Build 1.10.11.

Change-Id: I4717cc7864c1dce277eb16469647f5ec389a05e8
Reviewed-on: https://code.wireshark.org/review/5184
Reviewed-by: Gerald Combs <gerald@wireshark.org>

Fix a problem that causes build failures with CMake on Yosemite.

Change-Id: I52e81656470df887eb3c53b2632bf5a310b1b063
Reviewed-on: https://code.wireshark.org/review/5229
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit b2c2debd978518c5c39781567a0e00f6713cc94e)
Reviewed-on: https://code.wireshark.org/review/5231

[Automatic manuf, services and enterprise-numbers update for 2014-11-09]

Change-Id: I7c1008e62be0dba9a814a39d574707362b232c3b
Reviewed-on: https://code.wireshark.org/review/5208
Reviewed-by: Gerald Combs <gerald@wireshark.org>

simple fix for bug 10676

Bug: 10676
Change-Id: I442aa2ce44309655190752e715279d7325647336
Reviewed-on: https://code.wireshark.org/review/5199
Reviewed-by: Martin Kaiser <wireshark@kaiser.cx>
Tested-by: Martin Kaiser <wireshark@kaiser.cx>
(cherry picked from commit ee1c74c3487812b3b361c66dfa7c2055e4df1588)
Reviewed-on: https://code.wireshark.org/review/5202

amqp: Fix for crash reported in Bug 10582.

  Detect and handle overflows when advancing 'offset'.

  (This patch converts the previous temporary fix to be the
   permanent fix for Wireshark-1.10).

Ping Bug 10582
Change-Id: I6ed10c325d851ad9fa5e13e1852fe86ad6f09f6d
Reviewed-on: https://code.wireshark.org/review/5196
Reviewed-by: Bill Meier <wmeier@newsguy.com>

sigcomp-udvm: fix invalid access

Way back in 2006, svn r20041 (now gd408f2f6fc) worked around a problem by
limiting the amount of buffer dealt with in the udvm_state_access call (changing
it from state_minimum_access_length_buff[n] to STATE_MIN_ACCESS_LEN).

The very next line however, tries to read the same amount of buffer to display
to the user - since it seems nobody ever applied a proper fix for the original
case, just apply the same workaround to the second call.

Bug: 10662
Change-Id: I32c81ee2a8eb7b30ff7ab1fbbb6ea5f206385d4b
Reviewed-on: https://code.wireshark.org/review/5125
Reviewed-by: Evan Huus <eapache@gmail.com>
(cherry picked from commit 254436c43bdf59250780ff51fe07d1a8d0794ccd)
Conflicts: epan/sigcomp-udvm.c
Reviewed-on: https://code.wireshark.org/review/5177
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>

smb2: fully initialize dummy session

A few fields weren't being zeroed and were causing valgrind warnings.

Bug: 10569
Change-Id: I30e1e70c9445376806f385f3578b65bf6e320f8b
Reviewed-on: https://code.wireshark.org/review/4703
Reviewed-by: Evan Huus <eapache@gmail.com>
(cherry picked from commit d07c65de126f976190f1ce8d07ed8fe7deb90696)
Conflicts: epan/dissectors/packet-smb2.c
Reviewed-on: https://code.wireshark.org/review/5178
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>

tn5220: prevent another potential endless loop
exit the loop if dissect_unknown_data() returns 0

Change-Id: I681d170d4daf8efb06231790652ff7c941b46d89
Reviewed-on: https://code.wireshark.org/review/4882
Reviewed-by: Martin Kaiser <wireshark@kaiser.cx>
(cherry picked from commit 1463db37d9bbc9cd532afdf2817caaf8eb367831)
Reviewed-on: https://code.wireshark.org/review/5181
(cherry picked from commit d12ab6a48333f57710060c848340a97536ed2396)
Reviewed-on: https://code.wireshark.org/review/5182

mih: Fix some bugs; Add an XXX comment.

  One of the bugs: (2^32) is *not* 2 to the power of 32.

Change-Id: I0875ba5ac37717b778ee9252c7bce8a2ac442158
Reviewed-on: https://code.wireshark.org/review/5132
Reviewed-by: Bill Meier <wmeier@newsguy.com>

amqp: Fix a number of bugs in dissect_amqp_0_10_map().

 (Among other things: "2 ^ xxx" is *not* 2 to the xxx'th power).

Change-Id: I893b72e853ecbb5d7e6259e950fb347e89eed6c2
Reviewed-on: https://code.wireshark.org/review/5109
Reviewed-by: Bill Meier <wmeier@newsguy.com>
(cherry picked from commit 9ac261f205b7b6cba646368433eac550c7a6acd7)
Reviewed-on: https://code.wireshark.org/review/5111

[Automatic manuf, services and enterprise-numbers update for 2014-11-02]

Change-Id: I92f41798af384a9889a2804edab50ac1943554d6
Reviewed-on: https://code.wireshark.org/review/5076
Reviewed-by: Gerald Combs <gerald@wireshark.org>

Announce messages should only have a TLV if the message length is > 64.

Also, cleaned up a unnecessary global variable.

Bug: 10611

Reviewed-on: https://code.wireshark.org/review/5029
Reviewed-by: Evan Huus <eapache@gmail.com>
Petri-Dish: Evan Huus <eapache@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
Conflicts:
 epan/dissectors/packet-ptp.c

Change-Id: Ifb9bcde70405d591c21bd4e10195e1b192cba98b
Reviewed-on: https://code.wireshark.org/review/5062
Reviewed-by: Michael Mann <mmann78@netscape.net>

print: do not try to add an empty hex stream to PDML output for a text only hf

Bug: 10651
Change-Id: I4d7e79d6fad909453e3c6fa4973e11a5fd3f7ca4
Reviewed-on: https://code.wireshark.org/review/5060
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>

Convert a couple remaining isalpha() and isprint() calls that were left after
the merges to this branch.

