The attached patches fix the following issues:
1. CIP-Motion shouldn't be a heuristic dissector. CIP-Motion packets are
determined by specific fields in a CIP ForwardOpen. I ran into situations
where CIP-Motion's "vague" heuristics incorrectly dissected packets meant for
CIP-Safety. Of course this requires the ForwardOpen to be in the trace, but it
is the most deterministic way to have CIP-Motion packets.
2. Minor bugfix to the CIP-Safety dissector. A couple of fields were
"missing", causing errant malformed packets.
svn path=/trunk/; revision=41283
Modbus object support for CIP dissector
Support for the Modbus object within the CIP dissector. Essentially wraps the Modbus protocol within a CIP object.
Also included update to Modbus error code per
http://www.wireshark.org/lists/wireshark-dev/201111/msg00187.html
svn path=/trunk/; revision=40371
Don't set mr_mult_req_info->num_services until we're sure we've
initialized the full structure for that service. Otherwise if we
happen to throw an exception before initializing the whole
structure, we'll core someplace (like dissect_cip_generic_service_rsp())
which expects all num_services entries to be fully initialized.
svn path=/trunk/; revision=39626
cc1: warnings being treated as errors
packet-cip.c: In function 'dissect_cip_generic_service_req':
packet-cip.c:3281: warning: format not a string literal and no format arguments
packet-cip.c: In function 'dissect_cip_generic_service_rsp':
packet-cip.c:3631: warning: format not a string literal and no format arguments
packet-cip.c: In function 'dissect_cip_data':
packet-cip.c:4881: warning: comparison between signed and unsigned
svn path=/trunk/; revision=39599
1. If there's no character encoding (ENC_ASCII, ...) specified
then use ENC_ASCII.
2. For all but FT_UINT_STRING, always use ENC_NA
(replacing any existing True/1/FALSE/0
/ENC_BIG_ENDIAN/ENC_LITTLE_ENDIAN).
svn path=/trunk/; revision=39426
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=5929
From me:
packet-cipmotion.c:
FT_BOOLEAN fields with bitmasks need a bit-fieldwidth in the hf[] entry 'display' field;
Define attribute_size as guint32 since it has to store guint8*guint16;
Use ENC_NA as encoding arg in proto_tree_add_item() for FT_BYTES field types;
Remove trailing whitespace from lines;
Other minor cleanup and reformatting.
packet-enip.c:
Use ENC_NA as encoding arg in proto_tree_add_item() for FT_BYTES field types;
svn path=/trunk/; revision=39396
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
FT_NONE
FT_BYTES
FT_IPV6
FT_IPXNET
FT_OID
Note: Encoding field set to ENC_NA only if the field was previously TRUE|FALSE|ENC_LITTLE_ENDIAN|ENC_BIG_ENDIAN
svn path=/trunk/; revision=39260
Added support for the following services of the Connection Manager object:
LargeForwardOpen
GetConnectionOwner
Added support for the following services of the Connection Configuration
Object:
GetAttributeAll
SetAttributeAll
From me:
Replace tabs with spaces; adjust modelines for 8-space tabs instead of 3.
svn path=/trunk/; revision=37404
keys to have _uint in their names, to match the routines that handle
dissector tables with string keys. (Using _port can confuse people into
thinking they're intended solely for use with TCP/UDP/etc. ports when,
in fact, they work better for things such as Ethernet types, where the
binding of particular values to particular protocols are a lot
stronger.)
svn path=/trunk/; revision=35224
Bill Meier