diff options
Diffstat (limited to 'ChangeLog')
| -rw-r--r-- | ChangeLog | 1679 |
1 files changed, 450 insertions, 1229 deletions
@@ -1,1506 +1,727 @@ -commit 7263aefcb0 -Author: Gerald Combs <gerald@wireshark.org> -Date: Mon May 13 15:55:35 2024 -0700 - - Prep for 4.2.5 - -commit e02689beb6 -Author: John Thacker <johnthacker@gmail.com> -Date: Mon May 13 22:06:46 2024 +0000 - - Qt: Fix RTP Player crash - - marker_stream_ might not exist when playFinished is called. - In particular, deleting the RtpAudioStreams might cause - playFinished to be called when the audio streams delete - their audio output in certain cases, such as if they are - paused. (The behavior might be slightly different on Qt6, - where QAudioSink is used vs. Qt5 where QAudioOutput is used.) - - Fix #19596 - - (cherry picked from commit 7698e03ba13e2e9966c84fbd800f8ab6c25983ef) - - Co-authored-by: John Thacker <johnthacker@gmail.com> - -commit 8be6839369 -Author: Gerald Combs <gerald@wireshark.org> -Date: Mon May 13 13:07:00 2024 -0700 - - GitLab CI: Fix a couple of separators +commit c1e402598d +Author: Anders Broman <a.broman58@gmail.com> +Date: Wed Jul 10 06:27:36 2024 +0000 -commit 01a8f70b09 -Author: John Thacker <johnthacker@gmail.com> -Date: Wed Oct 18 20:59:53 2023 -0400 + Thrift: fix incorrect proto_depth calculation - GitLab CI: Switch the openSUSE builds to 15.5 + In case a sub-dissector handles a command containing structures, the + current depth was not incremented for the content of the structure + itself and, more importantly, was improperly decremented when exiting + the structure. + In most cases, the bug could be hidden by the parent element correctly + resetting the value to its expected current state but when the structure + is the element type of a container (list, set, map), the sequence of + structures would decrement the count below 0, resulting in an underflow + that triggers the ei_thrift_too_many_subtypes expert info. - (cherry picked from commit a274d763df69d2a3d91f6478f6deb29c0c039952) + This issue does not affect the standard Thrift dissector that follows a + different path. -commit 1b9e1eccc8 -Author: Gerald Combs <gerald@wireshark.org> -Date: Mon May 13 12:41:37 2024 -0700 + (cherry picked from commit 4a9802621e3c3b523c509383d604bbb2677b0214) - GitLab CI: Quote a variable + Co-authored-by: Triton Circonflexe <triton@kumal.info> -commit c637058453 +commit f1bb1ea947 Author: Gerald Combs <gerald@wireshark.org> -Date: Mon May 13 12:01:53 2024 -0700 - - GitLab CI: Try to add back Lua on macOS +Date: Tue Jul 9 18:14:59 2024 -0700 -commit 9118420bb8 -Author: John Thacker <johnthacker@gmail.com> -Date: Mon May 13 14:52:45 2024 +0000 + Prep for 4.2.6 - Packaging: Get MinGW cross-compiling working with Qt 6.7 - - Qt 6.7 renamed QWindowsVistaStylePlugin to QModernWindowsStylePlugin. - We have an explicit mapping because windeployqt6 doesn't work well - with cross-compiling. Skip files that don't exist, since only one - of the Windows plugin files will exist. - - (cherry picked from commit 169392ce60ebf3d1fdc090ebf08ecdb313edb145) - - Co-authored-by: John Thacker <johnthacker@gmail.com> + [skip ci] -commit e93ac94572 -Author: John Thacker <johnthacker@gmail.com> -Date: Mon May 13 13:37:23 2024 +0000 +commit 27a738f3cc +Author: Uli Heilmeier <uh@heilmeier.eu> +Date: Tue Jul 9 16:25:11 2024 +0000 - wslua: Fix segfault when registering a field / ei twice + SIP: decoding for quoted value of utran-cell-id-3gpp - We need to call Proto_commit as a protected function so that - errors don't call wslua_panic and abort, but instead are handled - gracefully. + According to RFC7315 value for utran-cell-id-3gpp is token / quoted-string - Fix #19194 + Fixes: #19917 - (cherry picked from commit 70a3c9f82913d6c70ddc57d19976f80a676c9c56) + (cherry picked from commit aa69a037691a97c67a754b81e06f0e6d948b1f13) - Co-authored-by: John Thacker <johnthacker@gmail.com> + Co-authored-by: Uli Heilmeier <uh@heilmeier.eu> -commit 7f380a2a9d +commit 21691ee1c7 