This gets it compiling cleanly again.

Change-Id: I7273c5bb684656b26a98cbcb4e279bc0499c812c
Reviewed-on: https://code.wireshark.org/review/4986
Reviewed-by: Jeff Morriss <jeff.morriss.ws@gmail.com>

Use g_ascii_isalnum() rather than isalnum().

Only *ASCII* alphanumerics are allowed in filterable field names, so use
g_ascii_isalnum() to check for them.  That avoids issues with characters
with the 8th bit set and avoids locale-dependent behavior.

Change-Id: I4e3c4eec907f5e576629229fcf154fcf728b7a2e
Reviewed-on: https://code.wireshark.org/review/4973
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 8ea4df97e052f3fb0af3336990f2bef6d882b258)
Reviewed-on: https://code.wireshark.org/review/4975

Add enumerated value 639 to vals_parm_PID.

Per http://www.nowsms.com/discus/messages/1/EMI_UCP_Specification_40-8156.pdf

Bug: 10463
Change-Id: I55957f7b77b3dcf8c65c288791fed6428a1f4ba5
Reviewed-on: https://code.wireshark.org/review/4944
Reviewed-by: Michael Mann <mmann78@netscape.net>
(cherry picked from commit ee7243239a56e27c75c896f727a27538112535e5)
Change-Id: I55957f7b77b3dcf8c65c288791fed6428a1f4ba5
Reviewed-on: https://code.wireshark.org/review/4946
Reviewed-by: Michael Mann <mmann78@netscape.net>

Add a "placeholder protocol" for H.223 to avoid confusion when using Decode As.

The Decode As dialog presents two H.223 options and only "trial and error" can determine if you want bitswapping or not.

Change-Id: I9303c73375d130d3097d347bb374019dfc714068
Reviewed-on: https://code.wireshark.org/review/4909
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Reviewed-by: Michael Mann <mmann78@netscape.net>
(cherry picked from commit 48da429a04956f18667d9de3022b11a60811663c)
Change-Id: I9303c73375d130d3097d347bb374019dfc714068
Reviewed-on: https://code.wireshark.org/review/4920
Reviewed-by: Michael Mann <mmann78@netscape.net>

Various Mobile IPv6 bugfixes

Most of this was found by Boaz Brickner, but to prevent a few more bug reports, I tried to find any remaining MIPv6 options that used -1 to read until "end of packet" instead of "end of option"

1. Option-Code and Reserved fields should be 1 byte, not 2 in Local Mobility Anchor Address mobility option
2. IPv6 AUTH mobility option parses Mobility SPI
3. IPv6 MESG-ID (should just be a timestamp field)
4. IPv6 Redirect Mobility Option K and N bits are parsed incorrectly

5. The following options had a field that read until the end of the packet instead of the end of the option:
DNS-UPDATE-TYPE
AUTH Authentication Data
Care Of Test
Permanent Home Keygen Token
CGA Parameters
Signature
Experimental Mobility Option

Bug: 10630
Bug: 10629
Bug: 10626
Bug: 10625
Bug: 10624
Bug: 10622
Bug: 10619

Reviewed-on: https://code.wireshark.org/review/4932
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
(cherry picked from commit 803a2f9518b9d207094dabfc01ad462e46588ef3)

Conflicts:
epan/dissectors/packet-mip6.c

Change-Id: I1bd4eeb4ed11d64d99c833ebde985515d9ffb6f5
Reviewed-on: https://code.wireshark.org/review/4936
Petri-Dish: Michael Mann <mmann78@netscape.net>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Tested-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Reviewed-by: Michael Mann <mmann78@netscape.net>

[Automatic manuf, services and enterprise-numbers update for 2014-10-26]

Change-Id: I4d835b812e835beb80ccc706833b2aa96e60f778
Reviewed-on: https://code.wireshark.org/review/4939
Reviewed-by: Gerald Combs <gerald@wireshark.org>

ncp2222: initialize buffer in build_expert_data

We do a bunch of conditional logic to fill in the buffer, and then were checking
strlen(buffer) to see if we'd actually filled it in or not, but if we hadn't
then the buffer was garbage and strlen(buffer) was throwing valgrind warnings.
Ensure this works as intended by setting the first byte of the buffer
unconditionally at the beginning, so strlen(buffer) returns 0 as expected in
that case.

Bug: 10628
Change-Id: I6f6c29fbbcb601b6ebccb3a4b312f0fa72c169ae
Reviewed-on: https://code.wireshark.org/review/4926
Reviewed-by: Evan Huus <eapache@gmail.com>

tn5250: fix an endless loop, exit when offset is not incremented

Bug: 10596
Change-Id: I64509aa49fce4893dee4e7a74a560e09a98830f6
Reviewed-on: https://code.wireshark.org/review/4878
Reviewed-by: Martin Kaiser <wireshark@kaiser.cx>
Tested-by: Martin Kaiser <wireshark@kaiser.cx>
(cherry picked from commit bc2726578156f3608960fc65ce1f691639e6addc)
Reviewed-on: https://code.wireshark.org/review/4880
Reviewed-by: Evan Huus <eapache@gmail.com>

Don't use <ctype.h> macros.

They're locale-dependent (although those *particular* ones might not
be).

Remove no-longer necessary (or not-even-necessary-before-this-change)
includes of <ctype.h>

(cherry picked from commit 0d9b449ae95f26e86f8b433c7f4f99341443b048)

Change-Id: Ic2be240369c62a06bd5152fda46e6ad9027570cf
Reviewed-on: https://code.wireshark.org/review/4869
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Get rid of another toupper() call in favor of g_ascii_toupper().

