on: https://code.wireshark.org/review/32706 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-04-03 21:32:30 +00:00
proto_item_set_generated(it);
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
/* TODO: Populate result_code in tap record from AVP 268 */
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
}
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
}
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
offset = 20;
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
Generate Diameter's expert info's even when there's no tree so that they show up in the Expert Infos dialog. Push the if(tree) check down into the basic type dissectors since we can't generate/fill the label (which won't be used anyway) when we're not building the tree (since the proto_item will be faked/NULL). Change-Id: Ie4f1f6856cfad0dabc7c58cdee2c16c8fc032c6d Reviewed-on: https://code.wireshark.org/review/10001 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-08-12 17:57:20 +00:00
/* Dissect AVPs until the end of the packet is reached */
while (offset < packet_len) {
EAP, Diameter: add EAP-TTLS and EAP-Message dissection support Add support for dissecting the decrypted TLS payload as Diameter. Add support for dissecting the EAP-Message attribute as EAP. Disable retransmission detection when EAP-Message is detected (EAP in TLS in EAP) since this results in false positives. Tested with captures from Bug 15603: * eap-ttls-pap.pcapng - ok, User-Name and User-Password AVPs. * eap-ttls-eap-gtc.pcapng, eap-ttls-eap-md5.pcapng - EAP-Message AVP. * eap-ttls-mschapv2.pcapng - partially supported, does not conform to Diameter AVP requirements as it is not padded. Microsoft vendor types are also not yet supported. To be fixed later. * eapttls-diameter-avp.pcapng (Bug 12880) - EAP-Message AVP. Bug: 12880 Bug: 15603 Change-Id: Ie7ea282d05c1d3ff8463c34bf259107562714440 Reviewed-on: https://code.wireshark.org/review/34281 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 01:47:09 +00:00
offset += dissect_diameter_avp(c, tvb, offset, diam_sub_dis_inf, FALSE);
Fix the compilation. I guess this is what is meant. Only compile tested... svn path=/trunk/; revision=52928
2013-10-28 18:59:33 +00:00
}
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
/* Handle requests for which no answers were found and
Fix a lot of typos and misspellings Change-Id: I8512cfa1d424f82a873a0e0e1d22c7b075fdd7f3 Reviewed-on: https://code.wireshark.org/review/13069 Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-01-06 00:58:42 +00:00
* answers for which no requests were found in the tap listener.
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
* In case if you don't need unpaired requests/answers use:
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
* if (diameter_pair->processing_request || !diameter_pair->req_frame)
Pass diam_sub_dis_t directly to the subdissectors instead of using pinfo->private_data. svn path=/trunk/; revision=52919
2013-10-28 15:59:26 +00:00
* return;
*/
tap_queue_packet(diameter_tap, pinfo, diameter_pair);
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
return tvb_reported_length(tvb);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
static guint
Lua: Expose tcp_dissect_pdus() to Lua Provide a way for Lua-based dissectors to invoke tcp_dissect_pdus() to make TCP-based dissection easier. Bug: 9851 Change-Id: I91630ebf1f1fc1964118b6750cc34238e18a8ad3 Reviewed-on: https://code.wireshark.org/review/6778 Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Reviewed-by: Hadriel Kaplan <hadrielk@yahoo.com> Tested-by: Hadriel Kaplan <hadrielk@yahoo.com>
2015-01-25 19:30:13 +00:00
get_diameter_pdu_len(packet_info *pinfo _U_, tvbuff_t *tvb,
int offset, void *data _U_)
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
{
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
/* Get the length of the Diameter packet. */
return tvb_get_ntoh24(tvb, offset + 1);
}
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
#define NOT_DIAMETER 0
#define IS_DIAMETER 1
#define NOT_ENOUGH_DATA 2
static gint
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
check_diameter(tvbuff_t *tvb)
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
{
Tweak the Diameter heuristics a bit more. Increase the max Diameter message size to 65534 and reject messages whose flags have both the E- and R-bits set. Change-Id: Ib11701a47d23ff042a346d59c56f9f0f4410e6b7 Reviewed-on: https://code.wireshark.org/review/990 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-04-06 23:59:52 +00:00
guint8 flags;
diameter: Strengthen heuristic slightly The Diameter message length must be a multiple of 4, something implicitly true in RFC 3588 and make explicit in RFC 6733.
2022-07-04 01:25:57 +00:00
guint32 msg_len;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Strengthen the Diameter heuristics to avoid trying to reassemble enormous messages. Fixes some heuristics mistakes pointed out in http://ask.wireshark.org/questions/31227/diameter-reassembly-malformed-packets Change-Id: Iffc97d46cee5dd532fec0031286927fbcb86c095 Reviewed-on: https://code.wireshark.org/review/949 Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-04-03 17:57:15 +00:00
/* Ensure we don't throw an exception trying to do these heuristics */
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
if (tvb_captured_length(tvb) < 5)
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
return NOT_ENOUGH_DATA;
Strengthen the Diameter heuristics to avoid trying to reassemble enormous messages. Fixes some heuristics mistakes pointed out in http://ask.wireshark.org/questions/31227/diameter-reassembly-malformed-packets Change-Id: Iffc97d46cee5dd532fec0031286927fbcb86c095 Reviewed-on: https://code.wireshark.org/review/949 Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-04-03 17:57:15 +00:00
/* Check if the Diameter version is 1 */
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
if (tvb_get_guint8(tvb, 0) != 1)
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
return NOT_DIAMETER;
Strengthen the Diameter heuristics to avoid trying to reassemble enormous messages. Fixes some heuristics mistakes pointed out in http://ask.wireshark.org/questions/31227/diameter-reassembly-malformed-packets Change-Id: Iffc97d46cee5dd532fec0031286927fbcb86c095 Reviewed-on: https://code.wireshark.org/review/949 Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-04-03 17:57:15 +00:00
When checking if a message is Diameter also check for a minimum message length. This improves the detection of Diameter messages within a TCP bytestream (i.e., when the Diameter PDUs don't neatly align with frames). Bug: 10362 Change-Id: I49a6e8cf076a6ab8a14761493aab9f3b11e4756e Reviewed-on: https://code.wireshark.org/review/3557 Petri-Dish: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com>
2014-08-12 14:09:03 +00:00
/* Diameter minimum message length:
*
* Version+Length - 4 bytes
* Flags+CC - 4 bytes
* AppID - 4 bytes
* HbH - 4 bytes
* E2E - 4 bytes
* 2 AVPs (Orig-Host, Orig-Realm), each including:
* * AVP code - 4 bytes
* * AVP flags + length - 4 bytes
* * (no data - what would a reasonable minimum be?)
*
* --> 36 bytes
*/
diameter: Strengthen heuristic slightly The Diameter message length must be a multiple of 4, something implicitly true in RFC 3588 and make explicit in RFC 6733.
2022-07-04 01:25:57 +00:00
msg_len = tvb_get_ntoh24(tvb, 1);
/* Diameter message length field must be a multiple of 4.
* This is implicit in RFC 3588 (based on the header and that each
* AVP must align on a 32-bit boundary) and explicit in RFC 6733.
*/
if ((msg_len < 36) || (msg_len & 0x3))
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
return NOT_DIAMETER;
When checking if a message is Diameter also check for a minimum message length. This improves the detection of Diameter messages within a TCP bytestream (i.e., when the Diameter PDUs don't neatly align with frames). Bug: 10362 Change-Id: I49a6e8cf076a6ab8a14761493aab9f3b11e4756e Reviewed-on: https://code.wireshark.org/review/3557 Petri-Dish: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com>
2014-08-12 14:09:03 +00:00
Tweak the Diameter heuristics a bit more. Increase the max Diameter message size to 65534 and reject messages whose flags have both the E- and R-bits set. Change-Id: Ib11701a47d23ff042a346d59c56f9f0f4410e6b7 Reviewed-on: https://code.wireshark.org/review/990 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-04-06 23:59:52 +00:00
flags = tvb_get_guint8(tvb, 4);
Strengthen the Diameter heuristics to avoid trying to reassemble enormous messages. Fixes some heuristics mistakes pointed out in http://ask.wireshark.org/questions/31227/diameter-reassembly-malformed-packets Change-Id: Iffc97d46cee5dd532fec0031286927fbcb86c095 Reviewed-on: https://code.wireshark.org/review/949 Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-04-03 17:57:15 +00:00
/* Check if any of the Reserved flag bits are set */
Tweak the Diameter heuristics a bit more. Increase the max Diameter message size to 65534 and reject messages whose flags have both the E- and R-bits set. Change-Id: Ib11701a47d23ff042a346d59c56f9f0f4410e6b7 Reviewed-on: https://code.wireshark.org/review/990 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-04-06 23:59:52 +00:00
if (flags & 0x0f)
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
return NOT_DIAMETER;
Tweak the Diameter heuristics a bit more. Increase the max Diameter message size to 65534 and reject messages whose flags have both the E- and R-bits set. Change-Id: Ib11701a47d23ff042a346d59c56f9f0f4410e6b7 Reviewed-on: https://code.wireshark.org/review/990 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-04-06 23:59:52 +00:00
/* Check if both the R- and E-bits are set */
if ((flags & DIAM_FLAGS_R) && (flags & DIAM_FLAGS_E))
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
return NOT_DIAMETER;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
return IS_DIAMETER;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
Tweak the Diameter heuristics a bit more. Increase the max Diameter message size to 65534 and reject messages whose flags have both the E- and R-bits set. Change-Id: Ib11701a47d23ff042a346d59c56f9f0f4410e6b7 Reviewed-on: https://code.wireshark.org/review/990 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-04-06 23:59:52 +00:00
/*****************************************************************/
/* Main dissection function */
/* Checks if the message looks like Diameter before accepting it */
/*****************************************************************/
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
static int
Add data parameter to tcp_dissect_pdus() as well as convert it to using "new" style dissectors. Now that "bytes consumed" can be determined, should tcp_dissect_pdus() take advantage of that? Should tcp_dissect_pdus return length (bytes consumed)? There are many dissectors that just call tcp_dissect_pdus() then return tvb_length(tvb). Seems like that could all be rolled into one. svn path=/trunk/; revision=53198
2013-11-09 17:46:28 +00:00
dissect_diameter(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
{
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
if (check_diameter(tvb) != IS_DIAMETER)
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
return 0;
Add data parameter to tcp_dissect_pdus() as well as convert it to using "new" style dissectors. Now that "bytes consumed" can be determined, should tcp_dissect_pdus() take advantage of that? Should tcp_dissect_pdus return length (bytes consumed)? There are many dissectors that just call tcp_dissect_pdus() then return tvb_length(tvb). Seems like that could all be rolled into one. svn path=/trunk/; revision=53198
2013-11-09 17:46:28 +00:00
return dissect_diameter_common(tvb, pinfo, tree, data);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
Add data parameter to tcp_dissect_pdus() as well as convert it to using "new" style dissectors. Now that "bytes consumed" can be determined, should tcp_dissect_pdus() take advantage of that? Should tcp_dissect_pdus return length (bytes consumed)? There are many dissectors that just call tcp_dissect_pdus() then return tvb_length(tvb). Seems like that could all be rolled into one. svn path=/trunk/; revision=53198
2013-11-09 17:46:28 +00:00
static int
Tweak the Diameter heuristics a bit more. Increase the max Diameter message size to 65534 and reject messages whose flags have both the E- and R-bits set. Change-Id: Ib11701a47d23ff042a346d59c56f9f0f4410e6b7 Reviewed-on: https://code.wireshark.org/review/990 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-04-06 23:59:52 +00:00
dissect_diameter_tcp(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
{
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
gint is_diam = check_diameter(tvb);
if (is_diam == NOT_DIAMETER) {
/* We've probably been given a frame that's not the start of
* a PDU.
*/
We need to start reassembly on a segment boundary. svn path=/trunk/; revision=37740
2011-06-21 11:53:33 +00:00
col_set_str(pinfo->cinfo, COL_PROTOCOL, "DIAMETER");
col_set_str(pinfo->cinfo, COL_INFO, "Continuation");
call_dissector(data_handle, tvb, pinfo, tree);
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
} else if (is_diam == NOT_ENOUGH_DATA) {
/* Since we're doing our heuristic checks before
Fix more repeated words throughout Fixes more repeated words seen throughout the code base.
2021-12-31 03:03:45 +00:00
* tcp_dissect_pdus() (since we can't do heuristics once
Fix Diameter reassembly when the captured tvb length is less than 5 bytes. Since Diameter does heuristic checks before calling tcp_dissect_pdus() we have to "manually" ask for more data if the tvb is too short for our heuristics. Bug: 11183 Change-Id: I14c36042306b532b53df80cc3971866b76094084 Reviewed-on: https://code.wireshark.org/review/8405 Reviewed-by: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-05-11 19:45:04 +00:00
* we're in there) we sometimes have to ask for more data...
*/
pinfo->desegment_offset = 0;
pinfo->desegment_len = DESEGMENT_ONE_MORE_SEGMENT;
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
} else {
We need to start reassembly on a segment boundary. svn path=/trunk/; revision=37740
2011-06-21 11:53:33 +00:00
tcp_dissect_pdus(tvb, pinfo, tree, gbl_diameter_desegment, 4,
Add data parameter to tcp_dissect_pdus() as well as convert it to using "new" style dissectors. Now that "bytes consumed" can be determined, should tcp_dissect_pdus() take advantage of that? Should tcp_dissect_pdus return length (bytes consumed)? There are many dissectors that just call tcp_dissect_pdus() then return tvb_length(tvb). Seems like that could all be rolled into one. svn path=/trunk/; revision=53198
2013-11-09 17:46:28 +00:00
get_diameter_pdu_len, dissect_diameter_common, data);
We need to start reassembly on a segment boundary. svn path=/trunk/; revision=37740
2011-06-21 11:53:33 +00:00
}
Add data parameter to tcp_dissect_pdus() as well as convert it to using "new" style dissectors. Now that "bytes consumed" can be determined, should tcp_dissect_pdus() take advantage of that? Should tcp_dissect_pdus return length (bytes consumed)? There are many dissectors that just call tcp_dissect_pdus() then return tvb_length(tvb). Seems like that could all be rolled into one. svn path=/trunk/; revision=53198
2013-11-09 17:46:28 +00:00
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
return tvb_reported_length(tvb);
Use port range for SCTP too. svn path=/trunk/; revision=34557
2010-10-18 15:56:06 +00:00
}
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
Diameter: Add a heuristic dissector over TCP Add a heuristic dissector for Diameter over TCP, disabled by default. We shouldn't need one for Diameter over SCTP, the PPID should take care of it.
