"Fossies" - the Fresh Open Source Software Archive  

Source code changes of the file "ChangeLog" between
wireshark-3.4.5.tar.xz and wireshark-3.4.6.tar.xz

About: Wireshark is a network protocol analyzer. It lets you capture and interactively browse the traffic running on a computer network.

ChangeLog  (wireshark-3.4.5.tar.xz):ChangeLog  (wireshark-3.4.6.tar.xz)
commit 6548abd191 commit 65a554b976
Author: Gerald Combs <gerald@wireshark.org> Author: Guy Harris <gharris@sonic.net>
Date: Mon Apr 19 12:23:07 2021 -0700 Date: Tue Jun 1 18:59:29 2021 -0700
Prep for 3.4.5.
commit 04f9d3e097
Author: Gerald Combs <gerald@wireshark.org>
Date: Mon Apr 19 10:39:01 2021 -0700
MS-WSP: Don't allocate huge amounts of memory.
Add a couple of memory allocation sanity checks, one of which
fixes #17331.
(cherry picked from commit b7a0650e061b5418ab4a8f72c6e4b00317aff623)
commit 19d03ef4ac
Author: Gerald Combs <gerald@wireshark.org>
Date: Wed Mar 31 11:30:45 2021 -0700
CMake: Enable AUTO{MOC,UIC,RCC} according to our CMake version.
As noted in be2b0fc810, we need to set CMAKE_AUTO* before searching for
Qt packages when using 3.20.0 and later. However, this fails if we're
using CMake 3.9.6 or earlier. Set CMAKE_AUTO* where needed depending on
our CMake version.
Ping #17314.
(cherry picked from commit a80ea46ff7f19e4c7f9ade4970b0f2a2202f2ea2)
commit eee907aeb5 pcapng: set the length of the options item.
Author: naf <naf@sdf.org>
Date: Tue Feb 2 13:25:52 2021 -0600
QT ByteViewText: calculate string widths consistently to prevent clipping It runs up to either the end of the option data or the terminating
end-of-options option (readers MUST handle lists of options that
contains an end-of-options option and lists of options that don't).
For QT >5.11, stringWidth() uses horizontalAdvance, which gives different (cherry picked from commit 2f5c0ffdb26d753943201029ff65e237c1a9776b)
(longer) widths than the old boundingRect().width() method.
Other locations use the boundRect().width() method directly, resulting commit 73fe213954
in underestimating line widths and clipping the last characters in Author: Guy Harris <gharris@sonic.net>
the byte view window. Date: Mon May 31 02:44:57 2021 -0700
Fix by forcing all width calculations to use stringWidth(). ascend: set rec->rec_type.
Closes #17087.
(cherry picked from commit 95f3d1b0750044967c6cf2f767d809453ca7819b) REC_TYPE_PACKET is 0, so if it's been initialized to 0, and never gets
overwritten, this fixes code withotu fixing a visible bug, but it should
be done anyway.
Conflicts: (backported from commit 162251176ae206430b81bd8b467bc22c6c7bcd8b)
commit cc954fae7e commit e8938c10a2
Author: Gerald Combs <gerald@wireshark.org> Author: Gerald Combs <gerald@wireshark.org>
Date: Sun Apr 18 09:52:11 2021 +0000 Date: Sun May 30 17:37:02 2021 +0000
[Automatic update for 2021-04-18] [Automatic update for 2021-05-30]
Update manuf, services enterprise numbers, translations, and other items. Update manuf, services enterprise numbers, translations, and other items.
commit e49430b4d3 commit 225352970b
Author: Stefan Metzmacher <metze@samba.org> Author: Pascal Quantin <pascal@wireshark.org>
Date: Tue Apr 6 20:52:17 2021 +0200 Date: Sat May 29 12:48:58 2021 +0200
packet-ldap: fix regression for SASL handling Windows: upgrade Npcap to 1.31
commit 19b3376a24192319a008399e8c44f1655f6dcebd commit 389025227e
("LDAP bogus malformed errors: decoding encrypted data") Author: Gerald Combs <gerald@wireshark.org>
introduced 2 problems: Date: Fri May 28 16:08:48 2021 -0700
- guint decr_len = tvb_reported_length(decr_tvb); was
always called with decr_tvb==NULL
- dissect_ldap_payload() was not called if sasl_tree is NULL,
it needs to be called even if the tree pointer are NULL
in order to have the COL_INFO setup correctly.
I guess this should also be backported to stable branches
(together with 2e6d3b571b50f6a0df31787cb95ba0f66c4fa85f
"LDAP: SASL Buffer doesn't include Length field")
Signed-off-by: Stefan Metzmacher <metze@samba.org>
(cherry picked from commit 1d623fd54116fbe234300712a9cbf04560802e9f)
commit 0c1743656f
Author: Uli Heilmeier <uh@heilmeier.eu>
Date: Mon Mar 22 22:34:20 2021 +0100
LDAP: SASL Buffer doesn't include Length field
SASL Buffer starts after the SASL Buffer Length field. Therefore
we should only mark the bytes without the Length field.