(cherry picked from commit 991743b0b1cce99305a4b3162d6c520a09877c15)

Change-Id: Ib7650b773504f7f4572f297a01301bdf895ed917
Reviewed-on: https://code.wireshark.org/review/4867
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Fix previous merge.

Change-Id: I3cbb84cb93612fee9b6eeddc6e9e3554c0b55c7a
Reviewed-on: https://code.wireshark.org/review/4862
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Clean up comments.

Change-Id: I38fb39d6444469fd17373a9b0703b553b620e439
Reviewed-on: https://code.wireshark.org/review/4860
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Clean up comments.

Change-Id: I1448062101f55d448f4e4c5e54aee089b7e1c1c5
Reviewed-on: https://code.wireshark.org/review/4854
Reviewed-by: Guy Harris <guy@alum.mit.edu>

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().

(cherry picked from commit 80e6f6251e96dd7b8a9ab0fdbf63b03a73ef6692)

Change-Id: Iabf1304205850934e1b880c0bb81064c76dcd06f
Reviewed-on: https://code.wireshark.org/review/4853
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Eliminate a use of a ctype.h function.

That way, we avoid locale-dependent behavior.

(cherry picked from commit 4d6b3db5d4edc8743151c599330441ad2f943094)

Change-Id: Iabafc8e9e66f8c0cad6c53cc8aa36cc9454ef37f
Reviewed-on: https://code.wireshark.org/review/4847
Reviewed-by: Guy Harris <guy@alum.mit.edu>

[Automatic manuf, services and enterprise-numbers update for 2014-10-19]

Change-Id: I7754169b9f84f4d39f7cf7f11e02bdcb4bfcf1e0
Reviewed-on: https://code.wireshark.org/review/4841
Reviewed-by: Gerald Combs <gerald@wireshark.org>

packet-amqp.c: Temporary/preliminary fix for Bug #10582 (crash)

This is a temporary/preliminary fix to prevent the buildbot fuzz-test
 failures seen when testing the capture file attached to Bug #10582.

As noted in the bug, a complete fix will require some reworking of
  the amqp dissector.

Change-Id: I2571f1eeb46a7eede04b8d809e87b95f1ff529df
Reviewed-on: https://code.wireshark.org/review/4835
Reviewed-by: Bill Meier <wmeier@newsguy.com>
(cherry picked from commit 8def8ef9c40189472a46d9b1ad95289780e09af5)
Reviewed-on: https://code.wireshark.org/review/4837
Reviewed-by: Evan Huus <eapache@gmail.com>
(cherry picked from commit 4655270af94dcf45195cd0206b3e10a32861fa2a)
Reviewed-on: https://code.wireshark.org/review/4838

MIPv6: Mobility Option Binding Authorization Data for FMIPv6 Authenticator field is read beyond the option data

There's a single IPv6 packet with Binding Authorization Data for FMIPv6 mobility option.
The option's data length is 14 but and since the SPI takes 4 bytes, the Authenticator field should take 10 bytes.
However, the Authenticator field is read until the end of the packet, which is 178 bytes.

Found by boaz
Bug: 10577
Change-Id: Ia79c5131d8477c1a7c111c47c2a94e990bbeae4d
Reviewed-on: https://code.wireshark.org/review/4817
Reviewed-by: Michael Mann <mmann78@netscape.net>
(cherry picked from commit 8a10f77ab4d7722efc95170771540eb93b2c2f16)
Change-Id: Ia79c5131d8477c1a7c111c47c2a94e990bbeae4d
Reviewed-on: https://code.wireshark.org/review/4822
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>

MIPv6: Mobility Option Mobile Node Link Layer Identifier Link-layer Identifier field is read beyond the option data

there's a single IPv6 packet with Mobile Node Link Layer Identifier mobility option.
The option's data length is 9 but and since the Reserved field takes 2 bytes, the Link Layer Identifier field should take 7 bytes.
However, the Link Layer Identifier field is read until the end of the packet, which is 519 bytes.

Bug:10578
Change-Id: I518de7f6b75e5b749cb029845fd4846954ca6d0e
Reviewed-on: https://code.wireshark.org/review/4818
Reviewed-by: Michael Mann <mmann78@netscape.net>
(cherry picked from commit 51f496220322ac518d9485afe030d0568fdae080)
Change-Id: I518de7f6b75e5b749cb029845fd4846954ca6d0e
Reviewed-on: https://code.wireshark.org/review/4824
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>

MIPv6: Mobility Option IPv6 Address/Prefix marks too many bytes for the address/prefix field

One of the mobility options is the IPv6 Address/Prefix.
The prefix length of this option is 52.
Note that according to RFC 5568, the prefix length is in bits and not in bytes.
The IPv6 Address/Prefix field of this option marks 52 bytes.

In addition, I'm not sure that the IPv6 Address/Prefix field is a variable size field.
RFC 5568 section 6.4.2 has a diagram of this options that shows this field as if it always takes 128 bits, and the prefix length field just says the length of the prefix (not the length of the field).

Found by boaz

Bug:10576
Change-Id: I2a9a35cb7c537ec3b91c9a089f14fa31f41d8c80
Reviewed-on: https://code.wireshark.org/review/4816
Reviewed-by: Michael Mann <mmann78@netscape.net>
(cherry picked from commit 31c119fa789782190cea2608decaa114318bb5d8)
Change-Id: I2a9a35cb7c537ec3b91c9a089f14fa31f41d8c80
Reviewed-on: https://code.wireshark.org/review/4820
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>

More replacement of ctype.h functions.

Change-Id: I6a1b5b4e621080490d2f3161d954425973faeb5c
Reviewed-on: https://code.wireshark.org/review/4810
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Don't use ctype.h routines.

That eliminates locale-dependent behavior when parsing packets.

While we're at it, rename a variable to reflect what it's being used for.