2022-07-03 16:36:31 +00:00
static gboolean
dissect_diameter_tcp_heur(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data)
{
if (check_diameter(tvb) != IS_DIAMETER) {
return FALSE;
}
conversation_set_dissector(find_or_create_conversation(pinfo), diameter_tcp_handle);
tcp_dissect_pdus(tvb, pinfo, tree, gbl_diameter_desegment, 4,
get_diameter_pdu_len, dissect_diameter_common, data);
return TRUE;
}
EAP, Diameter: add EAP-TTLS and EAP-Message dissection support Add support for dissecting the decrypted TLS payload as Diameter. Add support for dissecting the EAP-Message attribute as EAP. Disable retransmission detection when EAP-Message is detected (EAP in TLS in EAP) since this results in false positives. Tested with captures from Bug 15603: * eap-ttls-pap.pcapng - ok, User-Name and User-Password AVPs. * eap-ttls-eap-gtc.pcapng, eap-ttls-eap-md5.pcapng - EAP-Message AVP. * eap-ttls-mschapv2.pcapng - partially supported, does not conform to Diameter AVP requirements as it is not padded. Microsoft vendor types are also not yet supported. To be fixed later. * eapttls-diameter-avp.pcapng (Bug 12880) - EAP-Message AVP. Bug: 12880 Bug: 15603 Change-Id: Ie7ea282d05c1d3ff8463c34bf259107562714440 Reviewed-on: https://code.wireshark.org/review/34281 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 01:47:09 +00:00
static int
dissect_diameter_avps(tvbuff_t *tvb, packet_info *pinfo, proto_tree *tree, void *data _U_)
{
proto_item *pi;
proto_tree *diam_tree;
int offset = 0;
Replace instances of wmem_alloc with wmem_new This commit replaces instances of (myobj *)wmem_alloc(wmem_file_scope(), sizeof(myobj)) and replaces them with: wmem_new(wmem_file_scope(), myobj) to improve the readability of Wireshark's code. The replacement locations were identified with grep and replaced with the Python script below. grep command: egrep "wmem_alloc0?\(wmem_file_scope\(\), sizeof\([a-z_]+\)\)" . -R -l python script: import re import sys import fileinput pattern = r'\(([^\s]+) ?\*\) ?wmem_alloc(0?)\((wmem_[a-z]+_scope\(\)), sizeof\(\1\)\)' replacewith = r'wmem_new\2(\3, \1)' fname = sys.argv[1] for line in fileinput.input(fname, inplace=1, mode='rb'): output = re.sub(pattern, replacewith, line) sys.stdout.write(output) Change-Id: Ieac246c104bf01e32cbc6e11e53e81c7f639d870 Reviewed-on: https://code.wireshark.org/review/37158 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2020-05-07 12:35:45 +00:00
diam_ctx_t *c = wmem_new0(wmem_packet_scope(), diam_ctx_t);
EAP, Diameter: add EAP-TTLS and EAP-Message dissection support Add support for dissecting the decrypted TLS payload as Diameter. Add support for dissecting the EAP-Message attribute as EAP. Disable retransmission detection when EAP-Message is detected (EAP in TLS in EAP) since this results in false positives. Tested with captures from Bug 15603: * eap-ttls-pap.pcapng - ok, User-Name and User-Password AVPs. * eap-ttls-eap-gtc.pcapng, eap-ttls-eap-md5.pcapng - EAP-Message AVP. * eap-ttls-mschapv2.pcapng - partially supported, does not conform to Diameter AVP requirements as it is not padded. Microsoft vendor types are also not yet supported. To be fixed later. * eapttls-diameter-avp.pcapng (Bug 12880) - EAP-Message AVP. Bug: 12880 Bug: 15603 Change-Id: Ie7ea282d05c1d3ff8463c34bf259107562714440 Reviewed-on: https://code.wireshark.org/review/34281 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 01:47:09 +00:00
diam_sub_dis_t *diam_sub_dis_inf = wmem_new0(wmem_packet_scope(), diam_sub_dis_t);
/* Load header fields if not already done */
if (hf_diameter_code == -1)
proto_registrar_get_byname("diameter.code");
col_set_str(pinfo->cinfo, COL_PROTOCOL, "DIAMETER");
col_set_str(pinfo->cinfo, COL_INFO, "AVPs:");
pi = proto_tree_add_item(tree, proto_diameter, tvb, 0, -1, ENC_NA);
diam_tree = proto_item_add_subtree(pi, ett_diameter);
c->tree = diam_tree;
c->pinfo = pinfo;
/* Dissect AVPs until the end of the packet is reached */
while (tvb_reported_length_remaining(tvb, offset)) {
offset += dissect_diameter_avp(c, tvb, offset, diam_sub_dis_inf, TRUE);
}
return tvb_reported_length(tvb);
}
Vendor ID lookup fix, from David Frascone. svn path=/trunk/; revision=4494
2002-01-07 20:05:20 +00:00
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static char *
alnumerize(char *name)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
char *r = name;
char *w = name;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
char c;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
for (;(c = *r); r++) {
Get rid of calls to ctype.h functions. They don't handle values outside the range -1 to 127, and their behavior is locale-dependent. Use g_ascii_isXXX() and g_ascii_toXXX() instead of isXXX() and toXXX(). If you're checking for printable ASCII, don't use isascii() and don't use iscntrl(), use g_ascii_isprint(). If you're checking for graphical ASCII, i.e. printable ASCII except for a space, use g_ascii_isgraph(). Use ws_xton() to convert a hex digit character to the corresponding numeric value. Change-Id: Id3039bc586fbf66d8736c2df248c790c0d7a2330 Reviewed-on: https://code.wireshark.org/review/4851 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2014-10-20 06:14:50 +00:00
if (g_ascii_isalnum(c) || c == '_' || c == '-' || c == '.') {
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
*(w++) = c;
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
}
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
*w = '\0';
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
return name;
Make both the Diameter-over-TCP and Diameter-over-SCTP dissectors "new-style" dissectors, and have them reject packets that don't start with a version number of 1, in the hopes that it'll be less likely to try to dissect non-Diameter packets to/from port 1812 as Diameter and thus less likely to spew out a bunch of warnings that cause users to ask the Ethereal mailing lists what the problem is. Use "tcp_dissect_pdus()" to handle Diameter-over-TCP. svn path=/trunk/; revision=10165
2004-02-21 10:29:52 +00:00
}
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
static guint
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
reginfo(int *hf_ptr, const char *name, const char *abbr, const char *desc,
Make value_string_ext const-correct. There is still some const-incorrect usage of them but those can be ironed out after this change has been made. Change-Id: Iba0631c804bdab34d7c0232b49967130e3370488 Reviewed-on: https://code.wireshark.org/review/3199 Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-07-25 16:29:04 +00:00
enum ftenum ft, field_display_e base, value_string_ext *vs_ext,
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
guint32 mask)
{
Try to fix warning: initializer element is not computable at load time. svn path=/trunk/; revision=54832
2014-01-17 18:58:50 +00:00
hf_register_info hf;
When checking if a message is Diameter also check for a minimum message length. This improves the detection of Diameter messages within a TCP bytestream (i.e., when the Diameter PDUs don't neatly align with frames). Bug: 10362 Change-Id: I49a6e8cf076a6ab8a14761493aab9f3b11e4756e Reviewed-on: https://code.wireshark.org/review/3557 Petri-Dish: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com>
2014-08-12 14:09:03 +00:00
hf.p_id = hf_ptr;
Try to fix warning: initializer element is not computable at load time. svn path=/trunk/; revision=54832
2014-01-17 18:58:50 +00:00
hf.hfinfo.name = name;
hf.hfinfo.abbrev = abbr;
hf.hfinfo.type = ft;
hf.hfinfo.display = base;
hf.hfinfo.strings = NULL;
hf.hfinfo.bitmask = mask;
hf.hfinfo.blurb = desc;
/* HFILL */
As spotted by Joerg; make a HFILL_INIT(hf) macro and use it to init the private part of the hf structure. svn path=/trunk/; revision=54843
2014-01-19 09:40:04 +00:00
HFILL_INIT(hf);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
if (vs_ext) {
Use value_string ext for address family. svn path=/trunk/; revision=34887
2010-11-15 21:05:03 +00:00
hf.hfinfo.strings = vs_ext;
}
Build diameter's hf array in a wmem array instead of a garray. Fixes ~270KB of leak when running epan_cleanup(). svn path=/trunk/; revision=50405
2013-07-06 12:45:33 +00:00
wmem_array_append_one(build_dict.hf,hf);
return wmem_array_get_count(build_dict.hf);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
static void
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
basic_avp_reginfo(diam_avp_t *a, const char *name, enum ftenum ft,
Make value_string_ext const-correct. There is still some const-incorrect usage of them but those can be ironed out after this change has been made. Change-Id: Iba0631c804bdab34d7c0232b49967130e3370488 Reviewed-on: https://code.wireshark.org/review/3199 Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-07-25 16:29:04 +00:00
field_display_e base, value_string_ext *vs_ext)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Try to fix warning: initializer element is not computable at load time. svn path=/trunk/; revision=54832
2014-01-17 18:58:50 +00:00
hf_register_info hf;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
gint *ettp = &(a->ett);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
When checking if a message is Diameter also check for a minimum message length. This improves the detection of Diameter messages within a TCP bytestream (i.e., when the Diameter PDUs don't neatly align with frames). Bug: 10362 Change-Id: I49a6e8cf076a6ab8a14761493aab9f3b11e4756e Reviewed-on: https://code.wireshark.org/review/3557 Petri-Dish: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com>
2014-08-12 14:09:03 +00:00
hf.p_id = &(a->hf_value);
Try to fix warning: initializer element is not computable at load time. svn path=/trunk/; revision=54832
2014-01-17 18:58:50 +00:00
hf.hfinfo.name = NULL;
hf.hfinfo.abbrev = NULL;
hf.hfinfo.type = ft;
hf.hfinfo.display = base;
hf.hfinfo.strings = NULL;
hf.hfinfo.bitmask = 0x0;
hf.hfinfo.blurb = a->vendor->code ?
When checking if a message is Diameter also check for a minimum message length. This improves the detection of Diameter messages within a TCP bytestream (i.e., when the Diameter PDUs don't neatly align with frames). Bug: 10362 Change-Id: I49a6e8cf076a6ab8a14761493aab9f3b11e4756e Reviewed-on: https://code.wireshark.org/review/3557 Petri-Dish: Evan Huus <eapache@gmail.com> Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com>
2014-08-12 14:09:03 +00:00
wmem_strdup_printf(wmem_epan_scope(), "vendor=%d code=%d", a->vendor->code, a->code)
: wmem_strdup_printf(wmem_epan_scope(), "code=%d", a->code);
Try to fix warning: initializer element is not computable at load time. svn path=/trunk/; revision=54832
2014-01-17 18:58:50 +00:00
/* HFILL */
As spotted by Joerg; make a HFILL_INIT(hf) macro and use it to init the private part of the hf structure. svn path=/trunk/; revision=54843
2014-01-19 09:40:04 +00:00
HFILL_INIT(hf);
Try to fix warning: initializer element is not computable at load time. svn path=/trunk/; revision=54832
2014-01-17 18:58:50 +00:00
Use faster wmem_str* functions in a few places. Use wmem_strdup and wmem_strconcat instead of wmem_strdup_printf. This shaves a small amount of time off of register_all_protocols on Windows according to the Visual Studio profiler. Change-Id: Ib6991e8de5b4fc30e960c513a3028c09dfe6a0a4 Reviewed-on: https://code.wireshark.org/review/14770 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-04-01 19:48:22 +00:00
hf.hfinfo.name = wmem_strdup(wmem_epan_scope(), name);
hf.hfinfo.abbrev = alnumerize(wmem_strconcat(wmem_epan_scope(), "diameter.", name, NULL));
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
if (vs_ext) {
Try to fix warning: initializer element is not computable at load time. svn path=/trunk/; revision=54832
2014-01-17 18:58:50 +00:00
hf.hfinfo.strings = vs_ext;
Make value_string_ext for all AVP:s with enums. svn path=/trunk/; revision=34835
2010-11-10 20:11:38 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Try to fix warning: initializer element is not computable at load time. svn path=/trunk/; revision=54832
2014-01-17 18:58:50 +00:00
wmem_array_append(build_dict.hf,&hf,1);
The second argument to g_ptr_array_add() is the pointer to add, not a pointer to that pointer - unlike g_array_append_vals, which takes a pointer to an array of values and a count of the number of values in the array. svn path=/trunk/; revision=25753
2008-07-16 08:08:17 +00:00
g_ptr_array_add(build_dict.ett,ettp);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
Removed trailing whitespaces from .h and .c files using the winapi_cleanup tool written by Patrik Stridvall for the wine project. svn path=/trunk/; revision=6117
2002-08-28 21:04:11 +00:00
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static diam_avp_t *
build_address_avp(const avp_type_t *type _U_, guint32 code,
Don't make promises of constness that we can't honor. svn path=/trunk/; revision=54757
2014-01-14 03:24:37 +00:00
diam_vnd_t *vendor, const char *name,
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
const value_string *vs _U_, void *data _U_)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Use epan-scoped memory for building more of the diameter table stuff. svn path=/trunk/; revision=51754
2013-09-04 21:06:56 +00:00
diam_avp_t *a = wmem_new0(wmem_epan_scope(), diam_avp_t);
address_avp_t *t = wmem_new(wmem_epan_scope(), address_avp_t);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
gint *ettp = &(t->ett);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
a->code = code;
a->vendor = vendor;
Change dissecton of address type for AVP < 256 (RADIUS address). svn path=/trunk/; revision=23402
2007-11-08 22:36:19 +00:00
/*
Include padding length in the AVP's length. svn path=/trunk/; revision=48702
2013-04-02 18:15:42 +00:00
* It seems like the radius AVPs 1-255 will use the defs from RADIUS in which case:
Standardise IETF RFC and Draft URLs in dissectors. Prefer: - html (rather than txt) - https Also includes the script check_dissector_urls.py, that can be used to find links in code and test them. Change-Id: Iafd8bb8948674a38ad5232bf5b5432ffb2b1251b Reviewed-on: https://code.wireshark.org/review/36821 Petri-Dish: Martin Mathieson <martin.r.mathieson@googlemail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Peter Wu <peter@lekensteyn.nl>
2020-04-13 11:06:37 +00:00
* https://tools.ietf.org/html/rfc2685
Include padding length in the AVP's length. svn path=/trunk/; revision=48702
2013-04-02 18:15:42 +00:00
* Address
* The Address field is four octets. The value 0xFFFFFFFF indicates
* that the NAS Should allow the user to select an address (e.g.
* Negotiated). The value 0xFFFFFFFE indicates that the NAS should
* select an address for the user (e.g. Assigned from a pool of
* addresses kept by the NAS). Other valid values indicate that the
* NAS should use that value as the user's IP address.
*
* Where as in Diameter:
* RFC3588
* Address
* The Address format is derived from the OctetString AVP Base
* Format. It is a discriminated union, representing, for example a
* 32-bit (IPv4) [IPV4] or 128-bit (IPv6) [IPV6] address, most
* significant octet first. The first two octets of the Address
* AVP represents the AddressType, which contains an Address Family
* defined in [IANAADFAM]. The AddressType is used to discriminate
* the content and format of the remaining octets.
*/
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
if (code<256) {
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
a->dissector_rfc = address_radius_avp;
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
} else {
Change dissecton of address type for AVP < 256 (RADIUS address). svn path=/trunk/; revision=23402
2007-11-08 22:36:19 +00:00
a->dissector_rfc = address_rfc_avp;
}
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
a->ett = -1;
a->hf_value = -1;
a->type_data = t;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
t->ett = -1;
t->hf_address_type = -1;
t->hf_ipv4 = -1;
t->hf_ipv6 = -1;
Diameter: Handle (IP)address type E164 as string The Diameter type Address hase a two byte address type family field previously only IPv4 and IPv6 was handled. Add handling of E.164 when encoded as a string.
2020-10-12 08:34:49 +00:00
t->hf_e164_str = -1;
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
t->hf_other = -1;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
basic_avp_reginfo(a, name, FT_BYTES, BASE_NONE, NULL);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Use faster string functions at startup. Use wmem_strconcat and g_strconcat instead of wmem_strdup_printf and g_strdup_printf when we register various protocols. This shows a fairly significant speedup in the Visual Studio profiler. Change-Id: I98709329513daa66ad3665925dc69149c43df884 Reviewed-on: https://code.wireshark.org/review/14855 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-04-07 20:05:27 +00:00
reginfo(&(t->hf_address_type), wmem_strconcat(wmem_epan_scope(), name, " Address Family", NULL),
alnumerize(wmem_strconcat(wmem_epan_scope(), "diameter.", name, ".addr_family", NULL)),
Rename base_display_* to field_display This is preparation to support ->display for FT_STRING values. svn path=/trunk/; revision=53820
2013-12-07 10:41:42 +00:00
NULL, FT_UINT16, (field_display_e)(BASE_DEC|BASE_EXT_STRING), &diameter_avp_data_addrfamily_vals_ext, 0);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Use faster string functions at startup. Use wmem_strconcat and g_strconcat instead of wmem_strdup_printf and g_strdup_printf when we register various protocols. This shows a fairly significant speedup in the Visual Studio profiler. Change-Id: I98709329513daa66ad3665925dc69149c43df884 Reviewed-on: https://code.wireshark.org/review/14855 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-04-07 20:05:27 +00:00
reginfo(&(t->hf_ipv4), wmem_strconcat(wmem_epan_scope(), name, " Address", NULL),
alnumerize(wmem_strconcat(wmem_epan_scope(), "diameter.", name, ".IPv4", NULL)),
From Kovarththanan Rajaratnam: More hf_register_info related cleanup. svn path=/trunk/; revision=28832
2009-06-24 02:17:12 +00:00
NULL, FT_IPv4, BASE_NONE, NULL, 0);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Use faster string functions at startup. Use wmem_strconcat and g_strconcat instead of wmem_strdup_printf and g_strdup_printf when we register various protocols. This shows a fairly significant speedup in the Visual Studio profiler. Change-Id: I98709329513daa66ad3665925dc69149c43df884 Reviewed-on: https://code.wireshark.org/review/14855 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-04-07 20:05:27 +00:00
reginfo(&(t->hf_ipv6), wmem_strconcat(wmem_epan_scope(), name, " Address", NULL),
alnumerize(wmem_strconcat(wmem_epan_scope(), "diameter.", name, ".IPv6", NULL)),
From Kovarththanan Rajaratnam: More hf_register_info related cleanup. svn path=/trunk/; revision=28832
2009-06-24 02:17:12 +00:00
NULL, FT_IPv6, BASE_NONE, NULL, 0);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Diameter: Handle (IP)address type E164 as string The Diameter type Address hase a two byte address type family field previously only IPv4 and IPv6 was handled. Add handling of E.164 when encoded as a string.