Sample capture can be found in wireshark/wireshark#15128
(cherry picked from commit 2e6d3b571b50f6a0df31787cb95ba0f66c4fa85f)
commit d28ded7244
Author: Simon Holesch <simon@holesch.de>
Date: Sat Mar 6 01:56:06 2021 +0100
ptvcursor: Fix crash with deeply nested subtrees
If the proto tree is more than 8 levels deep, the subtree_lvl array Prep for 3.4.6
length is extended, by allocating a new area and copying everything into
that new area. However the old array length wasn't calculated correctly,
so only part of the subtree_lvl array was copied, causing a crash after
two ptvcursor_pop_subtree() calls.
(cherry picked from commit fa483ac191d4ec6b1994d600e37af6815a97ce18) commit 7308ab766b
Author: João Valverde <joao.valverde@tecnico.ulisboa.pt>
Date: Wed May 26 04:17:59 2021 +0100
commit 489fc40a49 dfilter: Fix handling of escaped quotes in macros
Author: Michael Tuexen <tuexen@wireshark.org>
Date: Wed Apr 14 14:33:29 2021 +0200
Fix conversation table when using epoch based time We can't unescape characters when expanding a display filter macro.
The escaping must be preserved until the expression is evaluated in
the display filter engine, otherwise it will likely generate a syntax
error in the parser.
Ensure that if using tshark -q -t e -z conv,tcp the reported In the macro body we allow '$' (or any other char) to be escaped
start time is relative to the epoch time and not relative to with backslash (preserving the backslash).
the time of the first packet in the capture file.
Thanks to Theresa Enghardt for reporting the issue and to Fixes #17160.
Peter Lei for initialy looking into it.
(cherry picked from commit f099bd179a6ff86e3f9fd6b779ccad707f379aad) (cherry picked from commit 1dba58789d7fbf6952d631774c94f63e2179d4d1)
commit cb432b878e commit e6eb1d7370
Author: Gerald Combs <gerald@wireshark.org> Author: Gerald Combs <gerald@wireshark.org>
Date: Sun Apr 11 09:53:17 2021 +0000 Date: Fri May 7 14:21:09 2021 -0700
[Automatic update for 2021-04-11]
Update manuf, services enterprise numbers, translations, and other items. Tools: Show only filenames when fuzzing.
commit cdca1f2d03
Author: Guy Harris <gharris@sonic.net>
Date: Thu Apr 8 19:38:26 2021 -0700
pcapng: expand a comment and tweak the version test.
Drop in the comment from libpcap about version 1.2 (I wrote that Show only the basename of each capture file when fuzzing, which is less
comment, and generously double-license it under the BSD license and the noisy than the full pathname.
GPL :-)).
Redo the version test as (cherry picked from commit 4266e1e9baa2ecce49eaaa20a5beecb39d21a0af)
if (!({version is one we handle})) commit 0d8be1fb79
Author: John Thacker <johnthacker@gmail.com>
to match the way it's done in libpcap. Date: Tue May 25 08:18:22 2021 -0400
(cherry picked from commit 7de6b0822ad764edfa0ddd76e6a383bd3408a4e6)
commit 716dd09605 DVB-S2-BB: Prevent infinite loop
Author: Loris Degioanni <loris@sysdig.com>
Date: Wed Apr 7 16:54:26 2021 -0700
sysdig: a couple more fixes Commit 4bf4ee88f0544727e7f89f3f288c6afd2f650a4c removed an else
statement that broke out of the BBFrame processing loop. Without
it, infinite loops might be possible if the GSE frames have bit errors
in the length field.
- parse the number of system call arguments in a way that works for both V1 (cherry picked from commit 0137c24d60934f131b25506a88c9464e4dc827de)
and V2 event blocks
- returned the correct error string when unable to read the nparams entry fr
om a sysdig event block V2
(cherry picked from commit 7894b1d0ea86224a16c3dbebae0c6db56f590f84) commit b20a77698d
Author: John Thacker <johnthacker@gmail.com>
Date: Mon Apr 5 09:03:36 2021 -0400
commit 4e7df5af01 ftype-protocol: Fix crash when comparing _ws.expert to literals
Author: Loris Degioanni <loris@sysdig.com>
Date: Tue Apr 6 14:42:39 2021 -0700
sysdig: implementation fixes The ftype-protocol has two components to its value - a tvb, which is
allowed to be be NULL (most notably in _ws.expert), and a string
description. They can also be created from string literals, such as
in display filters. It's possible to compare protocols with a NULL
tvb with protocol terms created from literals, e.g. entering the
display filter "_ws_expert < 1".
- make sure nparams is initialized Partially revert 69e2603c48d04a675785d9e7bad162ebb9a83b07 so that
- validate block_total_length in the correct way for sysdig event blocks this doesn't crash, by assigning proto_string to the empty string
instead of null when creating from a literal. Fixes #17316
(cherry picked from commit cbbe6605043b6137f0a7eb22693969586893637d) (cherry picked from commit 31297dbb82da0b3adf5c257398638d9b4da94931)
commit 180c96beb8 commit 13546c7a18
Author: Loris Degioanni <loris@sysdig.com> Author: Gerald Combs <gerald@wireshark.org>
Date: Tue Apr 6 13:35:58 2021 -0700 Date: Tue May 25 14:04:51 2021 -0700
sysdig: simplified the separate handling of the two different sysdig block t ypes CMake: Update autogen properties.