(cherry picked from commit c734c57a37e4a84e8b66cc054d4b2c07aecebc9e)

Change-Id: Ic5d5fb6d73005b1f8e8932fc1666e473d6a42a24
Reviewed-on: https://code.wireshark.org/review/4808
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Get rid of some no-longer-necessary casts.

g_ascii_isXXX() can be handed a char value safely; they won't try to use
it as an index into a table before checking it.

(cherry picked from commit e3477bd13635a637d1337093457dbad499bd5538)

Change-Id: I35aaeaabcc896adb747aad628d259dd0b5b09c2e
Reviewed-on: https://code.wireshark.org/review/4803
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Don't use ctype.h routines.

That avoids locale dependency and handles possibly-signed chars(which
we weren't always doing before).

(cherry picked from commit a91799ebfa3a3aac705132722e57acc92fe4eddb)

Change-Id: I9c91ea79d5313317e9a8a877b4ac806e96fdb38f
Reviewed-on: https://code.wireshark.org/review/4800
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Don't use ctype.h routines.

That avoids locale dependency and handles possibly-signed chars (which
we weren't always doing before).

(cherry picked from commit 033f096ee909b63c0692b80416797f743940d054)

Change-Id: Id2282486d37e7e88ed144e96f29f3c67014e762d
Reviewed-on: https://code.wireshark.org/review/4796
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Don't use ctype.h routines.

That avoids locale dependency and handles possibly-signed chars (which
we weren't doing before).

(cherry picked from commit 5fc398c3286fda408ad5cd4b5834d1b2778db86a)

Change-Id: I98c0e532d14e78cc44fef701fb2ad4f29d8edf4e
Reviewed-on: https://code.wireshark.org/review/4791
Reviewed-by: Guy Harris <guy@alum.mit.edu>

isprint() -> g_ascii_isprint()

(cherry picked from commit 3157bf6ba5e4896b28316695ecd3a4659631c8c2)

Change-Id: I0e4645b83289a4a12b7d1eaa275a17c411ebbfce
Reviewed-on: https://code.wireshark.org/review/4789
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Fix potential invalid-reads in catapult files

As caught by CppCheck, the conditions checking that n is in bounds should
occur *before* the conditions using n as an index so that if n is out of bounds
we don't try and use it anyways.

(cherry picked from commit 72fc075e9a6b3ba79e49668e74a8dd187fc04f8a)

Change-Id: I23de43d52984cb441fa08f5f7384c486c9de817c
Reviewed-on: https://code.wireshark.org/review/4785
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Some replacement of ctype.h routines missed by the previous commit.

Change-Id: I0d132fceb68141b4e587311f7043ea990b3d083a
Reviewed-on: https://code.wireshark.org/review/4784
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Don't use ctype.h routines.

That avoids locale dependency and handles possibly-signed chars (which
we weren't doing before).

(cherry picked from commit 5bfde7c6386b56db80fec85e2ca7dc1d271b461c)

Change-Id: Ia20fe28a77b136e0860cffa43cc36d994af04f13
Reviewed-on: https://code.wireshark.org/review/4783
Reviewed-by: Guy Harris <guy@alum.mit.edu>

isdigit -> g_ascii_isdigit to handle 8-bit bytes correctly.

(cherry picked from commit 9d1afef1dfd3af626c471debe7f09f41d67c401f)

Change-Id: Icee86ee3efd31df183ff259d605ab0ca0295ae88
Reviewed-on: https://code.wireshark.org/review/4780
Reviewed-by: Guy Harris <guy@alum.mit.edu>

GSM MAP: fix detection of SMS direction when application context version is < 3

Bug: 10547
Change-Id: I4708fd9977e635c66ef1350ce5098520e4c2ce1e
Reviewed-on: https://code.wireshark.org/review/4579
Reviewed-by: Michael Mann <mmann78@netscape.net>
(cherry picked from commit 6463d2a7e66e259dabbeffbab67bcbeedd98133f)
Reviewed-on: https://code.wireshark.org/review/4775
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>

Don't use <ctype.h> macros, and eliminate an include of <ctype.h>.

This avoids locale-dependent tests, and fixes cases where we passed
signed char values to those macros (which is not safe with char being
signed, as it is on most, but not all, platforms).

(cherry picked from commit 63ab36be9266f79d92ac37e632299c8f0e2244f1)

Change-Id: I5deeb4dcf6d3828cef3a879bf2aac7e8da6010b1
Reviewed-on: https://code.wireshark.org/review/4766
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Don't use <ctype.h> macros, and eliminate an include of <ctype.h>.

This avoids locale-dependent tests.

(cherry picked from commit 25d9a7b5f1b880ab283947da28a8216dc9b7e551)

Change-Id: I84fe8ca3885fc95ac55b3afdff00a62084d8121f
Reviewed-on: https://code.wireshark.org/review/4760
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Use g_ascii_isalnum() rather than isalnum().

That way, we don't have to worry about casting the argument (which, in
one place, was done wrong - casting to int preserves the sign-extension
done with signed chars), and don't have to worry about a locale in which
particular 8-bit byte values are considered alphanumeric characters.

Change-Id: I129b4bfdad70ade4ab6e0a1d2c13d59ae9e6f524
Reviewed-on: https://code.wireshark.org/review/4751
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit e6fcff1eb62b581eb7fb3d7af1d2e38612a00b30)
Reviewed-on: https://code.wireshark.org/review/4753

Make sort_alphabet() not fail too badly with characters in [128,255].

Treat all values in the alphabet as unsigned.

Change-Id: I4476c75352f32673a9cd131ea233465f3376fa25
Reviewed-on: https://code.wireshark.org/review/4747
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit a517d500b4d5bdedba68127d295a37c2dbe4bf2c)
Reviewed-on: https://code.wireshark.org/review/4749

Better fix - just check that the upper and lower nibbles are >= 0xA.