2020-10-12 08:34:49 +00:00
reginfo(&(t->hf_e164_str), wmem_strconcat(wmem_epan_scope(), name, " Address", NULL),
alnumerize(wmem_strconcat(wmem_epan_scope(), "diameter.", name, ".E164", NULL)),
NULL, FT_STRING, BASE_NONE, NULL, 0);
Use faster string functions at startup. Use wmem_strconcat and g_strconcat instead of wmem_strdup_printf and g_strdup_printf when we register various protocols. This shows a fairly significant speedup in the Visual Studio profiler. Change-Id: I98709329513daa66ad3665925dc69149c43df884 Reviewed-on: https://code.wireshark.org/review/14855 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-04-07 20:05:27 +00:00
reginfo(&(t->hf_other), wmem_strconcat(wmem_epan_scope(), name, " Address", NULL),
alnumerize(wmem_strconcat(wmem_epan_scope(), "diameter.", name, ".Bytes", NULL)),
From Kovarththanan Rajaratnam: More hf_register_info related cleanup. svn path=/trunk/; revision=28832
2009-06-24 02:17:12 +00:00
NULL, FT_BYTES, BASE_NONE, NULL, 0);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
The second argument to g_ptr_array_add() is the pointer to add, not a pointer to that pointer - unlike g_array_append_vals, which takes a pointer to an array of values and a count of the number of values in the array. svn path=/trunk/; revision=25753
2008-07-16 08:08:17 +00:00
g_ptr_array_add(build_dict.ett,ettp);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
return a;
}
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static diam_avp_t *
build_proto_avp(const avp_type_t *type _U_, guint32 code,
Don't make promises of constness that we can't honor. svn path=/trunk/; revision=54757
2014-01-14 03:24:37 +00:00
diam_vnd_t *vendor, const char *name _U_,
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
const value_string *vs _U_, void *data)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Replace instances of wmem_alloc with wmem_new This commit replaces instances of (myobj *)wmem_alloc(wmem_file_scope(), sizeof(myobj)) and replaces them with: wmem_new(wmem_file_scope(), myobj) to improve the readability of Wireshark's code. The replacement locations were identified with grep and replaced with the Python script below. grep command: egrep "wmem_alloc0?\(wmem_file_scope\(\), sizeof\([a-z_]+\)\)" . -R -l python script: import re import sys import fileinput pattern = r'\(([^\s]+) ?\*\) ?wmem_alloc(0?)\((wmem_[a-z]+_scope\(\)), sizeof\(\1\)\)' replacewith = r'wmem_new\2(\3, \1)' fname = sys.argv[1] for line in fileinput.input(fname, inplace=1, mode='rb'): output = re.sub(pattern, replacewith, line) sys.stdout.write(output) Change-Id: Ieac246c104bf01e32cbc6e11e53e81c7f639d870 Reviewed-on: https://code.wireshark.org/review/37158 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2020-05-07 12:35:45 +00:00
diam_avp_t *a = wmem_new0(wmem_epan_scope(), diam_avp_t);
proto_avp_t *t = wmem_new0(wmem_epan_scope(), proto_avp_t);
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
gint *ettp = &(a->ett);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
a->code = code;
a->vendor = vendor;
a->dissector_rfc = proto_avp;
a->ett = -1;
a->hf_value = -2;
a->type_data = t;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
t->name = (char *)data;
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
t->handle = NULL;
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
t->reassemble_mode = REASEMBLE_NEVER;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
The second argument to g_ptr_array_add() is the pointer to add, not a pointer to that pointer - unlike g_array_append_vals, which takes a pointer to an array of values and a count of the number of values in the array. svn path=/trunk/; revision=25753
2008-07-16 08:08:17 +00:00
g_ptr_array_add(build_dict.ett,ettp);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
return a;
}
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
static diam_avp_t *
Don't make promises of constness that we can't honor. svn path=/trunk/; revision=54757
2014-01-14 03:24:37 +00:00
build_simple_avp(const avp_type_t *type, guint32 code, diam_vnd_t *vendor,
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
const char *name, const value_string *vs, void *data _U_)
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
{
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
diam_avp_t *a;
Make value_string_ext const-correct. There is still some const-incorrect usage of them but those can be ironed out after this change has been made. Change-Id: Iba0631c804bdab34d7c0232b49967130e3370488 Reviewed-on: https://code.wireshark.org/review/3199 Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-07-25 16:29:04 +00:00
value_string_ext *vs_ext = NULL;
Rename base_display_* to field_display This is preparation to support ->display for FT_STRING values. svn path=/trunk/; revision=53820
2013-12-07 10:41:42 +00:00
field_display_e base;
Make value_string_ext for all AVP:s with enums. svn path=/trunk/; revision=34835
2010-11-10 20:11:38 +00:00
guint i = 0;
Check for AVPs with a list of values and a type that's not a 32-bit or shorter integral type. Fixes bug 2027. Rename the "bytes" pointer to "octetstring", and initialize it in a fashion that makes it clearer that it points to the first of the basic types, to make it clearer that it's for OctetString. svn path=/trunk/; revision=23615
2007-11-27 01:41:42 +00:00
/*
* Only 32-bit or shorter integral types can have a list of values.
*/
Rename base_display_* to field_display This is preparation to support ->display for FT_STRING values. svn path=/trunk/; revision=53820
2013-12-07 10:41:42 +00:00
base = (field_display_e)type->base;
Check for AVPs with a list of values and a type that's not a 32-bit or shorter integral type. Fixes bug 2027. Rename the "bytes" pointer to "octetstring", and initialize it in a fashion that makes it clearer that it points to the first of the basic types, to make it clearer that it's for OctetString. svn path=/trunk/; revision=23615
2007-11-27 01:41:42 +00:00
if (vs != NULL) {
switch (type->ft) {
case FT_UINT8:
case FT_UINT16:
case FT_UINT32:
case FT_INT8:
case FT_INT16:
case FT_INT32:
break;
default:
Diameter: Expand an error message. If we encounter the wrong ftype, print its name. Change-Id: I7405ccdd3e099f533c6a8aaf81b60faf4093741a Reviewed-on: https://code.wireshark.org/review/11790 Petri-Dish: Gerald Combs <gerald@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-11-13 01:39:39 +00:00
report_failure("Diameter Dictionary: AVP '%s' has a list of values but isn't of a 32-bit or shorter integral type (%s)\n",
name, ftype_name(type->ft));
Check for AVPs with a list of values and a type that's not a 32-bit or shorter integral type. Fixes bug 2027. Rename the "bytes" pointer to "octetstring", and initialize it in a fashion that makes it clearer that it points to the first of the basic types, to make it clearer that it's for OctetString. svn path=/trunk/; revision=23615
2007-11-27 01:41:42 +00:00
return NULL;
}
Make value_string_ext for all AVP:s with enums. svn path=/trunk/; revision=34835
2010-11-10 20:11:38 +00:00
while (vs[i].strptr) {
i++;
}
Use faster string functions at startup. Use wmem_strconcat and g_strconcat instead of wmem_strdup_printf and g_strdup_printf when we register various protocols. This shows a fairly significant speedup in the Visual Studio profiler. Change-Id: I98709329513daa66ad3665925dc69149c43df884 Reviewed-on: https://code.wireshark.org/review/14855 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-04-07 20:05:27 +00:00
vs_ext = value_string_ext_new(vs, i+1, wmem_strconcat(wmem_epan_scope(), name, "_vals_ext", NULL));
Rename base_display_* to field_display This is preparation to support ->display for FT_STRING values. svn path=/trunk/; revision=53820
2013-12-07 10:41:42 +00:00
base = (field_display_e)(base|BASE_EXT_STRING);
Check for AVPs with a list of values and a type that's not a 32-bit or shorter integral type. Fixes bug 2027. Rename the "bytes" pointer to "octetstring", and initialize it in a fashion that makes it clearer that it points to the first of the basic types, to make it clearer that it's for OctetString. svn path=/trunk/; revision=23615
2007-11-27 01:41:42 +00:00
}
Replace instances of wmem_alloc with wmem_new This commit replaces instances of (myobj *)wmem_alloc(wmem_file_scope(), sizeof(myobj)) and replaces them with: wmem_new(wmem_file_scope(), myobj) to improve the readability of Wireshark's code. The replacement locations were identified with grep and replaced with the Python script below. grep command: egrep "wmem_alloc0?\(wmem_file_scope\(\), sizeof\([a-z_]+\)\)" . -R -l python script: import re import sys import fileinput pattern = r'\(([^\s]+) ?\*\) ?wmem_alloc(0?)\((wmem_[a-z]+_scope\(\)), sizeof\(\1\)\)' replacewith = r'wmem_new\2(\3, \1)' fname = sys.argv[1] for line in fileinput.input(fname, inplace=1, mode='rb'): output = re.sub(pattern, replacewith, line) sys.stdout.write(output) Change-Id: Ieac246c104bf01e32cbc6e11e53e81c7f639d870 Reviewed-on: https://code.wireshark.org/review/37158 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2020-05-07 12:35:45 +00:00
a = wmem_new0(wmem_epan_scope(), diam_avp_t);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
a->code = code;
a->vendor = vendor;
a->dissector_rfc = type->rfc;
a->ett = -1;
a->hf_value = -1;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
basic_avp_reginfo(a, name, type->ft, base, vs_ext);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
return a;
}
Removed trailing whitespaces from .h and .c files using the winapi_cleanup tool written by Patrik Stridvall for the wine project. svn path=/trunk/; revision=6117
2002-08-28 21:04:11 +00:00
Add special handling of the (derived) AppId AVP type so that we can maintain one (and only one) Application ID list. This means we don't have to list all the Application IDs several times in the XML file. Collapse those lists into one (much more complete, now) list while also fixing URIs for several of the specifications (and making it clear for others that we don't have access to the specification--by declaring their URI to be "none"). Add a bunch more entries to the list of application IDs. Change-Id: Ia5c96b1f6f1fe3a9521b3d70142889e3881fae5e Reviewed-on: https://code.wireshark.org/review/8147 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-04-20 22:14:35 +00:00
static diam_avp_t *
build_appid_avp(const avp_type_t *type, guint32 code, diam_vnd_t *vendor,
diameter: fix 'pinfo/vs' was marked unused but was used [-Werror,-Wused-but-marked-unused] Change-Id: Ib7f1682a3cdeb8571a410db07261ff68bac03432 Reviewed-on: https://code.wireshark.org/review/14977 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-04-15 17:39:06 +00:00
const char *name, const value_string *vs, void *data _U_)
Add special handling of the (derived) AppId AVP type so that we can maintain one (and only one) Application ID list. This means we don't have to list all the Application IDs several times in the XML file. Collapse those lists into one (much more complete, now) list while also fixing URIs for several of the specifications (and making it clear for others that we don't have access to the specification--by declaring their URI to be "none"). Add a bunch more entries to the list of application IDs. Change-Id: Ia5c96b1f6f1fe3a9521b3d70142889e3881fae5e Reviewed-on: https://code.wireshark.org/review/8147 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-04-20 22:14:35 +00:00
{
diam_avp_t *a;
field_display_e base;
Support for desegmentation of Diameter-over-TCP, from David Frascone. svn path=/trunk/; revision=4141
2001-11-03 12:19:56 +00:00
Replace instances of wmem_alloc with wmem_new This commit replaces instances of (myobj *)wmem_alloc(wmem_file_scope(), sizeof(myobj)) and replaces them with: wmem_new(wmem_file_scope(), myobj) to improve the readability of Wireshark's code. The replacement locations were identified with grep and replaced with the Python script below. grep command: egrep "wmem_alloc0?\(wmem_file_scope\(\), sizeof\([a-z_]+\)\)" . -R -l python script: import re import sys import fileinput pattern = r'\(([^\s]+) ?\*\) ?wmem_alloc(0?)\((wmem_[a-z]+_scope\(\)), sizeof\(\1\)\)' replacewith = r'wmem_new\2(\3, \1)' fname = sys.argv[1] for line in fileinput.input(fname, inplace=1, mode='rb'): output = re.sub(pattern, replacewith, line) sys.stdout.write(output) Change-Id: Ieac246c104bf01e32cbc6e11e53e81c7f639d870 Reviewed-on: https://code.wireshark.org/review/37158 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2020-05-07 12:35:45 +00:00
a = wmem_new0(wmem_epan_scope(), diam_avp_t);
Add special handling of the (derived) AppId AVP type so that we can maintain one (and only one) Application ID list. This means we don't have to list all the Application IDs several times in the XML file. Collapse those lists into one (much more complete, now) list while also fixing URIs for several of the specifications (and making it clear for others that we don't have access to the specification--by declaring their URI to be "none"). Add a bunch more entries to the list of application IDs. Change-Id: Ia5c96b1f6f1fe3a9521b3d70142889e3881fae5e Reviewed-on: https://code.wireshark.org/review/8147 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-04-20 22:14:35 +00:00
a->code = code;
a->vendor = vendor;
a->dissector_rfc = type->rfc;
a->ett = -1;
a->hf_value = -1;
if (vs != NULL) {
report_failure("Diameter Dictionary: AVP '%s' (of type AppId) has a list of values but the list won't be used\n",
name);
}
base = (field_display_e)(type->base|BASE_EXT_STRING);
basic_avp_reginfo(a, name, type->ft, base, dictionary.applications);
return a;
}
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
static const avp_type_t basic_types[] = {
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
{"octetstring" , simple_avp , FT_BYTES , BASE_NONE , build_simple_avp },
{"octetstringorutf8" , simple_avp , FT_BYTES , BASE_SHOW_ASCII_PRINTABLE , build_simple_avp },
{"utf8string" , utf8_avp , FT_STRING , BASE_NONE , build_simple_avp },
{"grouped" , grouped_avp , FT_BYTES , BASE_NONE , build_simple_avp },
{"integer32" , integer32_avp , FT_INT32 , BASE_DEC , build_simple_avp },
{"unsigned32" , unsigned32_avp , FT_UINT32 , BASE_DEC , build_simple_avp },
{"integer64" , integer64_avp , FT_INT64 , BASE_DEC , build_simple_avp },
{"unsigned64" , unsigned64_avp , FT_UINT64 , BASE_DEC , build_simple_avp },
{"float32" , float32_avp , FT_FLOAT , BASE_NONE , build_simple_avp },
{"float64" , float64_avp , FT_DOUBLE , BASE_NONE , build_simple_avp },
{"ipaddress" , NULL , FT_NONE , BASE_NONE , build_address_avp },
{"diameteruri" , utf8_avp , FT_STRING , BASE_NONE , build_simple_avp },
{"diameteridentity" , utf8_avp , FT_STRING , BASE_NONE , build_simple_avp },
{"ipfilterrule" , utf8_avp , FT_STRING , BASE_NONE , build_simple_avp },
{"qosfilterrule" , utf8_avp , FT_STRING , BASE_NONE , build_simple_avp },
{"time" , time_avp , FT_ABSOLUTE_TIME , ABSOLUTE_TIME_UTC , build_simple_avp },
{"AppId" , simple_avp , FT_UINT32 , BASE_DEC , build_appid_avp },
{NULL, NULL, FT_NONE, BASE_NONE, NULL }
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
};
Make both the Diameter-over-TCP and Diameter-over-SCTP dissectors "new-style" dissectors, and have them reject packets that don't start with a version number of 1, in the hopes that it'll be less likely to try to dissect non-Diameter packets to/from port 1812 as Diameter and thus less likely to spew out a bunch of warnings that cause users to ask the Ethereal mailing lists what the problem is. Use "tcp_dissect_pdus()" to handle Diameter-over-TCP. svn path=/trunk/; revision=10165
2004-02-21 10:29:52 +00:00
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
strcasecmp(), strncasecmp(), g_strcasecmp(), and g_strncasecmp() delenda est. Use g_ascii_strcasecmp() and g_ascii_strncasecmp(), and supply our own versions if they're missing from GLib (as is the case with GLib 1.x). In the code to build the list of named fields for Diameter, don't use g_strdown(); do our own g_ascii_-style upper-case to lower-case mapping in the hash function and use g_ascii_strcasecmp() in the compare function. We do this because there is no guarantee that toupper(), tolower(), and functions that use them will, for example, map between "I" and "i" in all locales; in Turkish locales, for example, there are, in both upper case and lower case, versions of "i" with and without a dot, and the upper-case version of "i" is "I"-with-a-dot and the lower-case version of "I" is "i"-without-a-dot. This causes strings that should match not to match. This finishes fixing bug 2010 - an earlier checkin prevented the crash (as there are other ways to produce the same crash, e.g. a bogus dictionary.xml file), but didn't fix the case-insensitive string matching. svn path=/trunk/; revision=23623
2007-11-27 18:52:51 +00:00
/*
* This is like g_str_hash() (as of GLib 2.4.8), but it maps all
* upper-case ASCII characters to their ASCII lower-case equivalents.
* We can't use g_strdown(), as that doesn't do an ASCII mapping;
* in Turkish locales, for example, there are two lower-case "i"s
* and two upper-case "I"s, with and without dots - the ones with
* dots map between each other, as do the ones without dots, so "I"
* doesn't map to "i".
*/
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
static guint
strcase_hash(gconstpointer key)
{
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
const char *p = (const char *)key;
strcasecmp(), strncasecmp(), g_strcasecmp(), and g_strncasecmp() delenda est. Use g_ascii_strcasecmp() and g_ascii_strncasecmp(), and supply our own versions if they're missing from GLib (as is the case with GLib 1.x). In the code to build the list of named fields for Diameter, don't use g_strdown(); do our own g_ascii_-style upper-case to lower-case mapping in the hash function and use g_ascii_strcasecmp() in the compare function. We do this because there is no guarantee that toupper(), tolower(), and functions that use them will, for example, map between "I" and "i" in all locales; in Turkish locales, for example, there are, in both upper case and lower case, versions of "i" with and without a dot, and the upper-case version of "i" is "I"-with-a-dot and the lower-case version of "I" is "i"-without-a-dot. This causes strings that should match not to match. This finishes fixing bug 2010 - an earlier checkin prevented the crash (as there are other ways to produce the same crash, e.g. a bogus dictionary.xml file), but didn't fix the case-insensitive string matching. svn path=/trunk/; revision=23623
2007-11-27 18:52:51 +00:00
guint h = *p;
char c;
if (h) {
if (h >= 'A' && h <= 'Z')
h = h - 'A' + 'a';
for (p += 1; *p != '\0'; p++) {
c = *p;
if (c >= 'A' && c <= 'Z')
c = c - 'A' + 'a';
h = (h << 5) - h + c;
}
}
return h;
Support for desegmentation of Diameter-over-TCP, from David Frascone. svn path=/trunk/; revision=4141
2001-11-03 12:19:56 +00:00
}
strcasecmp(), strncasecmp(), g_strcasecmp(), and g_strncasecmp() delenda est. Use g_ascii_strcasecmp() and g_ascii_strncasecmp(), and supply our own versions if they're missing from GLib (as is the case with GLib 1.x). In the code to build the list of named fields for Diameter, don't use g_strdown(); do our own g_ascii_-style upper-case to lower-case mapping in the hash function and use g_ascii_strcasecmp() in the compare function. We do this because there is no guarantee that toupper(), tolower(), and functions that use them will, for example, map between "I" and "i" in all locales; in Turkish locales, for example, there are, in both upper case and lower case, versions of "i" with and without a dot, and the upper-case version of "i" is "I"-with-a-dot and the lower-case version of "I" is "i"-without-a-dot. This causes strings that should match not to match. This finishes fixing bug 2010 - an earlier checkin prevented the crash (as there are other ways to produce the same crash, e.g. a bogus dictionary.xml file), but didn't fix the case-insensitive string matching. svn path=/trunk/; revision=23623
2007-11-27 18:52:51 +00:00
/*
* Again, use g_ascii_strcasecmp(), not strcasecmp(), so that only ASCII
* letters are mapped, and they're mapped to the lower-case ASCII
* equivalents.