(cherry picked from commit 8c70dd8d176fca0c7e4ed759de0a05c2dd6424cf) Set the AUTOMOC, AUTOUIC, and AUTORCC properties for the qtui and
wireshark targets to match what we currently do in master. This should
keep us from running moc and uic on unwanted targets.
commit e94d9f4549 commit 113062025d
Author: Loris Degioanni <loris@sysdig.com> Author: Pascal Quantin <pascal@wireshark.org>
Date: Mon Apr 5 22:16:09 2021 -0700 Date: Mon Nov 23 22:08:04 2020 +0100
sysdig: support the most recent version of sysdig Qt: fix some Qt 5.15.2 deprecation warnings
Update the pcap-ng reader and sysdig event dissector to support the second v
ersion of the sysdig event block, which was introduced after Wireshark's origina
l implementation
(cherry picked from commit fbe8d3a00f1408960186572de96f82b4616eed49) (cherry picked from commit fb2414ae6dbdc3d81c9ccdd24eb65cd8324065ea)
commit 1be17e2d30 commit 1d22f8f6b6
Author: Gerald Combs <gerald@wireshark.org> Author: Gerald Combs <gerald@wireshark.org>
Date: Sun Apr 4 13:28:16 2021 -0700 Date: Tue May 25 12:44:08 2021 -0700
GitLab CI: Switch to our new Windows runner.
(cherry picked from commit 85b36ef4b6090b719bb46bc17ec58a07faff97ab)
commit f1677bd31c
Author: David Perry <boolean263@protonmail.com>
Date: Thu Jan 7 16:21:42 2021 -0500
maxmind: add optional synchronous lookups Release notes: The Windows installers now ship with Qt 5.15.2.
When tshark enables synchronous resolution of IP addresses to names, commit 6da1114152
forces calls to maxmind_db_lookup_ipv4()/_ipv6() to block-wait for the Author: Gerald Combs <gerald@wireshark.org>
maxmind response. Date: Mon May 24 13:45:31 2021 -0700
Proposed fix for #14691. GitLab CI: Enable ccache for fuzz builds.
(backported from commit c0abaa06f717a38ba15c24f10ee19fcc69b21537) Enable ccache for our fuzz builds. Add sections as described at
for the CMake step as well.
commit 3e35fecccb (cherry picked from commit 02f4dcb0ad13fd3b671c212aec739be96ca2dc41)
Author: Vahap Emin Agaogullari <vahapemin44@gmail.com>
Date: Tue Feb 23 11:15:14 2021 +0000
PROFINET: Multiple write in record dissection commit ffe59d91a1
Author: John Thacker <johnthacker@gmail.com>
Date: Mon May 24 19:43:20 2021 -0400
COContainerContent dissects PDInterfaceMrpDataAdjust and RANAP: Don't attempt to re-register heuristic dissectors
PDInterfaceMrpDataAdjust dissects remaining COContainerContent
because of offset problem. Offset problem is fixed.
(cherry picked from commit ccec04ede0591a5e83d2664c26ed5cb4481a6809) Move the RANAP heuristic dissector registration under the initialization
(cherry picked from commit ecbe04e0bbc6634eaa6c54968870fee61760107c) guard that they're only registered once. Prevents console warnings about
the dissectors already being registered to the sccp and sua tables if
a RANAP preference is changed. (Backported manually to regenerate the
dissector via asn2wrs.py)
commit 4a7ddb6b1a commit 67ca5e98c7
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Mon Mar 29 00:55:23 2021 +0000 Date: Sun May 23 23:14:27 2021 -0700
tvbuff_subset: fix its implementation of string scanning.
Both subset_find_guint8() and subset_pbrk_guint8() pass the parent
tvbuff to tvb_find_guint8()/tvb_ws_mempbrk_pattern_guint8(), along with
the offset in that tvbuff.
That means that the offset they get back is relative to that tvbuff, so k12: plug a memory leak.
it must be adjusted to be relative to the tvbuff *they* were handed.
For subsets of frame and "real data" tvbuffs, there's a single lump of If the two putative number-of-records values don't match (meaning one of
data containing the content of the subset tvbuff, so they go through the them is presumably the number of records and the other one isn't - we
"fast path" and get the offset correct, bypassing the broken code; don't know which is the case), free up the private data structure we
that's the vast majority of calls to those routines. allocated before returning an error.
For subsets of *composite* tvbuffs, however, they don't go through the (cherry picked from commit 7f6c5d0137c94bdc9f7b07ee249d1b3cc9396dd9)
"fast path", and this bug shows up.
This causes both crashes and misdissection of HTTP if the link-layer is commit 7438c67f06
PPP with Van Jacobson compression, as the decompression uses composite
Fixes #17254 and its many soon-to-be-duplicates.
(cherry picked from commit 2ba52cdc0e4216dafdfc32498fc0210c99449ec9)
commit 636f7f1c09
Author: Gerald Combs <gerald@wireshark.org>
Date: Sun Mar 28 09:54:39 2021 +0000
[Automatic update for 2021-03-28]
Update manuf, services enterprise numbers, translations, and other items.
commit 193b1fc545
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Sun Mar 28 02:31:37 2021 +0000 Date: Sun May 23 19:37:09 2021 -0700
wslua_tvb: correctly trim off the end of a tvbuff.
