Check length of Slow Protocol OAM Vendor Specific Event Notification and add an expert info if it is not at least 2 bytes long
svn path=/trunk/; revision=46336
where the edt is on the stack, in which case epan_dissect_free never gets
called and we were leaking the pool (although not any of the memory in it).
svn path=/trunk/; revision=46333
are like the non-TVB versions except that they take a TVB and an offset
instead of (frequently) a pointer into the TVB.
Calling tvb_get_ptr() before modifying the rest of the fields should help fix
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=7960 (though I can't
reproduce that problem).
Replace a bunch of calls like:
SET_ADDRESS(..., AT_XXX, length, tvb_get_ptr(tvb, offset, length));
with:
TVB_SET_ADDRESS(..., AT_XXX, tvb, offset, length);
svn path=/trunk/; revision=46324
ABORT_ON_DISSECTOR_BUG is set and we pass MAX_TREE_ITEMS.
If ABORT_ON_DISSECTOR_BUG is set and we get an unregistered hf, generate
an explanatory message (and a core).
svn path=/trunk/; revision=46323
Note the change in behavior: building the tree is now off by default.
Complain (and exit) if we get an argument we don't understand.
svn path=/trunk/; revision=46321
Don't store an AT_STRINGZ address in ep_ allocated memory: that memory is
freed before the addresses may be used. Use se_ memory instead (no, that's
not really ideal either).
It would appear that several other dissectors have the same problem.
svn path=/trunk/; revision=46320
https://www.wireshark.org/download/automated/captures/fuzz-2012-11-30-20336.pcap
(I can't reproduce the actual fuzz failure and this uninitialized-read
shouldn't have caused an out-of-memory condition--which is what the buildbot
said the problem with that capture was.)
Also stop marking a variable that is used as unused.
svn path=/trunk/; revision=46309
PATCH] USB: Add support for vendor_id/product_id,
bus_id/device_address and class/subclass/protocol tables
USB subdissectors can be decoded by class/subclass/protocol. If fail try vendor_id/product_id.
bus_id/device_address is intented to use by DecodeBy, so user can decode specified device as need.
Part of:
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=5032
svn path=/trunk/; revision=46306
Couple of coding style changes and fixes for:
warning: declaration of ‘k_frame_number’ shadows a previous local
warning: assignment discards qualifiers from pointer target type
Part of:
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=5032
svn path=/trunk/; revision=46304
The gsm rlcmac dissector truncates the 12 bit BSNs to 8 bits when storing them.
Fix this and the handling of BSN2 is refactored to clarify the difference between BSN2 (the stored value) and BSN2_offset (the value transferred at the air interface).
svn path=/trunk/; revision=46301
USB: Add support for vendor_id/product_id,
bus_id/device_address and class/subclass/protocol
tables
USB subdissectors can be decoded by class/subclass/protocol. If fail try vendor_id/product_id.
bus_id/device_address is intented to use by DecodeBy, so user can decode specified device as need.
Part of:
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=5032
svn path=/trunk/; revision=46283
Pascal Quantin