*/
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
static gboolean
strcase_equal(gconstpointer ka, gconstpointer kb)
{
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
const char *a = (const char *)ka;
const char *b = (const char *)kb;
strcasecmp(), strncasecmp(), g_strcasecmp(), and g_strncasecmp() delenda est. Use g_ascii_strcasecmp() and g_ascii_strncasecmp(), and supply our own versions if they're missing from GLib (as is the case with GLib 1.x). In the code to build the list of named fields for Diameter, don't use g_strdown(); do our own g_ascii_-style upper-case to lower-case mapping in the hash function and use g_ascii_strcasecmp() in the compare function. We do this because there is no guarantee that toupper(), tolower(), and functions that use them will, for example, map between "I" and "i" in all locales; in Turkish locales, for example, there are, in both upper case and lower case, versions of "i" with and without a dot, and the upper-case version of "i" is "I"-with-a-dot and the lower-case version of "I" is "i"-without-a-dot. This causes strings that should match not to match. This finishes fixing bug 2010 - an earlier checkin prevented the crash (as there are other ways to produce the same crash, e.g. a bogus dictionary.xml file), but didn't fix the case-insensitive string matching. svn path=/trunk/; revision=23623
2007-11-27 18:52:51 +00:00
return g_ascii_strcasecmp(a,b) == 0;
Make both the Diameter-over-TCP and Diameter-over-SCTP dissectors "new-style" dissectors, and have them reject packets that don't start with a version number of 1, in the hopes that it'll be less likely to try to dissect non-Diameter packets to/from port 1812 as Diameter and thus less likely to spew out a bunch of warnings that cause users to ask the Ethereal mailing lists what the problem is. Use "tcp_dissect_pdus()" to handle Diameter-over-TCP. svn path=/trunk/; revision=10165
2004-02-21 10:29:52 +00:00
}
Support for desegmentation of Diameter-over-TCP, from David Frascone. svn path=/trunk/; revision=4141
2001-11-03 12:19:56 +00:00
Revert "Revert "diameter: fix 400kb leaked memory on exit"" This reverts commit 5fea2b5f4198f1a36f313ef38532ddffd02ac5b1. I.e., it puts back the change; the reverted version passed the tests on which the versions with this change crashed. Change-Id: Idcc0eb11588cf14e2fe666de1905ee63917b0fcf Reviewed-on: https://code.wireshark.org/review/17413 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-08-31 08:16:24 +00:00
static gboolean
ddict_cleanup_cb(wmem_allocator_t* allocator _U_, wmem_cb_event_t event _U_, void *user_data)
{
ddict_t *d = (ddict_t *)user_data;
ddict_free(d);
return FALSE;
}
Use value_string_ext for AVP:s svn path=/trunk/; revision=32932
2010-05-24 14:20:53 +00:00
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
/* Note: Dynamic "value string arrays" (e.g., vs_avps, ...) are constructed using */
It's not kosher to alloc and init a GArray ourselves instead of calling g_array_new(). Fixes Bug #7138 (https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=7138) A puzzle: why did the buildbot fuzz-testing only start failing on this on 04/18/2012 since the bug has existed for quite some time ? svn path=/trunk/; revision=42200
2012-04-23 01:36:57 +00:00
/* "zero-terminated" GArrays so that they will have the same form as standard */
/* value_string arrays created at compile time. Since the last entry in a */
/* value_string array must be {0, NULL}, we are assuming that NULL == 0 (hackish). */
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
static int
dictionary_load(void)
Mark a local function static rather than extern. Get rid of an unused variable. Was going to use proto_register_prefix(), but am not sure if its appropriate. Even with tree == NULL, dictionary items are used to generate info column. Loading the dictionary lazily would mean doing a lot of work as part of dissecting the first diameter frame, which could decrease our chances of capturing frames that arrive very soon after that one... svn path=/trunk/; revision=24784
2008-04-05 05:41:59 +00:00
{
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
ddict_t *d;
ddict_application_t *p;
ddict_vendor_t *v;
ddict_cmd_t *c;
ddict_typedefn_t *t;
ddict_avp_t *a;
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
gboolean do_debug_parser = getenv("WIRESHARK_DEBUG_DIAM_DICT_PARSER") ? TRUE : FALSE;
gboolean do_dump_dict = getenv("WIRESHARK_DUMP_DIAM_DICT") ? TRUE : FALSE;
Get rid of some useless intermediate ep-memory when building Diameter arrays. svn path=/trunk/; revision=54421
2013-12-23 21:09:26 +00:00
char *dir;
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
const avp_type_t *type;
const avp_type_t *octetstring = &basic_types[0];
diam_avp_t *avp;
GHashTable *vendors = g_hash_table_new(strcase_hash,strcase_equal);
diam_vnd_t *vnd;
GArray *vnd_shrt_arr = g_array_new(TRUE,TRUE,sizeof(value_string));
Diameter: Allow diameter command code text to be added to custom column Create the command value string when loading the dictionary and assign it to the command code field when registering it. This makes it behave like any other value_string for adding to a custom column (allowing resolved and unresolved), etc. It could be changed to use a wmem_array instead of a GArray, as is currently done with the AVPs. Fix #16833. (Note that solving the analogous issue for the AVP code text is more complicated, since the interpretation depends on the vendor ID.)
2022-07-19 11:53:25 +00:00
GArray *all_cmds = g_array_new(TRUE,TRUE,sizeof(value_string));
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
[Diameter] Use ...array_sized_new() for hf:s and ett:s Change-Id: I5d3d3779ccf1fcd8a61e6cb6342305db2bc6e2be Reviewed-on: https://code.wireshark.org/review/18036 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-10-03 07:59:52 +00:00
/* Pre allocate the arrays big enough to hold the hf:s and etts:s*/
build_dict.hf = wmem_array_sized_new(wmem_epan_scope(), sizeof(hf_register_info), 4096);
build_dict.ett = g_ptr_array_sized_new(4096);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
build_dict.types = g_hash_table_new(strcase_hash,strcase_equal);
build_dict.avps = g_hash_table_new(strcase_hash,strcase_equal);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Use epan-scoped memory for building more of the diameter table stuff. svn path=/trunk/; revision=51754
2013-09-04 21:06:56 +00:00
dictionary.vnds = wmem_tree_new(wmem_epan_scope());
dictionary.avps = wmem_tree_new(wmem_epan_scope());
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
diameter: change GArray into wmem_array. This change fixes a leak in packet-diameter that loads a dictionary but doesn't free all the data. Found by valgrind. ==30481== 36,656 (960 direct, 35,696 indirect) bytes in 24 blocks are definitely lost in loss record 3,417 of 3,421 ==30481== at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==30481== by 0xA7FE610: g_malloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA81422D: g_slice_alloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA7CDC44: g_array_sized_new (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0x6863743: dictionary_load (packet-diameter.c:1980) ==30481== by 0x6863743: proto_register_diameter (packet-diameter.c:2344) ==30481== by 0x71C4BA4: register_all_protocols (register.c:323) ==30481== by 0x65EEFA7: proto_init (proto.c:521) ==30481== by 0x65CD621: epan_init (epan.c:126) ==30481== by 0x115330: main (tshark.c:1220) Change-Id: I3c0d19e1accab415355aa0f50c598f0c83356985 Reviewed-on: https://code.wireshark.org/review/13821 Petri-Dish: Dario Lombardo <lomato@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-02-18 10:30:41 +00:00
unknown_vendor.vs_avps = wmem_array_new(wmem_epan_scope(), sizeof(value_string));
wmem_array_set_null_terminator(unknown_vendor.vs_avps);
wmem_array_bzero(unknown_vendor.vs_avps);
no_vnd.vs_avps = wmem_array_new(wmem_epan_scope(), sizeof(value_string));
wmem_array_set_null_terminator(no_vnd.vs_avps);
wmem_array_bzero(no_vnd.vs_avps);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Use epan-scoped memory for building more of the diameter table stuff. svn path=/trunk/; revision=51754
2013-09-04 21:06:56 +00:00
wmem_tree_insert32(dictionary.vnds,0,&no_vnd);
Remove -Wwrite-strings compiler flag The "-Wwrite-strings" flag produces nuisance warnings. These warnings are not useful, they're impossible to fix in a sane way and therefore are being handled with casts of static strings to (char *). This just moves the warning to [-Wcast-qual] and a compiler pragma is in turn required (and used) to squelch that warning. Remove the Wwrite-strings warning. Let that responsibility fall on the programmer (as is done by casting). Change-Id: I5a44dfd9decd6d80797a521a3373593074962fb5 Reviewed-on: https://code.wireshark.org/review/12162 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-11-21 23:48:27 +00:00
g_hash_table_insert(vendors,"None",&no_vnd);
Add support for 64-bit signed integers in "int-64bit.[ch]", add an FT_INT64 type, and make the Diameter dissector use it. Handle the 64-bit integer types in the display filter semantics checks. svn path=/trunk/; revision=4125
2001-11-02 10:09:51 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
/* initialize the types hash with the known basic types */
for (type = basic_types; type->name; type++) {
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
g_hash_table_insert(build_dict.types,(gchar *)type->name,(void *)type);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
/* load the dictionary */
Diameter: free the diameter directory variable. It's only needed for a few milliseconds--there's no need for it to stick around until the epan scope ends. As discussed on I51813815babb4c40722c38459139ab9e3e3ccb42. Change-Id: I81fef351ef2d700cc5ec5866340605704173fbdb Reviewed-on: https://code.wireshark.org/review/15861 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-06-12 23:00:37 +00:00
dir = wmem_strdup_printf(NULL, "%s" G_DIR_SEPARATOR_S "diameter" G_DIR_SEPARATOR_S, get_datafile_dir());
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
d = ddict_scan(dir,"dictionary.xml",do_debug_parser);
Diameter: free the diameter directory variable. It's only needed for a few milliseconds--there's no need for it to stick around until the epan scope ends. As discussed on I51813815babb4c40722c38459139ab9e3e3ccb42. Change-Id: I81fef351ef2d700cc5ec5866340605704173fbdb Reviewed-on: https://code.wireshark.org/review/15861 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-06-12 23:00:37 +00:00
wmem_free(NULL, dir);
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
if (d == NULL) {
diameter: fix a memory leak on startup If we can't read the dictionary containing all our definitions, free necessary memory before returning. Change-Id: I814962d920852b9a82acb3bb2e7bc41addd835f7 Reviewed-on: https://code.wireshark.org/review/9131 Reviewed-by: Evan Huus <eapache@gmail.com> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-25 16:54:06 +00:00
g_hash_table_destroy(vendors);
g_array_free(vnd_shrt_arr, TRUE);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
return 0;
}
Revert "Revert "diameter: fix 400kb leaked memory on exit"" This reverts commit 5fea2b5f4198f1a36f313ef38532ddffd02ac5b1. I.e., it puts back the change; the reverted version passed the tests on which the versions with this change crashed. Change-Id: Idcc0eb11588cf14e2fe666de1905ee63917b0fcf Reviewed-on: https://code.wireshark.org/review/17413 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-08-31 08:16:24 +00:00
wmem_register_callback(wmem_epan_scope(), ddict_cleanup_cb, d);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
if (do_dump_dict) ddict_print(stdout, d);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
/* populate the types */
for (t = d->typedefns; t; t = t->next) {
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
const avp_type_t *parent = NULL;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
/* try to get the parent type */
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Fix crash (caused by a corrupted Diameter dictionary) reported in https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6664 : Add some sanity checks before adding types, vendors, commands, and AVPs. svn path=/trunk/; revision=41675
2012-03-19 17:12:04 +00:00
if (t->name == NULL) {
Fix some dissectors doing termio (fprintf(stderr,..), g_warning()). - Use report_...failure() (in most cases). - Also: Do some misc fixes in certain disectors - re-arrange order of #includes - Fixup preferences help text Change-Id: I385f6f97257f365f53ce611df02f57f9257dc5f9 Reviewed-on: https://code.wireshark.org/review/6039 Petri-Dish: Bill Meier <wmeier@newsguy.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-12-24 15:44:03 +00:00
report_failure("Diameter Dictionary: Invalid Type (empty name): parent==%s\n",
Fix crash (caused by a corrupted Diameter dictionary) reported in https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6664 : Add some sanity checks before adding types, vendors, commands, and AVPs. svn path=/trunk/; revision=41675
2012-03-19 17:12:04 +00:00
t->parent ? t->parent : "(null)");
continue;
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
if (g_hash_table_lookup(build_dict.types,t->name))
continue;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
if (t->parent) {
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
parent = (avp_type_t *)g_hash_table_lookup(build_dict.types,t->parent);
Don't copy the AVP data to a buffer from the tvbuff and then extract stuff from the buffer - extract what you need directly from the tvbuff. (And definitely don't extract stuff from the buffer if you haven't copied stuff to it.) For integral AVP items, check the length of the item before processing it. Put time items into the protocol tree as times, not as byte arrays. svn path=/trunk/; revision=4645
2002-01-31 01:55:14 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Check for AVPs with a list of values and a type that's not a 32-bit or shorter integral type. Fixes bug 2027. Rename the "bytes" pointer to "octetstring", and initialize it in a fashion that makes it clearer that it points to the first of the basic types, to make it clearer that it's for OctetString. svn path=/trunk/; revision=23615
2007-11-27 01:41:42 +00:00
if (!parent) parent = octetstring;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
/* insert the parent type for this type */
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
g_hash_table_insert(build_dict.types,t->name,(void *)parent);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
/* populate the applications */
if ((p = d->applications)) {
Don't mark a used parameter as unused. Also fix up some whitespace. Change-Id: I7cd78740199ce7b2682902a5687c4f05c2c963b2 Reviewed-on: https://code.wireshark.org/review/3716 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-19 13:01:43 +00:00
wmem_array_t *arr = wmem_array_new(wmem_epan_scope(), sizeof(value_string));
Revert "diameter: convert some GArrays to epan wmem_arrays" This reverts commit 59017a22be9a5a86f5c365cc22a2f1e23205597b. As indicated by Michael on the original change, this breaks diameter rather badly for some reason I haven't been able to determine. Reverting until I can track down the root cause. Change-Id: Iedfab546a726395c546a88138cd45b6667913a11 Reviewed-on: https://code.wireshark.org/review/9531 Reviewed-by: Evan Huus <eapache@gmail.com>
2015-07-07 02:06:34 +00:00
value_string term[1];
term[0].value = 0;
term[0].strptr = NULL;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
for (; p; p = p->next) {
Fix a couple of warning: initializer element is not computable at load time svn path=/trunk/; revision=54820
2014-01-15 19:29:03 +00:00
value_string item[1];
item[0].value = p->code;
item[0].strptr = p->name;
Fix nameless application in NokiaSolutionsAndNetworks.xml (from I4cd9bd7f7219e4d9ff1bb8a71fab32439a8a9a35). (The nameless application was causing known applications to be reported as unknown.) Add code to the Diameter dissector to report such problems at startup (similar code exists for other entities). Tweak the parser debug slightly. Change-Id: I6b28cda8660e6eb96648c7b3697d7fd85151ac96 Reviewed-on: https://code.wireshark.org/review/6927 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-02-03 02:28:01 +00:00
if (!p->name) {
report_failure("Diameter Dictionary: Invalid Application (empty name): id=%d\n", p->code);
continue;
}
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
wmem_array_append_one(arr,item);
Don't copy the AVP data to a buffer from the tvbuff and then extract stuff from the buffer - extract what you need directly from the tvbuff. (And definitely don't extract stuff from the buffer if you haven't copied stuff to it.) For integral AVP items, check the length of the item before processing it. Put time items into the protocol tree as times, not as byte arrays. svn path=/trunk/; revision=4645
2002-01-31 01:55:14 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Use wmem for application Ids Remove soft deprecated APIs Change-Id: Iecd2f552c1abb7bb2c6969073b3c245af22f4fcd Reviewed-on: https://code.wireshark.org/review/3230 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 18:18:49 +00:00
wmem_array_sort(arr, compare_avps);
wmem_array_append_one(arr,term);
Sort diameter.applicationId fix a typo. Change-Id: Id585371df929cdff17f98dddbebf063fa000ff56 Reviewed-on: https://code.wireshark.org/review/3229 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 17:35:02 +00:00
Don't mark a used parameter as unused. Also fix up some whitespace. Change-Id: I7cd78740199ce7b2682902a5687c4f05c2c963b2 Reviewed-on: https://code.wireshark.org/review/3716 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-19 13:01:43 +00:00
/* TODO: Remove duplicates */
Sort diameter.applicationId fix a typo. Change-Id: Id585371df929cdff17f98dddbebf063fa000ff56 Reviewed-on: https://code.wireshark.org/review/3229 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2014-07-28 17:35:02 +00:00
Add special handling of the (derived) AppId AVP type so that we can maintain one (and only one) Application ID list. This means we don't have to list all the Application IDs several times in the XML file. Collapse those lists into one (much more complete, now) list while also fixing URIs for several of the specifications (and making it clear for others that we don't have access to the specification--by declaring their URI to be "none"). Add a bunch more entries to the list of application IDs. Change-Id: Ia5c96b1f6f1fe3a9521b3d70142889e3881fae5e Reviewed-on: https://code.wireshark.org/review/8147 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-04-20 22:14:35 +00:00
dictionary.applications = value_string_ext_new((value_string *)wmem_array_get_raw(arr),
wmem_array_get_count(arr),