The length specified in a TvbRange is the *actual packet length*, not
the *sliced-to* length, so use tvb_new_subset_length() to cut it short.
This fixes the fix for #15655, and addresses at least some of the issues
in #17255.
(cherry picked from commit cda18f951ebfe53dc5bbb841618dc12e31b38ec6)
commit 6ef6e8f607
Author: Gerald Combs <gerald@wireshark.org>
Date: Fri Mar 26 12:52:01 2021 -0700
CMake: Enable AUTO{MOC,UIC,RCC} earlier.
for Qt packages. This is apparently required for CMake 3.20.0 and later.
Fixes #17314.
(cherry picked from commit be2b0fc810fb2f2a80e471138dc29f0cbe87b209) erf: set USERAPPL if we have the application version but not the name.
Conflicts: Just say "(Unknown application) <version>".
commit 3bca4e3b3f This also means that we don't leak the app_version string if there's no
Author: Nardi Ivan <nardi.ivan@gmail.com> app_name string.
Date: Tue Mar 16 17:06:20 2021 +0000
SMB2: fix two memory leaks
* Since c3342930 we don't free anymore the entries in the files hashtables.
The cleanest solution is probably to convert these hashtables into two
wmem_map_t structures and let the wmem core handling any cleanup.
* b0f5b2c1749 added supported for chained compression; the uncompressed
tvb must be freed
(cherry picked from commit e677a909e15126a61253004c88e48ae3c64420fc)
commit 6b072b63f0
Author: Uli Heilmeier <uh@heilmeier.eu>
Date: Tue Mar 23 11:32:42 2021 +0000
DNS: Fix pointer for is_multiple_responses
As discussed in wireshark/wireshark!2497 there is no need (cherry picked from commit 297b6c5407426aa33b197f0f5d280ce4c6c04dc3)
for a pointer to a pointer.
(cherry picked from commit 337bdf8eb34be3fc5ccd1197868077e6276a2d6c) commit ac7673bd0a
Author: Guy Harris <gharris@sonic.net>
commit 06e20df535 Date: Sun May 23 17:56:18 2021 -0700
Author: Uli Heilmeier <uh@heilmeier.eu>
Date: Sun Mar 21 10:15:56 2021 +0100
DNS: IXFR/AXFR queries with multiple responses
IXFR and AXFR queries can have multiple DNS responses. As all responses netscaler: plug a memory leak.
belong to one transaction, they have the same transaction ID.
We shouldn't handle them as retransmits. Free the read buffer if the first read fails.
Fix: wireshark/wireshark#17293 (cherry picked from commit 976ccc9a00d5abd4f1e1a6041378f8cb1292b27e)
(cherry picked from commit 07fb47111ee968ba132ce98764486dc825209e80)
commit 878e8058a6 commit 7c969e2deb
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Mon Mar 22 19:13:34 2021 -0700 Date: Sun May 23 17:04:59 2021 -0700
Do the LFS checks before processing any subdirectories. tacacs: free a buffer if it's not used.
That's necessary in order to make sure that the required -D flags show Plugs a leak.
up when building code from all subdirectories.
(cherry picked from commit 3f556a6e7639716a7254a4fcdf88e33ef3896fe8) (cherry picked from commit 1bb64b7e0ca46442b3f3a65e4b99893674b3c287)
commit d0fdecc6b6 commit f032d776e6
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Mon Mar 22 02:47:56 2021 -0700 Date: Sun May 23 13:55:48 2021 -0700
Replace the Large File Support CMake stuff. dnp: plug a memory leak.
The existing stuff doesn't appear to work (I tried it on 32-bit Ubuntu If we're throwing away the data, *throw away the data* - free it, as
18.04, and it did *not* add any flags to the compilation, as it appeared we're not using it as the backing data for a tvbuff.
not to conclude that they were necessary, even though they were).
Pull in the stuff from libpcap, which *does* appear to work. (it does (cherry picked from commit 618661b22e34a59b21117db723d8ff91e064d4ba)
so in my 32-bit Ubuntu testing).
This should fix #17301. commit 191e6744f9
While we're at it, fix cppcheck.sh so that it doesn't attempt to run
cppcheck on files that have been deleted.
(cherry picked from commit 0cc59d38ab6ef38e965d4de8a3ec4b80bcb5a6a3)
commit 76f01ea726
Author: Gerald Combs <gerald@wireshark.org> Author: Gerald Combs <gerald@wireshark.org>
Date: Sun Mar 21 09:57:41 2021 +0000 Date: Sun May 23 09:55:24 2021 +0000
[Automatic update for 2021-03-21] [Automatic update for 2021-05-23]
Update manuf, services enterprise numbers, translations, and other items. Update manuf, services enterprise numbers, translations, and other items.
commit b2d8858b9a commit ecf1616e35
Author: Darius Davis <darius@vmware.com> Author: Guy Harris <gharris@sonic.net>
Date: Fri Mar 19 23:37:53 2021 +0000 Date: Sun May 23 01:27:11 2021 -0700
Minor doc cleanups.
Here's a grab bag of trivial cleanup to the documentation. This change:
- Cleans up some comments in the asciidoctor macros which are no longer OID handling: fix a memory leak.
accurate (and do not appear in the build products anyway).