Author: John Thacker <johnthacker@gmail.com> -Date: Mon May 13 13:06:21 2024 +0000 +Date: Tue Jul 9 19:18:49 2024 +0000 - GitLab CI: Fix the Documentation job artifacts - - (cherry picked from commit 674e79c29e8144bb61ef4b3248be9fcb33c8dc9b) - - Co-authored-by: Gerald Combs <gerald@wireshark.org> + Shellcheck updates -commit 8c1ba39798 -Author: John Thacker <johnthacker@gmail.com> -Date: Mon May 13 12:35:29 2024 +0000 + Fix a few shellcheck warnings and add tools/rpm-setup.sh. - GitLab CI: Install ruby-dev in the Documentation job + [skip ci] - (cherry picked from commit 520fb7848ec713eadd9a0b149197d793789e9c52) + (cherry picked from commit e49d6dd7cb3b1bdc6c4f7e28cebe51d8a73df8d9) Co-authored-by: Gerald Combs <gerald@wireshark.org> -commit 235136144c -Author: John Thacker <johnthacker@gmail.com> -Date: Sun May 12 08:53:46 2024 -0400 - - Revert "Lua: Replace lua_pushnumber with lua_pushinteger where apropriate." - - This reverts commit 8cac5932f79b87deda26480769a266ff344036e6. - -commit f12b53ac0f -Author: John Thacker <johnthacker@gmail.com> -Date: Sun May 12 08:53:24 2024 -0400 - - Revert "wslua: Fix support for Lua 5.1 and 5.2 on 32bit" +commit 6637edf6d3 +Author: Gerald Combs <gerald@wireshark.org> +Date: Sat Jul 6 20:27:55 2024 -0500 - This reverts commit f8fd960cb8f5378c8ab6b20b4bc7f0b73498d057. + GitLab CI: Use Mino mc on Ubuntu -commit 4ed940da8c -Author: John Thacker <johnthacker@gmail.com> -Date: Sun May 12 08:53:11 2024 -0400 + Switch to minio mc on Ubuntu. Unlike `aws s3 cp` it can copy multiple + files. It should also reduce the Ubuntu image size a bit. - Revert "Lua: Revert a mistaken change to lua_pushinteger" + (cherry picked from commit d7eefd41a2b03bf72547901cd7e6b729f217793a) - This reverts commit f2dc894cb2e50668a3f80bfdffed1932f8e35459. + Conflicts: + .gitlab-ci.yml -commit a88ec0adc0 +commit 2e321b6357 Author: Gerald Combs <gerald@wireshark.org> -Date: Sun May 12 09:26:48 2024 +0000 +Date: Sun Jul 7 09:25:57 2024 +0000 - [Automatic update for 2024-05-12] + [Automatic update for 2024-07-07] Update manuf, services enterprise numbers, translations, and other items. -commit e8d6cf7c66 -Author: John Thacker <johnthacker@gmail.com> -Date: Sat May 11 13:12:25 2024 +0000 - - lua: Deregister heuristic lists added via heuristic_new on reload - - Heuristic dissectors lists don't necessarily (and don't usually) - have the name of a given dissector handle. +commit d4e7dfb50b +Author: Gerald Combs <gerald@wireshark.org> +Date: Sat Jul 6 22:17:58 2024 -0500 - There's still a leak when reloading plugins because the created - heuristic dissector list isn't returned to the caller. That also - means there's no way of knowing if it succeeded, and using the - heuristic list involves calling proto:register_heuristic with - the name, but rather than calling a function on a returned - object. + GitLab CI: Switch to clang 18 - Fix #19603 +commit 6606c20e92 +Author: Uli Heilmeier <uh@heilmeier.eu> +Date: Fri Jul 5 10:18:11 2024 +0000 - (backported from commit 2b2b032af2bdb33345534db33445a2aad24c6217) + DCHP: Fix Relay Agent Information Source Port -commit 0711fa7b1e -Author: mistral13 <mistral@gmx.ch> -Date: Wed Oct 11 22:17:50 2023 +0200 + Fixes: #19909 - btmesh: Various fixes from master + (cherry picked from commit acd8dca562a07a6811788617bdca03f4f3a27753) - 1. Register the company ID as big endian. - 2. The pinfo column of the own (lua-)dissector was overwritten - by the btmesh dissector. - 3. Output of the own (lua-)dissector as root-tree element. + Co-authored-by: Uli Heilmeier <uh@heilmeier.eu> - Fixed some code indent while already visiting this file. +commit ccfaadf39f +Author: Gerald Combs <gerald@wireshark.org> +Date: Sun Jun 30 09:25:42 2024 +0000 - Ping #19388 + [Automatic update for 2024-06-30] - (cherry picked from commit d3fab19401caf09e2e65047d2843db9d3caf2131) - (cherry picked from commit 50f322ec1702f4ce0e2d94a19509a82b3fb3df24) - (cherry picked from commit 38d05ad97a4a2158b834f973a5b6a034c78edc7b) - (cherry