Use faster string functions at startup. Use wmem_strconcat and g_strconcat instead of wmem_strdup_printf and g_strdup_printf when we register various protocols. This shows a fairly significant speedup in the Visual Studio profiler. Change-Id: I98709329513daa66ad3665925dc69149c43df884 Reviewed-on: https://code.wireshark.org/review/14855 Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Gerald Combs <gerald@wireshark.org>
2016-04-07 20:05:27 +00:00
wmem_strdup(wmem_epan_scope(), "applications_vals_ext"));
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
if ((v = d->vendors)) {
for ( ; v; v = v->next) {
Fix a couple of warning: initializer element is not computable at load time svn path=/trunk/; revision=54820
2014-01-15 19:29:03 +00:00
value_string item[1];
item[0].value = v->code;
item[0].strptr = v->name;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Fix crash (caused by a corrupted Diameter dictionary) reported in https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6664 : Add some sanity checks before adding types, vendors, commands, and AVPs. svn path=/trunk/; revision=41675
2012-03-19 17:12:04 +00:00
if (v->name == NULL) {
Fix nameless application in NokiaSolutionsAndNetworks.xml (from I4cd9bd7f7219e4d9ff1bb8a71fab32439a8a9a35). (The nameless application was causing known applications to be reported as unknown.) Add code to the Diameter dissector to report such problems at startup (similar code exists for other entities). Tweak the parser debug slightly. Change-Id: I6b28cda8660e6eb96648c7b3697d7fd85151ac96 Reviewed-on: https://code.wireshark.org/review/6927 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-02-03 02:28:01 +00:00
report_failure("Diameter Dictionary: Invalid Vendor (empty name): code==%d\n", v->code);
Fix crash (caused by a corrupted Diameter dictionary) reported in https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6664 : Add some sanity checks before adding types, vendors, commands, and AVPs. svn path=/trunk/; revision=41675
2012-03-19 17:12:04 +00:00
continue;
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
if (g_hash_table_lookup(vendors,v->name))
continue;
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
g_array_append_val(vnd_shrt_arr,item);
Clean up white space. Just use "proto_tree_add_item()" to put the session ID into the protocol string. svn path=/trunk/; revision=10431
2004-03-21 23:19:36 +00:00
Use epan-scoped memory for building more of the diameter table stuff. svn path=/trunk/; revision=51754
2013-09-04 21:06:56 +00:00
vnd = wmem_new(wmem_epan_scope(), diam_vnd_t);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
vnd->code = v->code;
diameter: change GArray into wmem_array. This change fixes a leak in packet-diameter that loads a dictionary but doesn't free all the data. Found by valgrind. ==30481== 36,656 (960 direct, 35,696 indirect) bytes in 24 blocks are definitely lost in loss record 3,417 of 3,421 ==30481== at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==30481== by 0xA7FE610: g_malloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA81422D: g_slice_alloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA7CDC44: g_array_sized_new (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0x6863743: dictionary_load (packet-diameter.c:1980) ==30481== by 0x6863743: proto_register_diameter (packet-diameter.c:2344) ==30481== by 0x71C4BA4: register_all_protocols (register.c:323) ==30481== by 0x65EEFA7: proto_init (proto.c:521) ==30481== by 0x65CD621: epan_init (epan.c:126) ==30481== by 0x115330: main (tshark.c:1220) Change-Id: I3c0d19e1accab415355aa0f50c598f0c83356985 Reviewed-on: https://code.wireshark.org/review/13821 Petri-Dish: Dario Lombardo <lomato@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-02-18 10:30:41 +00:00
vnd->vs_avps = wmem_array_new(wmem_epan_scope(), sizeof(value_string));
wmem_array_set_null_terminator(vnd->vs_avps);
wmem_array_bzero(vnd->vs_avps);
Rework "extended value strings": - Allow direct access when a range of values begins with a value other than 0; - Provide value_string_ext_new() for creating extended value strings at runtime; - Do access to value_string_ext members via a macro (all but value_string.c); - Update documentation. svn path=/trunk/; revision=34514
2010-10-14 17:50:35 +00:00
vnd->vs_avps_ext = NULL;
Use epan-scoped memory for building more of the diameter table stuff. svn path=/trunk/; revision=51754
2013-09-04 21:06:56 +00:00
wmem_tree_insert32(dictionary.vnds,vnd->code,vnd);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
g_hash_table_insert(vendors,v->name,vnd);
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Free g_array_free-related memory leaks g_array_free(a, FALSE) returns "a->data". Callers that do not handle this will leak memory. Convert other users to use the return value instead of direct access to "a->data". Change-Id: I0a29864e8106c0bf09e9573ef29e4474179c4171 Reviewed-on: https://code.wireshark.org/review/27438 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Peter Wu <peter@lekensteyn.nl>
2018-05-09 20:37:41 +00:00
vnd_short_vs = (value_string *)g_array_free(vnd_shrt_arr, FALSE);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
if ((c = d->cmds)) {
for (; c; c = c->next) {
Fix crash (caused by a corrupted Diameter dictionary) reported in https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6664 : Add some sanity checks before adding types, vendors, commands, and AVPs. svn path=/trunk/; revision=41675
2012-03-19 17:12:04 +00:00
if (c->vendor == NULL) {
Fix some dissectors doing termio (fprintf(stderr,..), g_warning()). - Use report_...failure() (in most cases). - Also: Do some misc fixes in certain disectors - re-arrange order of #includes - Fixup preferences help text Change-Id: I385f6f97257f365f53ce611df02f57f9257dc5f9 Reviewed-on: https://code.wireshark.org/review/6039 Petri-Dish: Bill Meier <wmeier@newsguy.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-12-24 15:44:03 +00:00
report_failure("Diameter Dictionary: Invalid Vendor (empty name) for command %s\n",
Fix crash (caused by a corrupted Diameter dictionary) reported in https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6664 : Add some sanity checks before adding types, vendors, commands, and AVPs. svn path=/trunk/; revision=41675
2012-03-19 17:12:04 +00:00
c->name ? c->name : "(null)");
continue;
}
diameter: Fix Clang Warning line 2198 Although the value stored to 'vnd' is used in the enclosing expression, the value is never actually read from 'vnd'
2022-09-06 11:25:15 +00:00
if ((diam_vnd_t *)g_hash_table_lookup(vendors,c->vendor)) {
Fix a couple of warning: initializer element is not computable at load time svn path=/trunk/; revision=54820
2014-01-15 19:29:03 +00:00
value_string item[1];
item[0].value = c->code;
item[0].strptr = c->name;
- add an all_cmds array for looking up 'RFC' packets - improve error/expert info when see unmatched AVPs, commands, appids svn path=/trunk/; revision=22363
2007-07-20 11:39:09 +00:00
g_array_append_val(all_cmds,item);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
} else {
Fix some dissectors doing termio (fprintf(stderr,..), g_warning()). - Use report_...failure() (in most cases). - Also: Do some misc fixes in certain disectors - re-arrange order of #includes - Fixup preferences help text Change-Id: I385f6f97257f365f53ce611df02f57f9257dc5f9 Reviewed-on: https://code.wireshark.org/review/6039 Petri-Dish: Bill Meier <wmeier@newsguy.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-12-24 15:44:03 +00:00
report_failure("Diameter Dictionary: No Vendor: %s\n",c->vendor);
From Martin Mathieson: packet-diameter.c -------------------------- I completely reindented dissect_avps() before I made any changes, but when ignoring white space (in tkdiff, -w plus checking 'Ignore blanks when diffing'), its easy to see the small changes I've made: - when fail to find AVP info, show code in tree parent in decimal (as specs do) - add an expert info (undecoded, note) to indicate unknown AVP codes diameter/imscxdx.xml ------------------------------- - added 'Associated-Identities' svn path=/trunk/; revision=18641
2006-07-03 21:18:36 +00:00
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
for (a = d->avps; a; a = a->next) {
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
ddict_enum_t *e;
value_string *vs = NULL;
const char *vend = a->vendor ? a->vendor : "None";
ddict_xmlpi_t *x;
void *avp_data = NULL;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Fix crash (caused by a corrupted Diameter dictionary) reported in https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6664 : Add some sanity checks before adding types, vendors, commands, and AVPs. svn path=/trunk/; revision=41675
2012-03-19 17:12:04 +00:00
if (a->name == NULL) {
Fix some dissectors doing termio (fprintf(stderr,..), g_warning()). - Use report_...failure() (in most cases). - Also: Do some misc fixes in certain disectors - re-arrange order of #includes - Fixup preferences help text Change-Id: I385f6f97257f365f53ce611df02f57f9257dc5f9 Reviewed-on: https://code.wireshark.org/review/6039 Petri-Dish: Bill Meier <wmeier@newsguy.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-12-24 15:44:03 +00:00
report_failure("Diameter Dictionary: Invalid AVP (empty name)\n");
Fix crash (caused by a corrupted Diameter dictionary) reported in https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=6664 : Add some sanity checks before adding types, vendors, commands, and AVPs. svn path=/trunk/; revision=41675
2012-03-19 17:12:04 +00:00
continue;
}
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
if ((vnd = (diam_vnd_t *)g_hash_table_lookup(vendors,vend))) {
Fix a couple of warning: initializer element is not computable at load time svn path=/trunk/; revision=54820
2014-01-15 19:29:03 +00:00
value_string vndvs[1];
vndvs[0].value = a->code;
vndvs[0].strptr = a->name;
Revert "diameter: convert some GArrays to epan wmem_arrays" This reverts commit 59017a22be9a5a86f5c365cc22a2f1e23205597b. As indicated by Michael on the original change, this breaks diameter rather badly for some reason I haven't been able to determine. Reverting until I can track down the root cause. Change-Id: Iedfab546a726395c546a88138cd45b6667913a11 Reviewed-on: https://code.wireshark.org/review/9531 Reviewed-by: Evan Huus <eapache@gmail.com>
2015-07-07 02:06:34 +00:00
diameter: change GArray into wmem_array. This change fixes a leak in packet-diameter that loads a dictionary but doesn't free all the data. Found by valgrind. ==30481== 36,656 (960 direct, 35,696 indirect) bytes in 24 blocks are definitely lost in loss record 3,417 of 3,421 ==30481== at 0x4C2AB80: malloc (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so) ==30481== by 0xA7FE610: g_malloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA81422D: g_slice_alloc (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0xA7CDC44: g_array_sized_new (in /lib/x86_64-linux-gnu/libglib-2.0.so.0.4002.0) ==30481== by 0x6863743: dictionary_load (packet-diameter.c:1980) ==30481== by 0x6863743: proto_register_diameter (packet-diameter.c:2344) ==30481== by 0x71C4BA4: register_all_protocols (register.c:323) ==30481== by 0x65EEFA7: proto_init (proto.c:521) ==30481== by 0x65CD621: epan_init (epan.c:126) ==30481== by 0x115330: main (tshark.c:1220) Change-Id: I3c0d19e1accab415355aa0f50c598f0c83356985 Reviewed-on: https://code.wireshark.org/review/13821 Petri-Dish: Dario Lombardo <lomato@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Evan Huus <eapache@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-02-18 10:30:41 +00:00
wmem_array_append_one(vnd->vs_avps,vndvs);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
} else {
Fix some dissectors doing termio (fprintf(stderr,..), g_warning()). - Use report_...failure() (in most cases). - Also: Do some misc fixes in certain disectors - re-arrange order of #includes - Fixup preferences help text Change-Id: I385f6f97257f365f53ce611df02f57f9257dc5f9 Reviewed-on: https://code.wireshark.org/review/6039 Petri-Dish: Bill Meier <wmeier@newsguy.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2014-12-24 15:44:03 +00:00
report_failure("Diameter Dictionary: No Vendor: %s\n",vend);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
vnd = &unknown_vendor;
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
if ((e = a->enums)) {
Build dynamic diameter value_string arrays with wmem arrays not garrays. Fixes another ~60KB of leaks. svn path=/trunk/; revision=50412
2013-07-06 18:01:57 +00:00
wmem_array_t *arr = wmem_array_new(wmem_epan_scope(), sizeof(value_string));
Revert "diameter: convert some GArrays to epan wmem_arrays" This reverts commit 59017a22be9a5a86f5c365cc22a2f1e23205597b. As indicated by Michael on the original change, this breaks diameter rather badly for some reason I haven't been able to determine. Reverting until I can track down the root cause. Change-Id: Iedfab546a726395c546a88138cd45b6667913a11 Reviewed-on: https://code.wireshark.org/review/9531 Reviewed-by: Evan Huus <eapache@gmail.com>
2015-07-07 02:06:34 +00:00
value_string term[1];
term[0].value = 0;
term[0].strptr = NULL;
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
for (; e; e = e->next) {
Fix a couple of warning: initializer element is not computable at load time svn path=/trunk/; revision=54820
2014-01-15 19:29:03 +00:00
value_string item[1];
item[0].value = e->code;
item[0].strptr = e->name;
Build dynamic diameter value_string arrays with wmem arrays not garrays. Fixes another ~60KB of leaks. svn path=/trunk/; revision=50412
2013-07-06 18:01:57 +00:00
wmem_array_append_one(arr,item);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
Build dynamic diameter value_string arrays with wmem arrays not garrays. Fixes another ~60KB of leaks. svn path=/trunk/; revision=50412
2013-07-06 18:01:57 +00:00
wmem_array_sort(arr, compare_avps);
wmem_array_append_one(arr,term);
vs = (value_string *)wmem_array_get_raw(arr);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
type = NULL;
for( x = d->xmlpis; x; x = x->next ) {
if ( (strcase_equal(x->name,"avp-proto") && strcase_equal(x->key,a->name))
|| (a->type && strcase_equal(x->name,"type-proto") && strcase_equal(x->key,a->type))
) {
Diameter: Remove dead code for pre Internet Draft 16 protocol There's some code intended to decode an older Internet Draft version of Diameter, before December 2002. It's supposed to be set by an obsolete enum preference, but due to a misunderstanding, it's been dead code for the last 15 years. No one has complained in that time, so instead of bringing the preference back and resurrecting it, remove it and shrink the memory usage of the diameter dissector. Related to #16833, as it will make adding the command code value string simpler.
2022-07-19 03:13:28 +00:00
static avp_type_t proto_type = {"proto", proto_avp, FT_UINT32, BASE_HEX, build_proto_avp};
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
type = &proto_type;
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
avp_data = x->value;
break;
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
if ( (!type) && a->type )
From beroset: remove C++ incompatibilities https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8416 svn path=/trunk/; revision=48013
2013-03-02 19:21:38 +00:00
type = (avp_type_t *)g_hash_table_lookup(build_dict.types,a->type);
Remove a couple of unused variables. svn path=/trunk/; revision=23266
2007-10-24 16:32:51 +00:00
Check for AVPs with a list of values and a type that's not a 32-bit or shorter integral type. Fixes bug 2027. Rename the "bytes" pointer to "octetstring", and initialize it in a fashion that makes it clearer that it points to the first of the basic types, to make it clearer that it's for OctetString. svn path=/trunk/; revision=23615
2007-11-27 01:41:42 +00:00
if (!type) type = octetstring;
- <?avp-proto and <?type-proto to instruct about which dissector to use for a given type or avp - dissect timestamps - add all the avps that were commented out in the dictionary svn path=/trunk/; revision=22360
2007-07-20 00:15:17 +00:00
avp = type->build( type, a->code, vnd, a->name, vs, avp_data);
Check for AVPs with a list of values and a type that's not a 32-bit or shorter integral type. Fixes bug 2027. Rename the "bytes" pointer to "octetstring", and initialize it in a fashion that makes it clearer that it points to the first of the basic types, to make it clearer that it's for OctetString. svn path=/trunk/; revision=23615
2007-11-27 01:41:42 +00:00
if (avp != NULL) {
g_hash_table_insert(build_dict.avps, a->name, avp);
{
Fix a couple of warning: initializer element is not computable at load time svn path=/trunk/; revision=54820
2014-01-15 19:29:03 +00:00
wmem_tree_key_t k[3];
k[0].length = 1;
k[0].key = &(a->code);
k[1].length = 1;
k[1].key = &(vnd->code);
k[2].length = 0;
k[2].key = NULL;
Use epan-scoped memory for building more of the diameter table stuff. svn path=/trunk/; revision=51754
2013-09-04 21:06:56 +00:00
wmem_tree_insert32_array(dictionary.avps,k,avp);
Check for AVPs with a list of values and a type that's not a 32-bit or shorter integral type. Fixes bug 2027. Rename the "bytes" pointer to "octetstring", and initialize it in a fashion that makes it clearer that it points to the first of the basic types, to make it clearer that it's for OctetString. svn path=/trunk/; revision=23615
2007-11-27 01:41:42 +00:00
}
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
}
}
g_hash_table_destroy(build_dict.types);
g_hash_table_destroy(build_dict.avps);
g_hash_table_destroy(vendors);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Diameter: Allow diameter command code text to be added to custom column Create the command value string when loading the dictionary and assign it to the command code field when registering it. This makes it behave like any other value_string for adding to a custom column (allowing resolved and unresolved), etc. It could be changed to use a wmem_array instead of a GArray, as is currently done with the AVPs. Fix #16833. (Note that solving the analogous issue for the AVP code text is more complicated, since the interpretation depends on the vendor ID.)
2022-07-19 11:53:25 +00:00
cmd_vs = (const value_string *)(void *)all_cmds->data;
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
return 1;
}
Updates from David Frascone: includes DIAMETER-over-SCTP support. svn path=/trunk/; revision=3073
2001-02-23 19:26:26 +00:00
Slightly less hacky way to arrange that dictionary_load() be called before hf_base[] is initialized. svn path=/trunk/; revision=46271
2012-11-28 23:11:08 +00:00
/*
Delay registration of Diameter fields until they're needed. ... Like the RADIUS and wimaxasncp dissectors do. Change-Id: Ifab019a0040d3938e52918a282a3beba9dfcfc70 Reviewed-on: https://code.wireshark.org/review/14900 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-04-13 03:01:18 +00:00
* This does most of the registration work; see register_diameter_fields()
Slightly less hacky way to arrange that dictionary_load() be called before hf_base[] is initialized. svn path=/trunk/; revision=46271
2012-11-28 23:11:08 +00:00
* for the reason why we split it off.