- Fixes a missing space in the text "Wireshark Q&A" in the release notes. There's a "break" in some code that appears to be copied and pasted from
a switch statement; the break would exit the loop (and leak memory
allocated within the loop), which does not appear to be the intent, so
it may have been copied over incorrectly. Remove it.
- Allows the "docbook" backend to produce hyperlinks too... That seems to be While we're at it, redo the "constant-time append to the end of a loop"
necessary if we want to start using our custom link macros in WSDG, which code to be a bit clearer, both to humans reading the code and code
seems like a reasonable thing to do. And fixes up a wrong variable name i analyzers reading the code.
the handling of the case where we are not able to produce a hyperlink.
(cherry picked from commit 4c513fb4ab880b2c63491d91b8ccf6ce17627f08) (cherry picked from commit c73ab16bef0c97dd67f03fdfa7063958d1712d8b)
commit f63d93957d commit faea31d0a1
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Thu Mar 18 04:22:41 2021 +0000 Date: Sun May 23 00:03:01 2021 -0700
caputils: pull the includes out of extern "C" { ... }. epl-profile-parser: plug a memory leak.
Have separate #ifdef HAVE_LIBPCAP ... #endif sections for the includes g_key_file_get_groups() returns a pointer to g_mallocated data; we need
and the definitions/declarations. to pass its return value to g_strfreev() when we're done with that data,
to free it up.
(There are no good solutions that don't require hopping in a time (cherry picked from commit 64f3f08702d09f912c2713ec830c91ead242507f)
machine and changing history.)
(cherry picked from commit 5a4cb7ce331830b74c703882c1738d6c1e2c5fa7) commit 54b8952150
commit fcefe78a67
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Wed Mar 17 21:59:55 2021 +0000 Date: Sat May 22 19:01:22 2021 -0700
caputils: get rid of the extern "C" { idiom.
Instead, declare each function with EXTERN_C, #defined as extern "C" in Plug another leak.
C++ and just extern in C.
This avoids all the thrashing to try to keep headers outside extern "C" If cf_export_specified_packets() succeeds, and it wrote to a temporary
{ by the simple expedient of not *having* extern "C" {. file, it leaks the name of the file to which it was writing. Free that
after we've renamed that file on top of the target file (safe save).
(cherry picked from commit 4584b9d1d396920468e0b8a71f1681c7cfda382b) (cherry picked from commit 8ca86b29bfa4780b17e6d15ab1f382f6f79a5065)
commit 2974b30150 commit 93e794c373
Author: Dario Lombardo <lomato@gmail.com> Author: Guy Harris <gharris@sonic.net>
Date: Wed Mar 17 21:12:42 2021 +0000 Date: Sat May 22 12:23:50 2021 -0700
caputils: fix compilation without pcap.
The extern declaration must be put outside the ifdef to match the Plug a memory leak.
closing statement as well as surrounding al the functions.
Fixes: 2820156fbd4b0213ca015216d8cd97c31b503c8c (Move still *more* headers o If the user aborted the process of exporting packets, if we're writing
utside of extern "C".) to a temporary file, we unlink the file, but we don't free the
g_mallocated name of the file, so it's leaked. Free it.
(cherry picked from commit 9ca121bcf87dc638d9f0bf1314bee7b906523b70) (cherry picked from commit 54508703b0ac4d85df4104f6114a247d68c42b8e)
commit 95c7c1c0f2 commit d072335658
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Tue Mar 16 21:57:30 2021 +0000 Date: Fri May 21 19:40:00 2021 -0700
kerberos: regenerate packet-kerberos.h. fuzzshark: close a leak.
We updated the template; regenerate the header. If init_progfile_dir() fails, it returns a g_mallocated string with an
error message. After printing the error message, free the string.
(cherry picked from commit 7efb2120bc37cbb3b23682c16f5bb73e05295383) (cherry picked from commit c22b857942ea3f123d9fa5c31e04e85c6f3928f5)
commit b5d65c0074 commit 0e6c7c7bb2
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Tue Mar 16 13:50:13 2021 -0700 Date: Fri May 21 19:14:03 2021 -0700
Move still *more* headers outside of extern "C". protobuf: close another leak.
If a header declares a function, or anything else requiring the extern Close the directory handle we've opened before returning a failure
"C" decoration, have it wrap the declaration itself; don't rely on the indication if pbw_load_proto_file() or load_all_files_in_dir() reports a
header itself being included inside extern "C". failure.
(cherry picked from commit 2820156fbd4b0213ca015216d8cd97c31b503c8c) (cherry picked from commit f0abd29e48c1bd724efea606000b86356a764011)
commit 0b8c953533 commit 161e765500
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Tue Mar 16 04:33:00 2021 -0700 Date: Fri May 21 18:33:56 2021 -0700
Move even more headers outside extern "C". protobuf: close a leak when file loading fails.
If a header declares a function, or anything else requiring the extern Free the path we've constructed before returning a failure indication if
"C" decoration, have it wrap the declaration itself; don't rely on the pbw_load_proto_file() or load_all_files_in_dir() reports a failure.
header itself being included inside extern "C".