An unsigned 8-bit integer is always <= 0xFF; that was the cause of the
warning.

To see whether a byte, when represented as hex, has letters for its
upper and lower nibbles, just check whether both nibbles are >= 0xA.
Cast the extracted nibbles to make sure there's no sign-extension.

Change-Id: If4c7717a5d2fe341c02e9309ee6b89973a6ac292
Reviewed-on: https://code.wireshark.org/review/4739
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit f6bc6f957a13471fecb4ce0a31ca09f30cdde44f)
Reviewed-on: https://code.wireshark.org/review/4741

gchar is not guaranteed to be signed; use gint8.

C does not guarantee that char is signed (that's why there's a keyword
"signed").  Use gint8, to indicate that it's not a character, it's an
8-bit signed integer.

Change-Id: Id632df23352840cbc86a4d88f4c7dade95ab7837
Reviewed-on: https://code.wireshark.org/review/4736
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 0db468e300287b8c8ee691f23873b299f64da0c7)
Reviewed-on: https://code.wireshark.org/review/4738

Fix issues in some ARM compilers, due to char being unsigned?

I'm not 100% certain the comparisons were right even with signed char;
make the comparisons unsigned vs. unsigned, regardless of whether char
is signed or not.  (No, C doesn't require it to be signed; that's why
there's a "signed" keyword.)

Change-Id: Icbbd1019a2f7d4ebb40d821255834f825cd7c5a7
Reviewed-on: https://code.wireshark.org/review/4731
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 353f6258f92ed7f8787cc32cd92ac3d44aa88f21)
Reviewed-on: https://code.wireshark.org/review/4733

It was called "interface_data" in 1.10.

Change-Id: Ibd918336c2c96c4bfede4564b6d147d472f6e0c2
Reviewed-on: https://code.wireshark.org/review/4714
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Don't allocate the interfaces array until we've read the SHB.

Nothing from the SHB gets stored in the interfaces array - it's filled
in from IDBs - so it doesn't need to exist when we read the first SHB,
and thus doesn't need to be freed if the attempt to read the SHB gets an
error or a "this isn't a pcap-ng file" indication.

Change-Id: I7e8c573efa67383c843177413cf5f3e2e279502c
Reviewed-on: https://code.wireshark.org/review/4713
Reviewed-by: Guy Harris <guy@alum.mit.edu>

NCP: cast to guint instead of gulong

Change-Id: Idc8f8f3177703ce5546b8a472034705cb6070707
Reviewed-on: https://code.wireshark.org/review/4666
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
(cherry picked from commit fb27a682a14ca653b36878186e0ddfdc1fea3a4e)
Reviewed-on: https://code.wireshark.org/review/4669

NCP: add a missing cast following g6ca40c5

Change-Id: I0b93c4faab52d830b8160d89b9bb8026ab2a216e
Reviewed-on: https://code.wireshark.org/review/4657
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
(cherry picked from commit 63faad686b49e02b86cd1961dd3bc8283f120da8)
Reviewed-on: https://code.wireshark.org/review/4660

ncp2222: don't overflow buffer constructing string

Fixes stack-smashing vuln.

Somehow there was already an XXX in the code about this, but nobody realized at
the time it was worth fixing... really?

Bug: 10552
Change-Id: I849068bd6c45146339444f295b72430d3c6e08f2
Reviewed-on: https://code.wireshark.org/review/4653
Reviewed-by: Evan Huus <eapache@gmail.com>
(cherry picked from commit b9988e2bbfc7c6c41ef82c559bd11a8c19170491)
Reviewed-on: https://code.wireshark.org/review/4654
(cherry picked from commit 6fa8338b381c0018f8dcc8286c554903a5fa4cb8)
Reviewed-on: https://code.wireshark.org/review/4655

[Automatic manuf, services and enterprise-numbers update for 2014-10-12]

Change-Id: I72dc592caa21049100926a5569a82355d2e0cede
Reviewed-on: https://code.wireshark.org/review/4617
Reviewed-by: Gerald Combs <gerald@wireshark.org>

[iSCSI] Fix incorrect heuristic.

The only valid check for Expected Data Transfer Length is :
R and W  MUST  NOT both be 0 when the Expected Data Transfer Length and/or
Bidirectional Read Expected Data Transfer Length are not 0

Backported with manual intervention from I5503a5ad524c4e237b143696fc8eda895bcb9172
with change to reflect v1.10 version of packet-isci.c code.

Ping-Bug: 10469

Change-Id: I56362010f354dac4a5e15978a288104fce2629af
Reviewed-on: https://code.wireshark.org/review/4529
Reviewed-by: Michael Mann <mmann78@netscape.net>

[Automatic manuf, services and enterprise-numbers update for 2014-10-05]

Change-Id: Id0448cce819b05a222d498996f7d4608fc33f15e
Reviewed-on: https://code.wireshark.org/review/4476
Reviewed-by: Gerald Combs <gerald@wireshark.org>

High Throughput is 11n, not 11g or 11a.

Change-Id: I86b6dac9cf3f1f524c1168d675bff4a04e361cee
Reviewed-on: https://code.wireshark.org/review/4374
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 4342d0f7a28cd31271b6d9f9116ee05e90ac44da)
Reviewed-on: https://code.wireshark.org/review/4376

mrcpv2: don't use an uninitialized buffer as a string

I have no idea what the original intent was with this magical
otherwise-totally-unused buffer, but it clearly serves no purpose now. Stop
reading garbage from it, and just read the data in the packet which is what the
field name suggests.