*/
static void
Delay registration of Diameter fields until they're needed. ... Like the RADIUS and wimaxasncp dissectors do. Change-Id: Ifab019a0040d3938e52918a282a3beba9dfcfc70 Reviewed-on: https://code.wireshark.org/review/14900 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-04-13 03:01:18 +00:00
real_register_diameter_fields(void)
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
{
Convert remaining proto_item_set_expert_flags in dissector directory. svn path=/trunk/; revision=51142
2013-08-05 01:54:15 +00:00
expert_module_t* expert_diameter;
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
guint i, ett_length;
Mark a local function static rather than extern. Get rid of an unused variable. Was going to use proto_register_prefix(), but am not sure if its appropriate. Even with tree == NULL, dictionary items are used to generate info column. Loading the dictionary lazily would mean doing a lot of work as part of dissecting the first diameter frame, which could decrease our chances of capturing frames that arrive very soon after that one... svn path=/trunk/; revision=24784
2008-04-05 05:41:59 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
hf_register_info hf_base[] = {
{ &hf_diameter_version,
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
{ "Version", "diameter.version", FT_UINT8, BASE_HEX, NULL, 0x00,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_length,
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
{ "Length","diameter.length", FT_UINT24, BASE_DEC, NULL, 0x0,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_flags,
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
{ "Flags", "diameter.flags", FT_UINT8, BASE_HEX, NULL, 0x0,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_flags_request,
Changed flags_set_truth -> tfs_set_notset svn path=/trunk/; revision=28989
2009-07-07 14:54:15 +00:00
{ "Request", "diameter.flags.request", FT_BOOLEAN, 8, TFS(&tfs_set_notset), DIAM_FLAGS_R,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_flags_proxyable,
Changed flags_set_truth -> tfs_set_notset svn path=/trunk/; revision=28989
2009-07-07 14:54:15 +00:00
{ "Proxyable", "diameter.flags.proxyable", FT_BOOLEAN, 8, TFS(&tfs_set_notset), DIAM_FLAGS_P,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_flags_error,
Changed flags_set_truth -> tfs_set_notset svn path=/trunk/; revision=28989
2009-07-07 14:54:15 +00:00
{ "Error","diameter.flags.error", FT_BOOLEAN, 8, TFS(&tfs_set_notset), DIAM_FLAGS_E,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_flags_T,
Fix pretty-printing of the application ID in the protocol-details pane by putting an "unused" variable (which was taken out in r24377) back in. This time add a (bogus) "use" of the variable and explain why we're doing what we're doing. Use VALS() (in one case anyway) and PINFO_FD_NUM. svn path=/trunk/; revision=46266
2012-11-28 22:45:31 +00:00
{ "T(Potentially re-transmitted message)","diameter.flags.T", FT_BOOLEAN, 8, TFS(&tfs_set_notset), DIAM_FLAGS_T,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_flags_reserved4,
Use some default true_false_string's. svn path=/trunk/; revision=27983
2009-04-07 12:20:41 +00:00
{ "Reserved","diameter.flags.reserved4", FT_BOOLEAN, 8, TFS(&tfs_set_notset),
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
DIAM_FLAGS_RESERVED4, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_flags_reserved5,
Use some default true_false_string's. svn path=/trunk/; revision=27983
2009-04-07 12:20:41 +00:00
{ "Reserved","diameter.flags.reserved5", FT_BOOLEAN, 8, TFS(&tfs_set_notset),
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
DIAM_FLAGS_RESERVED5, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_flags_reserved6,
Use some default true_false_string's. svn path=/trunk/; revision=27983
2009-04-07 12:20:41 +00:00
{ "Reserved","diameter.flags.reserved6", FT_BOOLEAN, 8, TFS(&tfs_set_notset),
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
DIAM_FLAGS_RESERVED6, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_flags_reserved7,
Use some default true_false_string's. svn path=/trunk/; revision=27983
2009-04-07 12:20:41 +00:00
{ "Reserved","diameter.flags.reserved7", FT_BOOLEAN, 8, TFS(&tfs_set_notset),
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
DIAM_FLAGS_RESERVED7, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_vendor_id,
Parse enterprise-numbers at run time "enterprise-numbers" is converted to tab-separated values and renamed "enterprises". Unused fields are stripped. PENs are stored in a hash table loaded at run-time. User "enterprises" file is loaded from the personal config dir. Misc make-sminmpec.pl improvements and fixes. Note: names of type "Entity (formerly ...)" have the formerly part commented out for a cleaner output. Change-Id: I60c533afbe3e399077fbf432088064471ad3e1e2 Reviewed-on: https://code.wireshark.org/review/22246 Petri-Dish: João Valverde <j@v6e.pt> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Reviewed-by: João Valverde <j@v6e.pt>
2017-06-24 13:25:41 +00:00
{ "VendorId", "diameter.vendorId", FT_UINT32, BASE_ENTERPRISES, STRINGS_ENTERPRISES,
Fix pretty-printing of the application ID in the protocol-details pane by putting an "unused" variable (which was taken out in r24377) back in. This time add a (bogus) "use" of the variable and explain why we're doing what we're doing. Use VALS() (in one case anyway) and PINFO_FD_NUM. svn path=/trunk/; revision=46266
2012-11-28 22:45:31 +00:00
0x0, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_application_id,
Use VALS_EXT_PTR to satisfy checkAPIs. Change-Id: Ic7a7cfb97be5c3fbca5944328d04d505b54242ad Reviewed-on: https://code.wireshark.org/review/8176 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-04-23 13:10:38 +00:00
{ "ApplicationId", "diameter.applicationId", FT_UINT32, BASE_DEC|BASE_EXT_STRING, VALS_EXT_PTR(dictionary.applications),
Fix pretty-printing of the application ID in the protocol-details pane by putting an "unused" variable (which was taken out in r24377) back in. This time add a (bogus) "use" of the variable and explain why we're doing what we're doing. Use VALS() (in one case anyway) and PINFO_FD_NUM. svn path=/trunk/; revision=46266
2012-11-28 22:45:31 +00:00
0x0, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_hopbyhopid,
Updates from David Frascone. svn path=/trunk/; revision=3287
2001-04-10 21:49:23 +00:00
{ "Hop-by-Hop Identifier", "diameter.hopbyhopid", FT_UINT32,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
BASE_HEX, NULL, 0x0, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_endtoendid,
Removed trailing whitespaces from .h and .c files using the winapi_cleanup tool written by Patrik Stridvall for the wine project. svn path=/trunk/; revision=6117
2002-08-28 21:04:11 +00:00
{ "End-to-End Identifier", "diameter.endtoendid", FT_UINT32,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
BASE_HEX, NULL, 0x0, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
{ "AVP","diameter.avp", FT_BYTES, BASE_NONE,
NULL, 0x0, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_len,
{ "AVP Length","diameter.avp.len", FT_UINT24, BASE_DEC,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, 0x0, NULL, HFILL }},
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
{ &hf_diameter_avp_code,
{ "AVP Code", "diameter.avp.code", FT_UINT32, BASE_DEC, NULL, 0, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_flags,
New tvbuffified DIAMETER dissector, from David Frascone. It doesn't do DIAMETER-over-UDP, so the RADIUS dissector no longer checks for DIAMETER packets and calls the DIAMETER dissector if it finds one. svn path=/trunk/; revision=3049
2001-02-19 23:14:02 +00:00
{ "AVP Flags","diameter.avp.flags", FT_UINT8, BASE_HEX,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, 0x0, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_flags_vendor_specific,
Changed flags_set_truth -> tfs_set_notset svn path=/trunk/; revision=28989
2009-07-07 14:54:15 +00:00
{ "Vendor-Specific", "diameter.flags.vendorspecific", FT_BOOLEAN, 8, TFS(&tfs_set_notset), AVP_FLAGS_V,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_flags_mandatory,
Changed flags_set_truth -> tfs_set_notset svn path=/trunk/; revision=28989
2009-07-07 14:54:15 +00:00
{ "Mandatory", "diameter.flags.mandatory", FT_BOOLEAN, 8, TFS(&tfs_set_notset), AVP_FLAGS_M,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_flags_protected,
Changed flags_set_truth -> tfs_set_notset svn path=/trunk/; revision=28989
2009-07-07 14:54:15 +00:00
{ "Protected","diameter.avp.flags.protected", FT_BOOLEAN, 8, TFS(&tfs_set_notset), AVP_FLAGS_P,
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_flags_reserved3,
Use some default true_false_string's. svn path=/trunk/; revision=27983
2009-04-07 12:20:41 +00:00
{ "Reserved","diameter.avp.flags.reserved3", FT_BOOLEAN, 8, TFS(&tfs_set_notset),
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
AVP_FLAGS_RESERVED3, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_flags_reserved4,
Use some default true_false_string's. svn path=/trunk/; revision=27983
2009-04-07 12:20:41 +00:00
{ "Reserved","diameter.avp.flags.reserved4", FT_BOOLEAN, 8, TFS(&tfs_set_notset),
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
AVP_FLAGS_RESERVED4, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_flags_reserved5,
Use some default true_false_string's. svn path=/trunk/; revision=27983
2009-04-07 12:20:41 +00:00
{ "Reserved","diameter.avp.flags.reserved5", FT_BOOLEAN, 8, TFS(&tfs_set_notset),
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
AVP_FLAGS_RESERVED5, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_flags_reserved6,
Use some default true_false_string's. svn path=/trunk/; revision=27983
2009-04-07 12:20:41 +00:00
{ "Reserved","diameter.avp.flags.reserved6", FT_BOOLEAN, 8, TFS(&tfs_set_notset),
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
AVP_FLAGS_RESERVED6, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_flags_reserved7,
Use some default true_false_string's. svn path=/trunk/; revision=27983
2009-04-07 12:20:41 +00:00
{ "Reserved","diameter.avp.flags.reserved7", FT_BOOLEAN, 8, TFS(&tfs_set_notset),
From Kovarththanan Rajaratnam via bug 3548: (1) Trailing/leading spaces are removed from 'name's/'blurb's (2) Duplicate 'blurb's are replaced with NULL (3) Empty ("") 'blurb's are replaced with NULL (4) BASE_NONE, NULL, 0x0 are used for 'display', 'strings' and 'bitmask' fields for FT_NONE, FT_BYTES, FT_IPv4, FT_IPv6, FT_ABSOLUTE_TIME, FT_RELATIVE_TIME, FT_PROTOCOL, FT_STRING and FT_STRINGZ field types (5) Only allow non-zero value for 'display' if 'bitmask' is non-zero svn path=/trunk/; revision=28770
2009-06-18 21:30:42 +00:00
AVP_FLAGS_RESERVED7, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_avp_vendor_id,
Parse enterprise-numbers at run time "enterprise-numbers" is converted to tab-separated values and renamed "enterprises". Unused fields are stripped. PENs are stored in a hash table loaded at run-time. User "enterprises" file is loaded from the personal config dir. Misc make-sminmpec.pl improvements and fixes. Note: names of type "Entity (formerly ...)" have the formerly part commented out for a cleaner output. Change-Id: I60c533afbe3e399077fbf432088064471ad3e1e2 Reviewed-on: https://code.wireshark.org/review/22246 Petri-Dish: João Valverde <j@v6e.pt> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Peter Wu <peter@lekensteyn.nl> Reviewed-by: João Valverde <j@v6e.pt>
2017-06-24 13:25:41 +00:00
{ "AVP Vendor Id","diameter.avp.vendorId", FT_UINT32, BASE_ENTERPRISES, STRINGS_ENTERPRISES,
0x0, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &(unknown_avp.hf_value),
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
{ "Value","diameter.avp.unknown", FT_BYTES, BASE_NONE, NULL, 0x0, NULL, HFILL }},
Check the length in unsigned32 AVP data fields. svn path=/trunk/; revision=22413
2007-07-27 18:32:09 +00:00
{ &hf_diameter_avp_data_wrong_length,
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
{ "Data","diameter.avp.invalid-data", FT_BYTES, BASE_NONE, NULL, 0x0, NULL, HFILL }},
{ &hf_diameter_avp_pad,
{ "Padding","diameter.avp.pad", FT_BYTES, BASE_NONE, NULL, 0x0, NULL, HFILL }},
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
{ &hf_diameter_code,
Diameter: Allow diameter command code text to be added to custom column Create the command value string when loading the dictionary and assign it to the command code field when registering it. This makes it behave like any other value_string for adding to a custom column (allowing resolved and unresolved), etc. It could be changed to use a wmem_array instead of a GArray, as is currently done with the AVPs. Fix #16833. (Note that solving the analogous issue for the AVP code text is more complicated, since the interpretation depends on the vendor ID.)
2022-07-19 11:53:25 +00:00
{ "Command Code", "diameter.cmd.code", FT_UINT32, BASE_DEC, VALS(cmd_vs), 0, NULL, HFILL }},
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
{ &hf_diameter_answer_in,
Use the new request/response framenum hints. Change-Id: Ic4a3750a55523a5cf8ea72002055ffea1f081dd1 Reviewed-on: https://code.wireshark.org/review/7565 Reviewed-by: Michael Mann <mmann78@netscape.net> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-03-06 23:31:41 +00:00
{ "Answer In", "diameter.answer_in", FT_FRAMENUM, BASE_NONE, FRAMENUM_TYPE(FT_FRAMENUM_RESPONSE), 0x0,
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
"The answer to this diameter request is in this frame", HFILL }},
{ &hf_diameter_answer_to,
Use the new request/response framenum hints. Change-Id: Ic4a3750a55523a5cf8ea72002055ffea1f081dd1 Reviewed-on: https://code.wireshark.org/review/7565 Reviewed-by: Michael Mann <mmann78@netscape.net> Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2015-03-06 23:31:41 +00:00
{ "Request In", "diameter.answer_to", FT_FRAMENUM, BASE_NONE, FRAMENUM_TYPE(FT_FRAMENUM_REQUEST), 0x0,
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
"This is an answer to the diameter request in this frame", HFILL }},
{ &hf_diameter_answer_time,
{ "Response Time", "diameter.resp_time", FT_RELATIVE_TIME, BASE_NONE, NULL, 0x0,
"The time between the request and the answer", HFILL }},
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
{ &hf_framed_ipv6_prefix_reserved,
{ "Framed IPv6 Prefix Reserved byte", "diameter.framed_ipv6_prefix_reserved",
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
FT_UINT8, BASE_HEX, NULL, 0, NULL, HFILL }},
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
{ &hf_framed_ipv6_prefix_length,
{ "Framed IPv6 Prefix length (in bits)", "diameter.framed_ipv6_prefix_length",
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
FT_UINT8, BASE_DEC, NULL, 0, NULL, HFILL }},
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
{ &hf_framed_ipv6_prefix_bytes,
{ "Framed IPv6 Prefix as a bytestring", "diameter.framed_ipv6_prefix_bytes",
And it happens again... In Yet Another Protocol, implementation A neglected to set the padding bytes to 0 and implementation B barfed on said padding (interestingly this protocol's spec does not include the IETF-normal "receiver MUST ignore the padding" blurb). So: Add the AVP to the dissection tree and add an expert info for when it's not zero. Also re-order a few of the hfs and remove a couple unneeded temporary variables. svn path=/trunk/; revision=48488
2013-03-22 18:13:31 +00:00
FT_BYTES, BASE_NONE, NULL, 0, NULL, HFILL }},
Add detailed decoding for Framed-IPv6-Prefix. svn path=/trunk/; revision=41532
2012-03-13 15:06:40 +00:00
{ &hf_framed_ipv6_prefix_ipv6,
{ "Framed IPv6 Prefix as an IPv6 address", "diameter.framed_ipv6_prefix_ipv6",
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
FT_IPv6, BASE_NONE, NULL, 0, "This field is present only if the prefix length is 128", HFILL }},
{ &hf_diameter_3gpp2_exp_res,
Fix a typo. svn path=/trunk/; revision=54597
2014-01-04 23:41:22 +00:00
{ "Experimental-Result-Code", "diameter.3gpp2.exp_res",
Show the Experimental-Result-Code if we don't know have a subdissector for that vendor ID. Otherwise the value is simply not shown to the user. Adding support for a vendor ID's Experimental Result Codes isn't as easy as modifying the XML so don't add an expert info about it. Change-Id: I65f2cb13853cc7141fb242fa03c6e474a6c02cb9 Reviewed-on: https://code.wireshark.org/review/11294 Reviewed-by: Michael Mann <mmann78@netscape.net> Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-10-26 20:44:18 +00:00
FT_UINT32, BASE_DEC, VALS(diameter_3gpp2_exp_res_vals), 0x0, NULL, HFILL }},
{ &hf_diameter_other_vendor_exp_res,
{ "Experimental-Result-Code", "diameter.other_vendor.Experimental-Result-Code",
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
FT_UINT32, BASE_DEC, NULL, 0x0, NULL, HFILL }},
{ &hf_diameter_mip6_feature_vector,
{ "MIP6-Feature-Vector", "diameter.mip6_feature_vector",
FT_UINT64, BASE_HEX, NULL, 0x0, NULL, HFILL }},
{ &hf_diameter_mip6_feature_vector_mip6_integrated,
{ "MIP6_INTEGRATED", "diameter.mip6_feature_vector.mip6_integrated.mip6_integrated",
FT_BOOLEAN, 64, TFS(&tfs_set_notset), 0x0000000000000001, NULL, HFILL }},
{ &hf_diameter_mip6_feature_vector_local_home_agent_assignment,
{ "LOCAL_HOME_AGENT_ASSIGNMENT", "diameter.mip6_feature_vector.local_home_agent_assignment",
FT_BOOLEAN, 64, TFS(&tfs_set_notset), 0x0000000000000002, NULL, HFILL }},
{ &hf_diameter_mip6_feature_vector_pmip6_supported,
{ "PMIP6_SUPPORTED", "diameter.mip6_feature_vector.pmip6_supported",
FT_BOOLEAN, 64, TFS(&tfs_set_notset), 0x0000010000000000, NULL, HFILL }},
{ &hf_diameter_mip6_feature_vector_ip4_hoa_supported,
{ "IP4_HOA_SUPPORTED", "diameter.mip6_feature_vector.ip4_hoa_supported",
FT_BOOLEAN, 64, TFS(&tfs_set_notset), 0x0000020000000000, NULL, HFILL }},
{ &hf_diameter_mip6_feature_vector_local_mag_routing_supported,
{ "LOCAL_MAG_ROUTING_SUPPORTED", "diameter.mip6_feature_vector.local_mag_routing_supported",
FT_BOOLEAN, 64, TFS(&tfs_set_notset), 0x0000040000000000,NULL, HFILL }},
{ &hf_diameter_3gpp_mip6_feature_vector,
{ "MIP6-Feature-Vector [3GPP]", "diameter.3gpp.mip6_feature_vector",
FT_UINT64, BASE_HEX, NULL, 0x0, NULL, HFILL }},
{ &hf_diameter_3gpp_mip6_feature_vector_assign_local_ip,
{ "MIP6_INTEGRATED", "diameter.3gpp.mip6_feature_vector.assign_local_ip",
FT_BOOLEAN, 64, TFS(&tfs_set_notset), 0x0000080000000000, NULL, HFILL }},
{ &hf_diameter_3gpp_mip6_feature_vector_mip4_supported,
{ "PMIP6_SUPPORTED", "diameter.3gpp.mip6_feature_vector.mip4_supported",
FT_BOOLEAN, 64, TFS(&tfs_set_notset), 0x0000100000000000, NULL, HFILL }},
{ &hf_diameter_3gpp_mip6_feature_vector_optimized_idle_mode_mobility,
{ "OPTIMIZED_IDLE_MODE_MOBILITY", "diameter.3gpp.mip6_feature_vector.optimized_idle_mode_mobility",
FT_BOOLEAN, 64, TFS(&tfs_set_notset), 0x0000200000000000, NULL, HFILL }},
{ &hf_diameter_3gpp_mip6_feature_vector_gtpv2_supported,
{ "GTPv2_SUPPORTED", "diameter.3gpp.mip6_feature_vector.gtpv2_supported",
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
FT_BOOLEAN, 64, TFS(&tfs_set_notset), 0x0000400000000000, NULL, HFILL }},
{ &hf_diameter_user_equipment_info_imeisv,
{ "IMEISV","diameter.user_equipment_info.imeisv", FT_STRING, BASE_NONE, NULL, 0x0, NULL, HFILL }},
{ &hf_diameter_user_equipment_info_mac,
{ "MAC","diameter.user_equipment_info.mac", FT_ETHER, BASE_NONE, NULL, 0x0, NULL, HFILL }},
{ &hf_diameter_user_equipment_info_eui64,
{ "EUI64","diameter.user_equipment_info.eui64", FT_EUI64, BASE_NONE, NULL, 0x0, NULL, HFILL }},
{ &hf_diameter_user_equipment_info_modified_eui64,
DIAMETER: Process BCD encoded IMEISV in User-Equipment-Info Some example captures of DIAMETER show the IMEISV in User-Equipment-Info being BCD encoded (as commonly seen in other protocols.) If the number of octets used is 8, assume it is BCD encoded. If the number of octets used is 16, assume it is ASCII. Otherwise, set an expert info for the wrong length like the other types. Use the defined constants to verify the lengths of the MAC, EUI64, and MODIFIED_EUI64 types. Correct the length of the Modified EUI64 type, as it is also 8 bytes, the same as EUI64 (there must have been confusion with IPv6 addresses, which contain a Modified EUI64 in their 8 least significant bytes.)