(cherry picked from commit e434d404d0db719440cc911729d225417a49b4f4) Also, explicitly compare pbw_load_proto_file()'s return value against 0,
to make it a little clearer that it's *not* a Boolean, it's a return
code (with 0 meaning success and different non-zero values meaning
failure; if it matters *which* failure it is, we should probably have
otherwise we should just make it a Boolean).
commit f1f937dd26 (cherry picked from commit f1ffe7d4215ac1cc80d9596e6604b30ddfa59fcf)
Author: Guy Harris <gharris@sonic.net>
Date: Tue Mar 16 11:21:16 2021 +0000
Move more headers outside extern "C". commit 4a17759938
Author: Nardi Ivan <nardi.ivan@gmail.com>
Date: Mon May 17 19:42:18 2021 +0200
If a header declares a function, or anything else requiring the extern QUIC: improve handling of unencrypted padding data
"C" decoration, have it wrap the declaration itself; don't rely on the
header itself being included inside extern "C".
(cherry picked from commit 1e1f4e6b5f9a309cef55d4459c8bba40d6acc104) 0af60377b4 added an heuristic to detect (unencrypted) padding data;
it is based on the fact that all coalesced QUIC packets must have the
same CID.
Unfortunately it doesn't work when the CID length is 0.
commit d0e6ad310b Treat decryption error of SH packets as a non fatal error, report them
Author: Michal Ruprich <michalruprich@gmail.com> as possible padding data misdetectd as coalesced packets and try
Date: Wed Feb 10 15:32:18 2021 +0100 decrypting next traffic.
Moving glib.h out of extern C Close #17383
(cherry picked from commit c8246c99737c7a844f45eb0e777382cc68397d17) (cherry picked from commit 389a899a18742185d14da729bf308505ca4f4a2e)
commit f6cfb8362c commit dd0c60a001
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Mon Mar 15 04:59:42 2021 +0000 Date: Wed May 19 14:56:30 2021 -0700
debian-setup: include GCC and G++ in the basic list. AUTHORS.src: fix Jeffrey Wildman's entry.
At least on my just-now-installed Kubuntu 20.04 VM, G++ wasn't installed [AT]ify his email address.
by default, and you need that to compile Wireshark (you can avoid it if
you're not building the GUI code, but the GUI code is Qt-based, so it's
in C++). Add both GCC and G++ to the basic list.
(cherry picked from commit 5f0dc153dd317ed1eaf8bb7fdf6b37268f951eac) This should address the recently-added problem mentioned #16658.
commit 82f93c42f7 commit 82b0eac6af
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Mon Mar 15 00:40:42 2021 +0000 Date: Tue May 18 01:03:04 2021 -0700
Fix the captions for the buttons in the "export specified packets" dialog. PortsModel: don't populate it by doing a lot of weird string-pushing.
The "Selected packet" caption is wrong, as there could be more than one When enumerating port-to-name entries, the callback to
packet selected. Furthermore, in the non-Windows "export specified wmem_map_foreach() gets passed:
packets" dialog, the caption is "Selected packets only", so go with the
full caption.
The "Marked packets" caption also lacks the "only" that it has in the - a key, which is the port number for the entry;
other dialog; add it as well. - a value, which is a pointer to a structure containing pointers to port
names for various transport protocols;
- a user data pointer.
(cherry picked from commit f785cf53401c38d695de7b961babc6b796466a50) That's sufficient (if you work around some C++ annoyances) to append a
row to a PortsModel, if the user data pointer is a pointer to the
commit 7e0360206e The existing code, instead, appended to a QStringList of lines (in
Author: Guy Harris <gharris@sonic.net> effect, undoing the effort of the code that read the services file and
Date: Sun Mar 14 06:51:36 2021 -0700 filled in the wmem_map, re-generating a set of lines) in the callback,
and then iterated over all the lines, splitting them with blanks and
Clean up "Export PDUs to File" code. appending rows.
Combine exp_pdu_file_open() is called only by do_export_pdu(); just Looking at that made my eyeballs bleed so badly that I decided not to
combine them into one routine. spend any time figuring out why it wasn't working.
Get rid of the exp_pdu_t * argument to do_export_pdu(); instead, have So I just make the callback just append rows, avoiding all the
the exp_pdu_t structure be a local variable in that routine. There's no string-pushing.
need to initialize exp_pdu_data.pkt_encap in
ExportPDUDialog::on_buttonBox_accepted() - do_export_pdu() already does
The return value of do_export_pdu() isn't used; don't return anything. Fixes #17395.
(backported from commit 01151ec332f6232cd7b86d1b9026e54decb88203) (cherry picked from commit 6e95a0aa47dd739edd852a3aa12f80d1d35ff270)
commit 6cbf30fa88 commit 4a5dece155
Author: Gerald Combs <gerald@wireshark.org> Author: Gerald Combs <gerald@wireshark.org>
Date: Sun Mar 14 10:32:22 2021 +0000 Date: Sun May 16 09:53:30 2021 +0000
[Automatic update for 2021-03-14] [Automatic update for 2021-05-16]
Update manuf, services enterprise numbers, translations, and other items. Update manuf, services enterprise numbers, translations, and other items.
commit 3d9603e051 commit 5254d5e822
Author: Guy Harris <gharris@sonic.net> Author: Gerald Combs <gerald@wireshark.org>
Date: Sun Mar 14 06:35:25 2021 +0000 Date: Tue May 11 10:56:10 2021 -0700
tap_export_pdu: free err_info for WTAP_ERR_INTERNAL.
