Have the makefiles do all the work setting up proto_register_<dissector> and proto_reg_handoff_<dissector>, rather than sprinkled throughout the dissectors with extern declarations.
More is coming to make more functions static (because of current makefile fixes) and cleanup MAC Management dissectors to remove unnecessary/duplicative code, but I wanted to get the build issues squared away.
svn path=/trunk/; revision=52167
explicit, and frees up the "generic" names (like tvb_memdup) for new signatures
that take the appropriate wmem pool.
Majority of the conversion done with sed.
svn path=/trunk/; revision=52164
1. Correct Interface Flag enumeration
2. Dissect ARP data without making it look like its an ARP packet by disabling column writing.
svn path=/trunk/; revision=52157
Decode the mesh formation information fields related to the number of mesh peerings
Signed-off-by: Chun-Yeow Yeoh <yeohchunyeow@cozybit.com>
svn path=/trunk/; revision=52152
Really add support for AEAD ciphers (GCM)
GCM uses counter mode with authentication tags (the latter is currently
not supported). As for the key material, there is no MAC (because the
auth tag is supposed to verify the authenticity).
Finally, correct the GCM cipher suite definitions: IV block size of
4 bytes and GCM instead of CBC mode.
svn path=/trunk/; revision=52150
Use IV from record for CBC mode, add padding/IV length check
Add summary of RFCs to make it more obvious why certain parts (IV, MAC,
padding) are used. Merge DTLS and TLS blocks for extracting IV. This
saves an unnecessary memmove() because the input pointer is, well, just
a local variable and can therefore be incremented.
Validate padding and IV lengths before using it. A crash could occur
if the explicit IV is missing (this would make memmove write before its
buffer). The missing padding check had as implication that a misleading
error is returning with a negative length (not exploitable).
Use IV from record for CBC mode, previously it decrypted the first block
incorrectly and then threw this "decrypted" IV away. Now it extracts the
IV and uses this for decrypting the first fragment block. (remember that
CBC xor's the output of the block cipher with the previous ciphertext
(or IV for the first block)).
This is a preparation for GCM which does not have a MAC. The skip_mac
branch is necessary to make the compiler happy in this patch, 'mac'
could otherwise be uninitialised.
svn path=/trunk/; revision=52149
Correct cipher suites list, add TLS_ annotations
Add official TLS_ names as comment and correct:
- 6: RC2 is a block cipher using MD5, not stream+SHA.
- 25,26,27: should be SHA instead of MD5
- 98: DES export is a block cipher using 56-bits[1], not stream.
- 99: DES export should be using 56-bits[1].
- 138: removed commented RC4 cipher because it is not a block cipher
Besides these comments and corrections, there are no further changes.
[1]: http://tools.ietf.org/html/draft-ietf-tls-56-bit-ciphersuites-00
svn path=/trunk/; revision=52148
Drop export_cipher and dig_len, cleaner digest access
Removed dig_len as this magic number is dependent on dig. The digests
variable is converted from a string to a structure holding the digest
name and length because of its close dependency.
Introduce another struct+function to get rid of the magic number 0x40
(DIG_MD5).
Removed export_cipher bit as this is dependent on eff_bits < bits.
Verified with:
grep ,KEX_ packet-ssl-utils.c | awk -F, '{bits=$6!=$7;ex=$9;
if ((bits && !ex) || (!bits && ex))print $6, $7, $8, "###", $0}'.
Removed space before SIG_RSA for cipher 51 for consistency with others.
svn path=/trunk/; revision=52147
the tvb_memcpy on the next line should be taking more data, but I don't know
enough about the protocol to be sure. This is the least disruptive way to fix
the last valgrind error from
https://bugs.wireshark.org/bugzilla/show_bug.cgi?id=8941
svn path=/trunk/; revision=52145
the "unittest" module that comes with Python. Specifically, this
takes advantage of a couple of features in the "unittest" in
Python 2.7. The tests are all the same as before, but much
better managed.
This is in preparation for some work on the display filter code.
svn path=/trunk/; revision=52136
This should hopefully fix linking with Solaris Studio, which doesn't
remove unused inline functions by default. Fix a copy+paste error.
svn path=/trunk/; revision=52122
be incorrect on the first pass and only be correct when remade.
The reason was that disp-exp.cnf was created incorrectly on first pass.
disp-exp.cnf required dop-exp.cnf to build correctly, so add a mechanism
to add dependencies for export files.
Note: Only the cmake change has been tested, the autofoo stuff requires
testing.
svn path=/trunk/; revision=52120