Bug: 10510
Change-Id: I05d0b98c04e59ea70247811168c4c8a64861f43d
Reviewed-on: https://code.wireshark.org/review/4354
Reviewed-by: Michael Mann <mmann78@netscape.net>

[Automatic manuf, services and enterprise-numbers update for 2014-09-28]

Change-Id: Ibd595940dee64e170048d3b47ae6f8c534667c11
Reviewed-on: https://code.wireshark.org/review/4341
Reviewed-by: Gerald Combs <gerald@wireshark.org>

Update a comment as per the previous commit.

Change-Id: Ic7adba4302d4d4abff7012cf09d68246ebade922
Reviewed-on: https://code.wireshark.org/review/4296
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 95dc9a927895db975e89bf3a65e4faf6ef28476b)
Reviewed-on: https://code.wireshark.org/review/4300

Renaming libtool* to glibtool* might require root permission.

Define DO_MV as "mv" or "sudo mv", depending on whether the user has
write access to /usr/local/bin, and use it to rename libtool* to
glibtool* and vice versa.

Change-Id: Ifbef4daa4521852bc200522743df9d23e7f0d1c2
Reviewed-on: https://code.wireshark.org/review/4295
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 92a0cb85f350d547b0816e1bdaee131768390d00)
Reviewed-on: https://code.wireshark.org/review/4298

Fix presumed cut-and-pasteo.

On errors *other* than a short read, we were driving on and processing
the non-data that we didn't read.

Change-Id: I6289ddf31ff7896918a030af9d1261bdc194e7d3
Reviewed-on: https://code.wireshark.org/review/4270
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 5f096acb454c3d57d23a786f49a884dd4afb0499)
Reviewed-on: https://code.wireshark.org/review/4272

6LoWPAN: fix encapsulating address decoding when a Mesh header is present

Bug: 10462
Change-Id: I37d532edc41193123b1a7bb380d48cb76c118bf4
Reviewed-on: https://code.wireshark.org/review/4161
Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com>
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>
(cherry picked from commit 1e47e050dcdf30eced478e6909408d79548975c6)
Reviewed-on: https://code.wireshark.org/review/4227

[Automatic manuf, services and enterprise-numbers update for 2014-09-21]

Change-Id: I5e6d324a1394c7ee05c236bfebaba3c0d27aceec
Reviewed-on: https://code.wireshark.org/review/4224
Reviewed-by: Gerald Combs <gerald@wireshark.org>

Align the error messages.

Change-Id: Icb4b3462b815d50cbdb7b13681bf4819665eb046
Reviewed-on: https://code.wireshark.org/review/4219
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Report (and free) err_info for WTAP_ERR_DECOMPRESS.

Change-Id: I61ecc58be4b0688c43df0b0f006fb899bdc1721c
Reviewed-on: https://code.wireshark.org/review/4216
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit be7f0539848238627db0e18acdc2dae5715c3182)
Reviewed-on: https://code.wireshark.org/review/4218

Free err_info on WTAP_ERR_DECOMPRESS.

Change-Id: I57fbb17baf43b0f1365fef72778ac0e65935456a
Reviewed-on: https://code.wireshark.org/review/4210
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 22a0d65a9a6691bcec24905baefeaba370f04043)
Reviewed-on: https://code.wireshark.org/review/4212

The previous cherry-pick required some manual intervention.

Change-Id: I3f81a0e002a74d9e64ae06428d40dc787a7f8e39
Reviewed-on: https://code.wireshark.org/review/4208
Reviewed-by: Guy Harris <guy@alum.mit.edu>

Add file_error() calls after file_read().

If file_read() doesn't return the exact number of bytes you asked for,
and you really need all those bytes to be there, you have to call
file_error() to find out what the problem is.

Change-Id: I4cc87bc1b6cc5a49bbcbf93b56106f57af290d20
Reviewed-on: https://code.wireshark.org/review/4205
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 32c38a4ecc7fc4f07349b0fbc13f7754be897546)
Reviewed-on: https://code.wireshark.org/review/4207

Set err_info if the attempts to read record headers fail.

If it fails due to, for example, the file being gzipped and having a bad
gzip CRC, the error returned is WTAP_ERR_DECOMPRESS and, for that error,
err_info is expected to be set to a string giving details of the
problem, so we need to pass back to our caller the string in question.

Bug: 10484
Change-Id: Ifa3139d2f28fadbc745c51611488865e97bcccb3
Reviewed-on: https://code.wireshark.org/review/4203
Reviewed-by: Guy Harris <guy@alum.mit.edu>

GTPv2: further fixes for Trace Information IE dissection

Bug: 10470
Change-Id: I8b61e5601379a568da39cce2335122e91ab37ec4
Reviewed-on: https://code.wireshark.org/review/4142
Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
(cherry picked from commit c4c82577d321ee288f2cba776560ef28b7143974)
Reviewed-on: https://code.wireshark.org/review/4154

GTPv2: fix dissection of Triggering Events IE

Bug: 10470
Change-Id: Ieb9819e578f078e956861a4e2d7ab3d8b96416d6
Reviewed-on: https://code.wireshark.org/review/4130
Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Evan Huus <eapache@gmail.com>
(cherry picked from commit 4cf9e71730df47281f53d736e43e76ba2beb96c5)
Reviewed-on: https://code.wireshark.org/review/4153
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>

Fix the patch-bzip2 target.

The main site URI scheme is now https. Update the URL in some other
places while we're here.

Change-Id: Ib03d4fd1c58dabd3cf5050dc4f79216e0b94d525
Reviewed-on: https://code.wireshark.org/review/4133
Reviewed-by: Gerald Combs <gerald@wireshark.org>
(cherry picked from commit 983ae6cadee518be589b53d9b149fd2e156e8ada)
(cherry picked from commit f4c3d07bd82c5ce146148c48dd9402517091a130)
Reviewed-on: https://code.wireshark.org/review/4135

1.10.10 → 1.10.11.