WTAP_ERR_INTERNAL also returns an err_info script describing the
internal error, so it should be freed to avoid leaking it.
(cherry picked from commit d4c55ba464f2b6842a0a05fc623030b80322367a)
commit 7a7da50fc3
Author: Guy Harris <gharris@sonic.net>
Date: Sun Mar 14 06:00:21 2021 +0000
pcapng: make sure the packet encapsulation matches the interface encapsulati
If an attempt is made to write a packet with one encapsulation for an
interface with a different encapsulation, report an internal error.
(cherry picked from commit 610dffbb19532034309a0ee326128a71608ebe6a)
commit 4482949ddc
Author: John Thacker <johnthacker@gmail.com>
Date: Fri Mar 12 22:12:40 2021 -0500
export_pdu: set pkt_encap to a WTAP_ENCAP value, not a pcap LINKTYPE_ value. Tools: Try to exit more gracefully from fuzzing.
rec.rec_header.packet_header.pkt_encap expects WTAP encapsulation types, If we catch a signal while we're fuzzing, exit with a successful status
not PCAP encapsulation types, so don't call and clean up after ourselves.
Also, set the output file encapsulation, and the link-layer type of the (cherry picked from commit dacbfc4ae96fca71935e3f987dc3530f615a665f)
fake interface, to exp_pdu_tap_data->pkt_encap rather than wiring it to
WTAP_ENCAP_WIRESHARK_UPPER_PDU (we set exp_pdu_tap_data->pkt_encap to
WTAP_ENCAP_WIRESHARK_UPPER_PDU, but this is a bit cleaner).
(backported from commit ea60a57826624bcdf8f955af85af7130c8864018) commit b45185f1c0
Author: Gerald Combs <gerald@wireshark.org>
Date: Sun May 9 09:53:41 2021 +0000
commit 53a7984431 [Automatic update for 2021-05-09]
Author: Guy Harris <gharris@sonic.net>
Date: Sat Mar 13 23:32:14 2021 +0000
tshark: clean u the way the -U option lists available taps. Update manuf, services enterprise numbers, translations, and other items.
Allow "-U ?" as well as an empty argument; an empty argument is a bit commit 28aecbba66
counterintuitive. Author: Gerald Combs <gerald@wireshark.org>
Date: Sat May 8 09:30:33 2021 -0700
Simplify the introductory line of output - asking for a list of taps GitLab CI: Set our Clang version.
isn't an error in which the user failed to supply a tap name, it's a
case where the user suplied a request for a list of tap names.
Just use fprintf() to print the list, and indent the elements of the commit 728ae2de5f
list, as we do with other lists of valid arguments. Author: Gerald Combs <gerald@wireshark.org>
Date: Fri May 7 17:09:42 2021 -0700
List the valid arguments if the user specified an invalid argument as GitLab CI: Fix our fuzzing resource group.
(cherry picked from commit 9bd144b8ea30a7345eca3c0cb590bc146f45f97c) commit 2410d8b6eb
Author: Gerald Combs <gerald@wireshark.org>
Date: Fri May 7 12:43:16 2021 -0700
commit 392d01c2cc GitLab CI: Add fuzzing to the 3.4 branch.
Author: Guy Harris <gharris@sonic.net>
Date: Sat Mar 13 12:58:18 2021 +0000
tshark: fix handling of "you're writing to a closed pipe" errors on Windows. Backport part of d7bdd77a4c, which adds a -t option to
tools/fuzz-test.sh which lets you specify a maximum fuzz time.
On Windows, a write to a pipe where the read side has been closed Copy over the current (as of 7c6df3848f) fuzz jobs from the master
apparently may return the Windows error ERROR_BROKEN_PIPE, which the branch.
Visual Studio C library maps to EPIPE, or may return the Windows error
ERROR_NO_DATA, which the Visual Studio C library maps to EINVAL.
So, on Windows, for errors other than the ones for which we're reporting commit 66cba93e39
a special error message, check for EINVAL with a *Windows* error of Author: Gerald Combs <gerald@wireshark.org>
ERROR_NO_DATA and, if that's what we have, don't print an error message; Date: Fri May 7 12:48:31 2021 -0700
otherwise, print an error message that reports a message based on the
Windows error (rather than a relatively uninformative "Invalid argument"
This should fix issue #16192. GitLab CI: Give our jobs proper names.
Clean up indentation while we're at it. GitLab's job YAML parser allows spaces in key names and our jobs are
visible in the web UI, so give them proper names.
(cherry picked from commit 1eb4a62dbc190b0f0cfcf7b945bb0e82ddf2246c) commit c1cf92283c
Author: Gerald Combs <gerald@wireshark.org>
Date: Fri May 7 11:49:51 2021 -0700
commit 20a013b835 GitLab CI: Restore the ability to run pipelines from the web UI.
Author: Alexis La Goutte <alexis.lagoutte@gmail.com>
Date: Fri Mar 12 08:51:11 2021 +0000
NAN(WiFi): Fix wrong variable size (cherry picked from commit bedf0eb21c31b64c0bcfaa00f5632e13eabe8479)
NDPE Attribute is not dissected commit 219978c50d
Author: Gerald Combs <gerald@wireshark.org>
Date: Tue May 4 15:36:22 2021 -0700
Issue (and fix) reported by Darren Chen GitLab CI: Miscellaneous updates.