2021-03-31 21:58:30 +00:00
{ "Modified EUI64","diameter.user_equipment_info.modified_eui64", FT_EUI64, BASE_NONE, NULL, 0x0, NULL, HFILL }}
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
};
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
gint *ett_base[] = {
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
&ett_diameter,
From David Frascone: update to the DIAMETER dissector for draft-ietf-aaa-diameter-07. His changenotes: Changes: o Flags are now done properly. (I changed it to dissect them like packet-tcp.c does) o Completely re-did the dictionary. I think it is much more maintainable now. And, it is no longer dynamically generated. o Added some lookup routines to work with new dictionary. o Removed lots of old lookup stuff that had been cloned from packet-radius.c. It's much cleaner now. svn path=/trunk/; revision=3799
2001-07-30 20:08:44 +00:00
&ett_diameter_flags,
&ett_diameter_avp_flags,
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
&ett_diameter_avpinfo,
Add Address decoding, two debug environment variables (WIRESHARK_DEBUG_DIAM_DICT_PARSER and WIRESHARK_DUMP_DIAM_DICT) and fix some issues pointed out by Martin svn path=/trunk/; revision=22345
2007-07-17 22:19:54 +00:00
&ett_unknown,
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
&ett_diameter_mip6_feature_vector,
&ett_diameter_3gpp_mip6_feature_vector,
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
&(unknown_avp.ett)
David Frascone's DIAMETER dissector. svn path=/trunk/; revision=2176
2000-07-30 07:16:11 +00:00
};
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Convert remaining proto_item_set_expert_flags in dissector directory. svn path=/trunk/; revision=51142
2013-08-05 01:54:15 +00:00
static ei_register_info ei[] = {
{ &ei_diameter_reserved_bit_set, { "diameter.reserved_bit_set", PI_MALFORMED, PI_WARN, "Reserved bit set", EXPFILL }},
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
{ &ei_diameter_avp_code, { "diameter.avp.code.unknown", PI_UNDECODED, PI_WARN, "Unknown AVP, if you know what this is you can add it to dictionary.xml", EXPFILL }},
{ &ei_diameter_avp_vendor_id, { "diameter.unknown_vendor", PI_UNDECODED, PI_WARN, "Unknown Vendor, if you know whose this is you can add it to dictionary.xml", EXPFILL }},
{ &ei_diameter_avp_no_data, { "diameter.avp.no_data", PI_UNDECODED, PI_WARN, "Data is empty", EXPFILL }},
{ &ei_diameter_avp_pad, { "diameter.avp.pad.non_zero", PI_MALFORMED, PI_NOTE, "Padding is non-zero", EXPFILL }},
Diameter: Add MS-CHAP/MS-CHAPv2 AVP support for EAP-TLS Created the Microsoft Diameter file based on MS-CHAP-* AVPs listed at https://www.iana.org/assignments/eap-numbers/eap-numbers.xhtml#eap-numbers-10 Many values are displayed as bytes for simplicit. The MS-CHAP2-Success attribute could for example be dissected further as 1 byte followed by a string, but that requires more effort. Allow padding to be missing since the eap-ttls-mschapv2.pcapng capture would throw a Malformed Packet exception otherwise. Bug: 15603 Change-Id: I9efc322a86802e78bb6cd4bc3df1c1282a45fe9e Reviewed-on: https://code.wireshark.org/review/34291 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 23:19:33 +00:00
{ &ei_diameter_avp_pad_missing, { "diameter.avp.pad.missing", PI_MALFORMED, PI_NOTE, "Padding is missing", EXPFILL }},
Batch of filterable expert info. svn path=/trunk/; revision=51625
2013-09-01 13:05:27 +00:00
{ &ei_diameter_avp_len, { "diameter.avp.invalid-len", PI_MALFORMED, PI_WARN, "Wrong length", EXPFILL }},
{ &ei_diameter_application_id, { "diameter.applicationId.unknown", PI_UNDECODED, PI_WARN, "Unknown Application Id, if you know what this is you can add it to dictionary.xml", EXPFILL }},
{ &ei_diameter_version, { "diameter.version.unknown", PI_UNDECODED, PI_WARN, "Unknown Diameter Version (decoding as RFC 3588)", EXPFILL }},
{ &ei_diameter_code, { "diameter.cmd.code.unknown", PI_UNDECODED, PI_WARN, "Unknown command, if you know what this is you can add it to dictionary.xml", EXPFILL }},
[Diameter] Improve dissection of malformed packets by continnuing dissection and display the problem more prominetly. Change-Id: Ia1a32667a18e1e5b60b5c167da9b6dd945ba3dfc Reviewed-on: https://code.wireshark.org/review/17385 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-08-30 12:55:54 +00:00
{ &ei_diameter_invalid_ipv6_prefix_len, { "diameter.invalid_ipv6_prefix_len", PI_MALFORMED, PI_ERROR, "Invalid IPv6 Prefix length", EXPFILL }},
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
{ &ei_diameter_invalid_avp_len,{ "diameter.invalid_avp_len", PI_MALFORMED, PI_ERROR, "Invalid AVP length", EXPFILL }},
Diameter: improve IMEI/IMEISV dissection in User-Equipment-Info AVP 3GPP.User-Equipment-Info AVP should contain the IMEISV (if type is 0). However some implementations put the IMEI there: decode it but display a protocol error indication too.
2022-03-19 12:44:56 +00:00
{ &ei_diameter_invalid_user_equipment_info_value_len,{ "diameter.invalid_user_equipment_info_value_len", PI_MALFORMED, PI_ERROR, "Invalid User-Equipment-Info-Value length", EXPFILL }},
{ &ei_diameter_unexpected_imei_as_user_equipment_info,{ "diameter.unexpected_imei_as_user_equipment_info", PI_MALFORMED, PI_ERROR, "Found IMEI as User-Equipment-Info-Value but IMEISV was expected", EXPFILL }},
Convert remaining proto_item_set_expert_flags in dissector directory. svn path=/trunk/; revision=51142
2013-08-05 01:54:15 +00:00
};
Build diameter's hf array in a wmem array instead of a garray. Fixes ~270KB of leak when running epan_cleanup(). svn path=/trunk/; revision=50405
2013-07-06 12:45:33 +00:00
wmem_array_append(build_dict.hf, hf_base, array_length(hf_base));
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
ett_length = array_length(ett_base);
Fix pretty-printing of the application ID in the protocol-details pane by putting an "unused" variable (which was taken out in r24377) back in. This time add a (bogus) "use" of the variable and explain why we're doing what we're doing. Use VALS() (in one case anyway) and PINFO_FD_NUM. svn path=/trunk/; revision=46266
2012-11-28 22:45:31 +00:00
for (i = 0; i < ett_length; i++) {
The second argument to g_ptr_array_add() is the pointer to add, not a pointer to that pointer - unlike g_array_append_vals, which takes a pointer to an array of values and a count of the number of values in the array. svn path=/trunk/; revision=25753
2008-07-16 08:08:17 +00:00
g_ptr_array_add(build_dict.ett, ett_base[i]);
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
}
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
Build diameter's hf array in a wmem array instead of a garray. Fixes ~270KB of leak when running epan_cleanup(). svn path=/trunk/; revision=50405
2013-07-06 12:45:33 +00:00
proto_register_field_array(proto_diameter, (hf_register_info *)wmem_array_get_raw(build_dict.hf), wmem_array_get_count(build_dict.hf));
Handle hop-by-hop-id collisions (when matching requests and answers) better: instead of storing the requests in an se_tree (keyed by hop-by-hop-id), store an se_tree which is itself keyed by frame number. When looking for a matching request, first find the tree of pdus (by hop-by-hop-id) and then look for the largest frame number less than or equal to the answer's frame number. Also verify that the end-to-end-id matches. Move the conversation structure out of packet-diameter.h: it's only used in packet-diameter.c svn path=/trunk/; revision=46553
2012-12-16 01:12:59 +00:00
proto_register_subtree_array((gint **)build_dict.ett->pdata, build_dict.ett->len);
Convert remaining proto_item_set_expert_flags in dissector directory. svn path=/trunk/; revision=51142
2013-08-05 01:54:15 +00:00
expert_diameter = expert_register_protocol(proto_diameter);
expert_register_field_array(expert_diameter, ei, array_length(ei));
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
g_ptr_array_free(build_dict.ett,TRUE);
Have "proto_register_protocol()" build a list of data structures for protocols, in addition to adding structures to the list of filterable fields. Give it an extra argument that specifies a "short name" for the protocol, for use in such places as pinfo->current_proto; the dialog box for constructing filters; the preferences tab for the protocol; and so on (although we're not yet using it in all those places). Make the preference name that appears in the preferences file and the command line for the DIAMETER protocol "diameter", not "Diameter"; the convention is that the name in question be all-lower-case. Make some routines and variables that aren't exported static. Update a comment in the ICP dissector to make it clear that the dissector won't see fragments other than the first fragment of a fragmented datagram. svn path=/trunk/; revision=2810
2001-01-03 06:56:03 +00:00
Delay registration of Diameter fields until they're needed. ... Like the RADIUS and wimaxasncp dissectors do. Change-Id: Ifab019a0040d3938e52918a282a3beba9dfcfc70 Reviewed-on: https://code.wireshark.org/review/14900 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-04-13 03:01:18 +00:00
}
static void
register_diameter_fields(const char *unused _U_)
{
/*
* The hf_base[] array for Diameter refers to a variable
* that is set by dictionary_load(), so we need to call
* dictionary_load() before hf_base[] is initialized.
*
* To ensure that, we call dictionary_load() and then
* call a routine that defines hf_base[] and does all
* the registration work.
*/
dictionary_load();
real_register_diameter_fields();
}
void
proto_register_diameter(void)
{
module_t *diameter_module;
WSUG: Update the Service Response Time section. Update the Service Response Time section of the User's Guide. The sample captures page had a nice SMB2 capture file, so use that for the example. Use title case for "Diameter" to match the RFCs. Change-Id: Icff510dbe2d77db9cd42548ad58439d17282e851 Reviewed-on: https://code.wireshark.org/review/36421 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2020-03-14 22:17:39 +00:00
proto_diameter = proto_register_protocol ("Diameter Protocol", "Diameter", "diameter");
Delay registration of Diameter fields until they're needed. ... Like the RADIUS and wimaxasncp dissectors do. Change-Id: Ifab019a0040d3938e52918a282a3beba9dfcfc70 Reviewed-on: https://code.wireshark.org/review/14900 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-04-13 03:01:18 +00:00
From Martin Mathieson: - Many DCT2000 protocols can be embedded within an IP primitive message. Add a heuristic to see if we can find the protocol payload within in IP primitive message, and look for an ethereal dissector matching the DCT2000 protocol name (this is useful for simple protocol testing where no physical links are involved) - Make some more of these protocols (diameter, http, mgcp) findable by name - Adds protocol 'variant' number to stub and dissector - Break the duplicated writing of the stub header out into a separate function svn path=/trunk/; revision=18212
2006-05-23 05:48:00 +00:00
/* Allow dissector to find be found by name. */
Dissectors don't need a journey of self discovery. They already know who they are when they register themselves. Saving the handle then to avoid finding it later. Not sure if this will increase unnecessary register_dissector functions (instead of using create_dissector_handle in proto_reg_handoff function) when other dissectors copy/paste, but it should make startup time a few microseconds better. Change-Id: I3839be791b32b84887ac51a6a65fb5733e9f1f43 Reviewed-on: https://code.wireshark.org/review/19481 Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-12-17 01:06:11 +00:00
diameter_sctp_handle = register_dissector("diameter", dissect_diameter, proto_diameter);
Use `register_dissector()` for more protocols
2022-09-29 11:21:10 +00:00
diameter_udp_handle = create_dissector_handle(dissect_diameter, proto_diameter);
diameter: Register the [D]TLS port Port 5868 is IANA registered for Diameter over TLS/TCP and DTLS/SCTP. Register the diameter TCP handle by name so that it can be registered to tls.