Change-Id: I604a8a69bfe0c2ede76889e26276c0c3442df057
Reviewed-on: https://code.wireshark.org/review/4132
Reviewed-by: Gerald Combs <gerald@wireshark.org>

Build 1.10.10.

Change-Id: I96268aa473668a675ab600ab9a0092fc4d8a2415
Reviewed-on: https://code.wireshark.org/review/4129
Reviewed-by: Gerald Combs <gerald@wireshark.org>

Updates for 1.10.10.

Change-Id: Id9c110fbd45f45ed366fec7b606a022c6a9929ca
Reviewed-on: https://code.wireshark.org/review/4120
Reviewed-by: Gerald Combs <gerald@wireshark.org>

dissect the alphanumeric message that's part of operation 30
Bug: 10464

Change-Id: I1a42877ee549b40e0c739dd2d1cbe4615ed9b9bd
Reviewed-on: https://code.wireshark.org/review/4098
Reviewed-by: Martin Kaiser <wireshark@kaiser.cx>
Tested-by: Martin Kaiser <wireshark@kaiser.cx>
(cherry picked from commit ed7c3282b1434641ecd28311167c0a3b3f1eae74)
Reviewed-on: https://code.wireshark.org/review/4116
(cherry picked from commit d73f2728eabfbf4e78b1ed0d4f2c07de9a73b1a4)
Reviewed-on: https://code.wireshark.org/review/4117

[Automatic manuf, services and enterprise-numbers update for 2014-09-14]

Change-Id: I4ce149fe4d156a5315900d33bfba55a0fdb34e93
Reviewed-on: https://code.wireshark.org/review/4108
Reviewed-by: Gerald Combs <gerald@wireshark.org>

Add some additional checks in SnifferDecompress().

Check the input pointer in the while clause of the loop, so that we
handle an empty input buffer.

When reading a bit mask, check before fetching the bit mask that we have
two bytes of bit mask and the byte after it.

Before putting an uncompressed input byte into the output, make sure we
wouldn't run past the end of the output buffer.

Before copying an earlier string from the output buffer, make sure it
doesn't run past the end of the data we've decompressed so far.

Bug: 10461
Change-Id: I8bb8d0d291368ae8bf0ac26970ff54d3262a7e6e
Reviewed-on: https://code.wireshark.org/review/4083
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 47c592938ba9f0caeacc4c2ccadb370e72f293a2)
Reviewed-on: https://code.wireshark.org/review/4085

packet-netflow.c: fix bug: "top-of-stack" --> "bottom-of-stack"

Bug #10458
Change-Id: I733bb54b13da8dd5b96837b57c1893e14f9622ec
Reviewed-on: https://code.wireshark.org/review/4067
Reviewed-by: Bill Meier <wmeier@newsguy.com>
(cherry picked from commit 0164b7582106ceaf244e8792ee635257ab4bace4)
Reviewed-on: https://code.wireshark.org/review/4069

packet-sip.c: col_...() functions must not be called under 'if(...tree)' (even indirectly).

Fixes Bug #10453

Change-Id: Ib3e5b480284ec1649b3728263f8fe164d53059cd
Reviewed-on: https://code.wireshark.org/review/4053
Reviewed-by: Bill Meier <wmeier@newsguy.com>

ses: initialize pres_ctx_id in session struct

Bug: 10454
Change-Id: Ib381646cad0a039514117735c48b461c9950f705
Reviewed-on: https://code.wireshark.org/review/4033
Reviewed-by: Evan Huus <eapache@gmail.com>
(cherry picked from commit 433a444d148f86f2562f804d25a57d00dc277cc0)
Reviewed-on: https://code.wireshark.org/review/4036
(cherry picked from commit 0942b2c13362b2e0a07a4583b46b59db6b44d7a8)
Reviewed-on: https://code.wireshark.org/review/4037

[Automatic manuf, services and enterprise-numbers update for 2014-09-07]

Change-Id: Ie94e7a30ff38c54175acc6cfb2dd2bce6ad5e726
Reviewed-on: https://code.wireshark.org/review/4026
Reviewed-by: Gerald Combs <gerald@wireshark.org>

6LoWPAN: fix storage of contexts coming from ICMPv6 Router Advertisement messages

Bug: 10443
Change-Id: I1a74b4d3036c61f0c8e1fca6373845ea40d87808
Reviewed-on: https://code.wireshark.org/review/4008
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
Tested-by: Pascal Quantin <pascal.quantin@gmail.com>
(cherry picked from commit b515cd3aec86f529643293995097f5a38468efa3)#
Reviewed-on: https://code.wireshark.org/review/4010

Work around an apparent bug in GtkFileChooserButton.

It looks like some versions of GTK+ intialize GtkFileChooserButton with a
location set to NULL and crash when you select an item from the dropdown
menu. Make sure we have a valid location set in that case.

Change-Id: If2556ef939b95d8af12581a7116a48fb46586f57
Ping-Bug: 10434
Reviewed-on: https://code.wireshark.org/review/3977
Reviewed-by: Gerald Combs <gerald@wireshark.org>
(cherry picked from commit 5e3a8fa85201b7f886ad41b85e6cf8c9eca4a978)
Reviewed-on: https://code.wireshark.org/review/3979
Reviewed-by: Evan Huus <eapache@gmail.com>

Don't print the packet counter when capturing if we're also printing
packet information to a terminal (which we assume is the same terminal
as the one to which the packet counts are being printed), as they get in
the way of each other.

Don't print it if we're sending the standard error to a terminal, or if
-q is specified, either.

Put all the setting of print_packet_counts together; it looks as if the
default value of print_packet_counts may have been changed to TRUE and
the code to handle -q wasn't changed to set it to FALSE if -q was
specified rather than setting it to TRUE if it wasn't specified.