Closed #17278 Copy over various updates from the master branch including:
(cherry picked from commit 73de7b132559ad876f4be113b7356979a14730c2) - Setting a git clone depth.
- Running manually in forks.
- Rule reuse.
- Using "extends" instead of YAML anchors.
commit fbbaa3d03d commit b191395e0e
Author: Alexis La Goutte <alexis.lagoutte@gmail.com> Author: Gerald Combs <gerald@wireshark.org>
Date: Fri Mar 12 08:52:21 2021 +0000 Date: Sun May 2 09:36:02 2021 +0000
EAP: fix a memory leak [Automatic update for 2021-05-02]
(cherry picked from commit 20a6fea312342802cc76178cec673acf67ba7a26) Update manuf, services enterprise numbers, translations, and other items.
commit fd940a364c commit 347f60c37b
Author: Guy Harris <gharris@sonic.net> Author: Guy Harris <gharris@sonic.net>
Date: Thu Mar 11 02:11:02 2021 -0800 Date: Thu Apr 29 14:27:49 2021 -0700
pcapng: clean up pcapng_block_read().
Instead of pcapng_open() calling pcap_block_read() to do all the work of
reading the initial SHB, have it do the read of the initial SHB itself,
by calling the same routines that pcap_block_read() calls.
That way, pcap_block_read() doesn't have to be prepared to be called to tap_export_pdu: we don't have packet flags, don't claim that we do.
read that block, so it can treat all issues with an SHB that it reads as
errors, rather than possibly reporting them as "not a pcapng file", and
it doesn't have to support being called without a pointer to the
pcapng_t for the file being read, as it no longer ever is. It can now
just return a gboolean success/failure indication.
That makes pcapng_open() a little more complicated but it makes We don't set rec.rec_header.packet_header.pack_flags, so don't set
pcap_block_read() less complicated. WTAP_HAS_PACK_FLAGS in the presence flags. (Copy-and-pasteo?)
Fix some use of : as ' in comments, and fix some indentation, while (cherry picked from commit faf2e62db8b7cd3720ec472aa303761d6f97eaca)
we're at it.
(backported from commit aab0bd12a59a1f59e8e0f9cf8010b29e8092406c) commit ab13858b7f
Author: Uli Heilmeier <uh@heilmeier.eu>
commit 3a657cb4fb Date: Tue Apr 20 22:47:43 2021 +0200
Author: Dr. Lars Völker <lars.voelker@technica-engineering.de>
Date: Tue Mar 9 12:12:14 2021 +0000
TECMP: Adding missing reserved flag to timestamp (BUGFIX) PTP: Check for enough bytes
This patch fixes a bug in the current TECMP dissector that leads to Make sure we have enough bytes for Length and Type fields before we read
wrong timestamps, whenever the reserved flag is set to true. from tvb.
Using existing msg_len for the checks.
Closes: #17279 Closes: wireshark/wireshark#17355
(cherry picked from commit 5d709459c4728a460888ee8602f3990f17e91318) (cherry picked from commit fd14396972733e8e2e1023150f628e3aa7370b18)
commit e8421b1a0e commit aa169ab224
Author: Gerald Combs <gerald@wireshark.org> Author: Gerald Combs <gerald@wireshark.org>
Date: Wed Mar 10 13:15:15 2021 -0800 Date: Sun Apr 25 09:52:21 2021 +0000
GitLab CI: Add a `tshark -G` artifact.
Dump `tshark -G fields` to a file and add it as an artifact. This
automates a step in the release process.
commit 580812874a [Automatic update for 2021-04-25]
Author: Guy Harris <gharris@sonic.net>
Date: Wed Mar 10 20:55:50 2021 +0000
pcapng: get rid of the local pcapng_t in pcapng_open().
Pass a null pointer to pcapng_read_block(), instead. In
pcapng_read_block(), treat that as the indication that we're trying to
read the purported first SHB, rather than treating a null section_info
pointer as that indication.
This addreses one, but not all, of the problems reported in issue #17281.
(cherry picked from commit 7446a16247a8f04b1fa476e526cdaa97be00f32e) Update manuf, services enterprise numbers, translations, and other items.
commit 2f7cf6e528
Author: Martin Mathieson <martin.mathieson@keysight.com>
Date: Wed Mar 10 21:21:56 2021 +0000
DECT: "tranceiver" -> "transceiver"
(cherry picked from commit 933e7f5eb6798e72e349099be3cd16a890be7fc7)
commit 45aed9966c commit 31a3ec04a8
Author: Gerald Combs <gerald@wireshark.org> Author: Gerald Combs <gerald@wireshark.org>
Date: Wed Mar 10 12:23:46 2021 -0800 Date: Wed Apr 21 11:55:57 2021 -0700
Version: 3.4.4 → 3.4.5. Version: 3.4.5 → 3.4.6.
[skip ci] [skip ci]
 End of changes. 184 change blocks. 
548 lines changed or deleted 312 lines changed or added

Home  |  About  |  Features  |  All  |  Newest  |  Dox  |  Diffs  |  RSS Feeds  |  Screenshots  |  Comments  |  Imprint  |  Privacy  |  HTTP(S)