2022-07-02 01:03:05 +00:00
diameter_tcp_handle = register_dissector("diameter.tcp", dissect_diameter_tcp, proto_diameter);
EAP, Diameter: add EAP-TTLS and EAP-Message dissection support Add support for dissecting the decrypted TLS payload as Diameter. Add support for dissecting the EAP-Message attribute as EAP. Disable retransmission detection when EAP-Message is detected (EAP in TLS in EAP) since this results in false positives. Tested with captures from Bug 15603: * eap-ttls-pap.pcapng - ok, User-Name and User-Password AVPs. * eap-ttls-eap-gtc.pcapng, eap-ttls-eap-md5.pcapng - EAP-Message AVP. * eap-ttls-mschapv2.pcapng - partially supported, does not conform to Diameter AVP requirements as it is not padded. Microsoft vendor types are also not yet supported. To be fixed later. * eapttls-diameter-avp.pcapng (Bug 12880) - EAP-Message AVP. Bug: 12880 Bug: 15603 Change-Id: Ie7ea282d05c1d3ff8463c34bf259107562714440 Reviewed-on: https://code.wireshark.org/review/34281 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 01:47:09 +00:00
/* Diameter AVPs without Diameter header, for EAP-TTLS (RFC 5281, Section 10) */
register_dissector("diameter_avps", dissect_diameter_avps, proto_diameter);
Mark a local function static rather than extern. Get rid of an unused variable. Was going to use proto_register_prefix(), but am not sure if its appropriate. Even with tree == NULL, dictionary items are used to generate info column. Loading the dictionary lazily would mean doing a lot of work as part of dissecting the first diameter frame, which could decrease our chances of capturing frames that arrive very soon after that one... svn path=/trunk/; revision=24784
2008-04-05 05:41:59 +00:00
Delay registration of Diameter fields until they're needed. ... Like the RADIUS and wimaxasncp dissectors do. Change-Id: Ifab019a0040d3938e52918a282a3beba9dfcfc70 Reviewed-on: https://code.wireshark.org/review/14900 Petri-Dish: Jeff Morriss <jeff.morriss.ws@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2016-04-13 03:01:18 +00:00
/* Delay registration of Diameter fields */
proto_register_prefix("diameter", register_diameter_fields);
Include padding length in the AVP's length. svn path=/trunk/; revision=48702
2013-04-02 18:15:42 +00:00
/* Register dissector table(s) to do sub dissection of AVPs (OctetStrings) */
Fix up dissector tables' UI names. This was inspired by using the Decode-As UI to decode Field "SSL TCP Dissector" Value (port) XXX as YYY. "SSL Port" makes more sense as the UI name. Change-Id: Id6398a5dc79e32bddc4f1bfcf0a468ae1364808f Reviewed-on: https://code.wireshark.org/review/19573 Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2017-01-07 02:49:50 +00:00
diameter_dissector_table = register_dissector_table("diameter.base", "Diameter Base AVP", proto_diameter, FT_UINT32, BASE_DEC);
diameter_3gpp_avp_dissector_table = register_dissector_table("diameter.3gpp", "Diameter 3GPP AVP", proto_diameter, FT_UINT32, BASE_DEC);
diameter_ericsson_avp_dissector_table = register_dissector_table("diameter.ericsson", "Diameter Ericsson AVP", proto_diameter, FT_UINT32, BASE_DEC);
[Diameter] Add a table for subdissectiong AVPs for VENDOR_VERIZON Change-Id: I37397e590f294d209b89fe868b365cafb5343464 Reviewed-on: https://code.wireshark.org/review/21363 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2017-04-27 11:03:42 +00:00
diameter_verizon_avp_dissector_table = register_dissector_table("diameter.verizon", "DIAMETER_VERIZON_AVPS", proto_diameter, FT_UINT32, BASE_DEC);
From Martin Mathieson: - Many DCT2000 protocols can be embedded within an IP primitive message. Add a heuristic to see if we can find the protocol payload within in IP primitive message, and look for an ethereal dissector matching the DCT2000 protocol name (this is useful for simple protocol testing where no physical links are involved) - Make some more of these protocols (diameter, http, mgcp) findable by name - Adds protocol 'variant' number to stub and dissector - Break the duplicated writing of the stub header out into a separate function svn path=/trunk/; revision=18212
2006-05-23 05:48:00 +00:00
Fix up dissector tables' UI names. This was inspired by using the Decode-As UI to decode Field "SSL TCP Dissector" Value (port) XXX as YYY. "SSL Port" makes more sense as the UI name. Change-Id: Id6398a5dc79e32bddc4f1bfcf0a468ae1364808f Reviewed-on: https://code.wireshark.org/review/19573 Petri-Dish: Michael Mann <mmann78@netscape.net> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2017-01-07 02:49:50 +00:00
diameter_expr_result_vnd_table = register_dissector_table("diameter.vnd_exp_res", "Diameter Experimental-Result-Code", proto_diameter, FT_UINT32, BASE_DEC);
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* Register configuration options */
diameter_module = prefs_register_protocol(proto_diameter, NULL);
Add a mechanism to register aliases for preference modules. That way, we don't need a chunk of code in epan/prefs.c that knows about various preference module mappings; individual dissectors can register aliases as appropriate. (The Nortel Discovery Protocol never *had* any preferences, even when it was the SynOptics Network Management Protocol, so there's no need for it to register an alias.) Change-Id: I4a718dac6bb06801cc06a6ee5a28d7ed81e67e5d Reviewed-on: https://code.wireshark.org/review/29914 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-09-29 03:23:39 +00:00
/* For reading older preference files with "Diameter." preferences */
prefs_register_module_alias("Diameter", diameter_module);
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
Support for desegmentation of Diameter-over-TCP, from David Frascone. svn path=/trunk/; revision=4141
2001-11-03 12:19:56 +00:00
/* Desegmentation */
Fix some preferences to eliminate the extra copy of the protocol name at the beginning, and to use underscores rather than periods where the preference's name really isn't part of a hierarchical namespace. Use "%u" rather than "%d" to print unsigned quantities. svn path=/trunk/; revision=4543
2002-01-15 10:01:21 +00:00
prefs_register_bool_preference(diameter_module, "desegment",
Check preference titles and descriptions. When registering preferences, make sure our titles and descriptions are valid UTF-8. Make sure our titles are short and only contain printable characters. Fix problematic titles and descriptions. Change-Id: I20d3f93438f2b3c30266f934297feb79897f2ee5 Reviewed-on: https://code.wireshark.org/review/18998 Reviewed-by: Gerald Combs <gerald@wireshark.org> Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-09-20 21:11:29 +00:00
"Reassemble Diameter messages spanning multiple TCP segments",
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
"Whether the Diameter dissector should reassemble messages spanning multiple TCP segments."
" To use this option, you must also enable \"Allow subdissectors to reassemble TCP streams\" in the TCP protocol settings.",
&gbl_diameter_desegment);
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
GArrays' data is properly aligned for whatever they are holding (despite the fact that 'data' is a pointer to a char). GLIB fixed this in their accessor macro here: http://svn.gnome.org/viewvc/glib?view=revision&revision=6092 The bug report confirms that the data is properly aligned: http://bugzilla.gnome.org/show_bug.cgi?id=502927 So, add some intermediate (void *) casts to avoid "cast increases required alignment of target type" on SPARC. (No, we can't use the accessor macro because we want to access the whole array not just the ith element.) Since build_dict->ett is actually an array of pointers, change it to a GPtrArray. Reindent some. svn path=/trunk/; revision=25749
2008-07-15 21:06:31 +00:00
/* Register some preferences we no longer support, so we can report
* them as obsolete rather than just illegal.
*/
Add port ranges. svn path=/trunk/; revision=23265
2007-10-24 15:04:01 +00:00
prefs_register_obsolete_preference(diameter_module, "version");
Allow a dissector to register preferences that it no longer supports as obsolete; we silently ignore attempts to set those in a preferences file, so that we don't spam the user with error messages caused by them having saved preferences in an earlier release that contained those preferences. Make the Diameter and iSCSI dissectors register obsolete preferences. Crash if some code tries to register a preferences module with a name that contains something other than lower-case ASCII letters, numbers, or underscores, or that has already been registered, or if some code tries to register a preference with a name that contains something other than lower-case ASCII letters, numbers, underscores, or periods, or that has already been registered, so that we don't put code like that in a release and have to shovel code into "prefs.c" to fix it up later. (The problem with multiple modules or preferences with the same name should be obvious; the problem with names with blanks, or upper-case letters, or punctuation, is that they're a pain to type on the command line.) svn path=/trunk/; revision=4148
2001-11-04 02:50:21 +00:00
prefs_register_obsolete_preference(diameter_module, "command_in_header");
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
prefs_register_obsolete_preference(diameter_module, "dictionary.name");
prefs_register_obsolete_preference(diameter_module, "dictionary.use");
prefs_register_obsolete_preference(diameter_module, "allow_zero_as_app_id");
prefs_register_obsolete_preference(diameter_module, "suppress_console_output");
If we can't find dictionary.xml, fail instead of trying to read from stdin (which makes distcheck unhappy). Fixup whitespace. svn path=/trunk/; revision=22338
2007-07-17 21:29:18 +00:00
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
/* Register tap */
diameter_tap = register_tap("diameter");
Further refactor SRT stats. Create "common" SRT tap data collection intended for all GUIs. Refactor/merge functionality of existing dissectors that have SRT support (AFP, DCERPC, Diameter, FC, GTP, LDAP, NCP, RPC, SCIS, SMB, and SMB2) for both TShark and GTK. SMB and DCERPC "tap packet filtering" were different between TShark and GTK, so I went with GTK filter logic. CAMEL "tap packet filtering" was different between TShark and GTK, so GTK filtering logic was pushed to the dissector and the TShark tap was left alone. Change-Id: I7d6eaad0673fe628ef337f9165d7ed94f4a5e1cc Reviewed-on: https://code.wireshark.org/review/8894 Petri-Dish: Michael Mann <mmann78@netscape.net> Reviewed-by: Gerald Combs <gerald@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2015-06-16 21:07:04 +00:00
register_srt_table(proto_diameter, NULL, 1, diameterstat_packet, diameterstat_init, NULL);
From Abhik Sarkar: Support for Diameter conversations and service response time. svn path=/trunk/; revision=26785
2008-11-15 15:52:39 +00:00
From David Frascone: duUpdate to Diameter dissector to load the dictionary as an XML file rather than building it in, and various Diameter updates. svn path=/trunk/; revision=4122
2001-11-01 21:52:44 +00:00
} /* proto_register_diameter */
Rewrite of the diameter dissector to use the dictionary for creating hfids, drop libxml dependency. The work is still incomplete (anything but strings and numbers appears as bytes) but I want others to start testing it. TODO: builders and decoders for: - (ntp) timestamps - addresses - diameteruris - diameteridentities - ipfilterrules - qosfilterrules - mipregistrationrequests svn path=/trunk/; revision=22318
2007-07-16 05:41:58 +00:00
Move proto_reg_handoff... to the end of the file. Whitespace cleanup: remove trailing spaces from a few lines. svn path=/trunk/; revision=32569
2010-04-26 22:19:34 +00:00
void
proto_reg_handoff_diameter(void)
{
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
data_handle = find_dissector("data");
eap_handle = find_dissector_add_dependency("eap", proto_diameter);
Register for Diameter's SCTP PPID. svn path=/trunk/; revision=44501
2012-08-14 20:44:58 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
dissector_add_uint("sctp.ppi", DIAMETER_PROTOCOL_ID, diameter_sctp_handle);
Diameter: Add a heuristic dissector over TCP Add a heuristic dissector for Diameter over TCP, disabled by default. We shouldn't need one for Diameter over SCTP, the PPID should take care of it.
2022-07-03 16:36:31 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
heur_dissector_add("tcp", dissect_diameter_tcp_heur, "Diameter over TCP", "diameter_tcp", proto_diameter, HEURISTIC_DISABLE);
diameter: Register the [D]TLS port Port 5868 is IANA registered for Diameter over TLS/TCP and DTLS/SCTP. Register the diameter TCP handle by name so that it can be registered to tls.
2022-07-02 01:03:05 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
ssl_dissector_add(DEFAULT_DIAMETER_TLS_PORT, diameter_tcp_handle);
dtls_dissector_add(DEFAULT_DIAMETER_TLS_PORT, diameter_sctp_handle);
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* Register special decoding for some AVPs */
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 1 User-Name */
dissector_add_uint("diameter.base", 1, create_dissector_handle(dissect_diameter_user_name, proto_diameter));
EAP, Diameter: add EAP-TTLS and EAP-Message dissection support Add support for dissecting the decrypted TLS payload as Diameter. Add support for dissecting the EAP-Message attribute as EAP. Disable retransmission detection when EAP-Message is detected (EAP in TLS in EAP) since this results in false positives. Tested with captures from Bug 15603: * eap-ttls-pap.pcapng - ok, User-Name and User-Password AVPs. * eap-ttls-eap-gtc.pcapng, eap-ttls-eap-md5.pcapng - EAP-Message AVP. * eap-ttls-mschapv2.pcapng - partially supported, does not conform to Diameter AVP requirements as it is not padded. Microsoft vendor types are also not yet supported. To be fixed later. * eapttls-diameter-avp.pcapng (Bug 12880) - EAP-Message AVP. Bug: 12880 Bug: 15603 Change-Id: Ie7ea282d05c1d3ff8463c34bf259107562714440 Reviewed-on: https://code.wireshark.org/review/34281 Petri-Dish: Peter Wu <peter@lekensteyn.nl> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-08-14 01:47:09 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 79 EAP-Message (defined in RFC 2869, but used for EAP-TTLS, RFC 5281) */
dissector_add_uint("diameter.base", 79, create_dissector_handle(dissect_diameter_eap_payload, proto_diameter));
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 97 Framed-IPv6-Address */
dissector_add_uint("diameter.base", 97, create_dissector_handle(dissect_diameter_base_framed_ipv6_prefix, proto_diameter));
DIAMETER: Add MIP6-Feature-Vector flag decode Change-Id: I1b4e4567dcafef88fcb5d2157899dfb88a8e2640 Reviewed-on: https://code.wireshark.org/review/33657 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-05-30 18:26:24 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 124 MIP6-Feature-Vector */
dissector_add_uint("diameter.base", 124, create_dissector_handle(dissect_diameter_mip6_feature_vector, proto_diameter));
[Diamater] Sub dissect Suported-Vendor-Id Change-Id: I5c0014449ef9886ba74abdf8f609860ef37a7831 Reviewed-on: https://code.wireshark.org/review/8582 Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-05-22 10:29:53 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 265 Supported-Vendor-Id */
dissector_add_uint("diameter.base", 265, create_dissector_handle(dissect_diameter_vendor_id, proto_diameter));
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 266 Vendor-Id */
dissector_add_uint("diameter.base", 266, create_dissector_handle(dissect_diameter_vendor_id, proto_diameter));
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 443 Subscription-Id */
dissector_add_uint("diameter.base", 443, create_dissector_handle(dissect_diameter_subscription_id, proto_diameter));
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 450 Subscription-Id-Type */
dissector_add_uint("diameter.base", 450, create_dissector_handle(dissect_diameter_subscription_id_type, proto_diameter));
Decode IMSIs found in Subscription-Id AVPs. Move decode of the User-Name AVP out of the 3GPP-specific file. Add a couple of macros for 3GPP AppIDs (rather than using the number directly) in both files. Change-Id: I496b4ae86b05264462167b6e41ce1451392de11b Reviewed-on: https://code.wireshark.org/review/3753 Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>
2014-08-20 19:08:21 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 444 Subscription-Id-Data */
dissector_add_uint("diameter.base", 444, create_dissector_handle(dissect_diameter_subscription_id_data, proto_diameter));
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 458 User-Equipment-Info */
dissector_add_uint("diameter.base", 458, create_dissector_handle(dissect_diameter_user_equipment_info, proto_diameter));
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 459 User-Equipment-Info-Type */
dissector_add_uint("diameter.base", 459, create_dissector_handle(dissect_diameter_user_equipment_info_type, proto_diameter));
DIAMETER: Add User-Equipment-Info dissector Change-Id: I5a1eb1a431cad6ca07ef0db00fbe1c3192709d58 Reviewed-on: https://code.wireshark.org/review/34922 Petri-Dish: Pascal Quantin <pascal@wireshark.org> Tested-by: Petri Dish Buildbot Reviewed-by: Pascal Quantin <pascal@wireshark.org>
2019-11-03 11:12:19 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 460 User-Equipment-Info-Value */
dissector_add_uint("diameter.base", 460, create_dissector_handle(dissect_diameter_user_equipment_info_value, proto_diameter));
Move proto_reg_handoff... to the end of the file. Whitespace cleanup: remove trailing spaces from a few lines. svn path=/trunk/; revision=32569
2010-04-26 22:19:34 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* AVP Code: 462 EAP-Payload */
dissector_add_uint("diameter.base", 462, create_dissector_handle(dissect_diameter_eap_payload, proto_diameter));
/* AVP Code: 463 EAP-Reissued-Payload */
dissector_add_uint("diameter.base", 463, create_dissector_handle(dissect_diameter_eap_payload, proto_diameter));
"RFC 6733 [5] specifies the Experimental-Result AVP containing Vendor-Id AVP and Experimental-Result-Code AVP. The Experimental-Result-Code AVP (AVP Code 298) is of type Unsigned32 and contains a vendor-assigned value representing the result of processing a request." This means we have to store the value of the Vendor Id AVP in the grouped Experimental-Result AVP to be able to dissect the value of Experimental-Result-Code according to the vendor specified values(if known). This patch implements that with a dissector table to register vendor dissectors for the result code. svn path=/trunk/; revision=53785
2013-12-04 20:55:52 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
/* Register dissector for Experimental result code, with 3GPP2's vendor Id */
dissector_add_uint("diameter.vnd_exp_res", VENDOR_THE3GPP2, create_dissector_handle(dissect_diameter_3gpp2_exp_res, proto_diameter));
Move proto_reg_handoff... to the end of the file. Whitespace cleanup: remove trailing spaces from a few lines. svn path=/trunk/; revision=32569
2010-04-26 22:19:34 +00:00
prefs: More port prefs to auto prefs with ranges Move the rest of the SCTP port preferences to auto prefs with ranges. Ping #14319.
2022-08-06 20:03:33 +00:00
dissector_add_uint_range_with_preference("tcp.port", DEFAULT_DIAMETER_PORT_RANGE, diameter_tcp_handle);
dissector_add_uint_range_with_preference("udp.port", "", diameter_udp_handle);
dissector_add_uint_range_with_preference("sctp.port", DEFAULT_DIAMETER_PORT_RANGE, diameter_sctp_handle);
Add export_pdu_tap to packet-diameter.c svn path=/trunk/; revision=49448
2013-05-20 18:30:32 +00:00
exported_pdu_tap = find_tap_id(EXPORT_PDU_TAP_NAME_LAYER_7);
Move proto_reg_handoff... to the end of the file. Whitespace cleanup: remove trailing spaces from a few lines. svn path=/trunk/; revision=32569
2010-04-26 22:19:34 +00:00
}
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
/*
HTTPS (almost) everywhere. Change all wireshark.org URLs to use https. Fix some broken links while we're at it. Change-Id: I161bf8eeca43b8027605acea666032da86f5ea1c Reviewed-on: https://code.wireshark.org/review/34089 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-07-26 18:43:17 +00:00
* Editor modelines - https://www.wireshark.org/tools/modelines.html
Do some constification; Create/use value_string_ext_free(); Display certain numbers also as hex in the "forced to fall back to linear search: ..." value-string warning msg Add editor-modelines to some files; Do some whitespace changes. svn path=/trunk/; revision=53484
2013-11-21 18:46:40 +00:00
*
* Local variables:
* c-basic-offset: 8
* tab-width: 8
* indent-tabs-mode: t
* End:
*
* vi: set shiftwidth=8 tabstop=8 noexpandtab:
* :indentSize=8:tabSize=8:noTabs=false:
*/