Change-Id: Ifa8fd70c1694235f0ee8cd0e52eb860286329138
Reviewed-on: https://code.wireshark.org/review/3951
Reviewed-by: Guy Harris <guy@alum.mit.edu>

[Automatic manuf, services and enterprise-numbers update for 2014-08-31]

Change-Id: I19cc887da4b07edd9fc0a7d0bcce33a888869a2b
Reviewed-on: https://code.wireshark.org/review/3930
Reviewed-by: Gerald Combs <gerald@wireshark.org>

6LoWPAN: fix dissection when using multicast address compression

Also update the value_string arrays with the various SAM/DAM combinations

Bug: 10426
Change-Id: I87f13c29f42770ec655d85e2247b847bfe28e3ba
Reviewed-on: https://code.wireshark.org/review/3915
Petri-Dish: Pascal Quantin <pascal.quantin@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>
(cherry picked from commit 2eced3689d809e6a80a4d00a6bebd8b481e5ebc9)
Reviewed-on: https://code.wireshark.org/review/3922
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>

NTP: fix parsing of control assignments with empty value

Consider that ',' is a separator between assignments and make value presence optional

Bug: 10417
Change-Id: I23f2b4029548a1263d65ee11c6356270e7a89bd7
Reviewed-on: https://code.wireshark.org/review/3884
Reviewed-by: Evan Huus <eapache@gmail.com>
Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
(cherry picked from commit edcd2dd57f1649dbb25f1ad0f4a3a77c4d454322)
Reviewed-on: https://code.wireshark.org/review/3904
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>

Increase the error buffer size for rpcap.

If the rpcap port is unreachable pcap_findalldevs_ex can write more
than PCAP_ERRBUF_SIZE bytes to errbuf. E.g. if we try to capture from
Google's all-eights public DNS server we get:

----
Can't get list of interfaces: Is the server properly installed on 8.8.8.8?
connect() failed: A connection attempt failed because the connected
party did not properly respond after a period of time, or established
connection failed because connected host has failed to respond.  (code 1
----

Set the buffer to PCAP_ERRBUF_SIZE*4 bytes. Hopefully that's large enough.

Change-Id: I19f34cda16050c1ba8b9d7d6ed2d8e77b945a2af
Reviewed-on: https://code.wireshark.org/review/3880
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Reviewed-by: Evan Huus <eapache@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Gerald Combs <gerald@wireshark.org>
(cherry picked from commit 5ee328e90b26f82fbd00316110c5edd9a5c852e6)
Reviewed-on: https://code.wireshark.org/review/3886
(cherry picked from commit fa483e652110f1e7f812f252eb7dcab75ec891e3)
Reviewed-on: https://code.wireshark.org/review/3887

RPKI RTR: Wrong decoding of RPKI RTR End of Data PDU

The decoding of the End of Data PDU ignores the Serial Number, i.e., the last 32 bit of this PDU. Instead of that the current dissector implementation starts to decode a new RTR PDU.

Details of packet format see http://tools.ietf.org/html/rfc6810#section-5.8

Found by Matthias Wählisch

Bug: 10411
Change-Id: I8451b5fbda18a034022b97ff442ddb2a2dabed7b
Reviewed-on: https://code.wireshark.org/review/3832
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
(cherry picked from commit f44e082fa8403023425e27b28cb295f4e40e34e6)
Reviewed-on: https://code.wireshark.org/review/3834

[Automatic manuf, services and enterprise-numbers update for 2014-08-24]

Change-Id: I7b2bef480ab3aa0f6bacb8653a2261be966fca7f
Reviewed-on: https://code.wireshark.org/review/3813
Reviewed-by: Gerald Combs <gerald@wireshark.org>

0 returned from an open routine is not an error; don't provide an error.

If we return 0 from an open routine, we should not set err and should
not return an error message through err_info; it just means "this isn't
one of our files", which is *NOT* an error.

Change-Id: I0fc97c878e34a5e4072fa5dcbf7506c138753115
Reviewed-on: https://code.wireshark.org/review/3682
Reviewed-by: Guy Harris <guy@alum.mit.edu>

[Automatic manuf, services and enterprise-numbers update for 2014-08-17]

Change-Id: I8ea7f95f57804c810c78ce41c8e671440fe23a44
Reviewed-on: https://code.wireshark.org/review/3673
Reviewed-by: Gerald Combs <gerald@wireshark.org>

rtsp: parse the correct token for the status code

Don't call get_token_len on next_token *and* pass in next_token to store the
subsequent pointer - the token we want to parse is the *current* value of
next_token, not the next next token (which may be beyond the end of the buffer,
if next_token happens to be the *last* token).

Bug: 10381
Change-Id: I9fb27e8bdaf2f9556f61841de30cec04b98ffb96
Reviewed-on: https://code.wireshark.org/review/3663
Reviewed-by: Evan Huus <eapache@gmail.com>
(cherry picked from commit 73959159dbf34b4a0b50fbd19e05cb1b470be9b0)
Reviewed-on: https://code.wireshark.org/review/3664
(cherry picked from commit 949cca0810f7ed2f92adc16c57d2e4badb10a6a4)
Reviewed-on: https://code.wireshark.org/review/3665

An NLPID of 0xc0 refers to TRILL.

This will, at least, make it show up as the meaning of a field
containing an NLPID; whether it's actually used to say "a TRILL packet
follows" is another matter.

At least partially addresses bug 10382, and fully addresses it if it's
never actually used to say "a TRILL packet follows".

Change-Id: Ic9dc8ec432532bde692fad3aa9cc97a3109aa634
Reviewed-on: https://code.wireshark.org/review/3602
Reviewed-by: Guy Harris <guy@alum.mit.edu>
(cherry picked from commit 2cab8766c35b4f1a9b699fbc194aa57e9bef3e8a)
Reviewed-on: https://code.wireshark.org/review/3628