picked from commit a4527fa0418605318c8986bc8947a606e72043a9) + Update manuf, services enterprise numbers, translations, and other items. -commit 949aa570f3 +commit d29df12f71 Author: John Thacker <johnthacker@gmail.com> -Date: Fri May 10 10:42:03 2024 -0400 +Date: Fri Jun 28 11:14:09 2024 +0000 - Lua: Fix DissectorTable on certain Lua versions (e.g., 32 bit) + Qt: Keep axes in sync for Window Scaling graph - The port value is a uint32_t. Calling checkinteger and implicitly - casting to an unsigned integer later does not give the correct - result on a Lua install where Lua_Integer is 32 bit for unsigned - integers greater than INT32_MAX. Lua will truncate to the integer - range (sometimes by using inline assembly, sometimes explicitly.) + The unacked bytes makes sense as a percentage of the window size, + so keep the two axes in sync, while one axis has the color and + title for the unacked bytes, while the other has the color and + title of the window size. - Note that Lua_Integer can 32 bit on 64 bit platforms; it can also - be 64 bit on 32 bit platforms. + Fixup 259bbbff2a5257319ff69aa32cc377fc501eb022 - Fix #18367 + (cherry picked from commit ac81cdbc92b62f04f979ffbe40679fae7f948c7a) - (backported from commit 4049ea7e6c4009158cbf91b0bf3300759d2106d6) + Co-authored-by: John Thacker <johnthacker@gmail.com> -commit f2dc894cb2 +commit f97cd68ea4 Author: John Thacker <johnthacker@gmail.com> -Date: Fri May 10 10:08:00 2024 -0400 - - Lua: Revert a mistaken change to lua_pushinteger - - The value read from a file here is allowed to be a floating point number, - it should not be pushed as a Lua_Integer. - - (backported from commit 1428fe28e663c8c6057d583d336b84014f169de1) +Date: Wed Jun 26 11:01:44 2024 +0000 -commit 077104f6dc -Author: Gerald Combs <gerald@wireshark.org> -Date: Wed May 8 14:17:01 2024 -0700 - - Windows: Upgrade GnuTLS to 3.8.4 - - (cherry picked from commit b119b9d6010f7ec0210f83297df365c2e1dec1ae) + idl2wrs.c: Fix generation of idl2wrs-dissectors -commit 6ab3e8ed88 -Author: John Thacker <johnthacker@gmail.com> -Date: Thu May 9 14:08:33 2024 +0000 + This hasn't worked since 8ca846436d14da6079a51da4111a860e0f357c36 + (September 24, 2016) because the code actually depended on atoi + failing silently and returning 0 when faced with a non integer. - GitLab CI: Remove LUA_FIND_VERSIONS ANY + Switch the ws_strtou32 call to being *after* the checks for the + special values '*' and ']'. - That was added briefly but support removed in - commit 759c0feab351de2b20946ececf2949ab52592ccd and it - results in Lua not being found. + We don't need to copy the result into a string buffer and compare, + because ws_strtou32 with a NULL end of string pointer will err in + the cases where the comparison would fail. - (cherry picked from commit eccf1104d6e310eaf78a0bf063425179c81a7150) + (cherry picked from commit f3f19e8fbb972a49fa3d141e59ef3d9a007a24aa) Co-authored-by: John Thacker <johnthacker@gmail.com> -commit 38c5a89ac4 -Author: Gerald Combs <gerald@wireshark.org> -Date: Tue May 7 16:48:37 2024 -0700 - - Windows: Upgrade nghttp2 to 1.61.0 +commit ad95284883 +Author: Guy Harris <gharris@sonic.net> +Date: Wed Jun 26 05:58:34 2024 +0000 - (cherry picked from commit fe51f8424f9a5dcae514745b919b6139c3c568fd) + prefs: Add one space back to gui.debounce_timer description -commit 5e96f900a9 -Author: John Thacker <johnthacker@gmail.com> -Date: Tue May 7 06:38:31 2024 -0400 + The previous fix accidentally missed adding one trailing space in + a line. - gitlab-ci: Remove docker tag from jobs + Fixes: fe98af0e30 ("prefs: Get rid of line continuations for gui.debounce.timer description") - The small SaaS runner on Linux had the docker (plus all other tags) - removed, causing all the jobs with that tag to get stuck. + (cherry picked from commit 2f19d0db5c22635171f07c4bd1858ae5ba9e9609) - https://about.gitlab.com/blog/2023/08/15/removing-tags-from-small-saas-runner-on-linux/ - https://docs.gitlab.com/ee/update/deprecations.html#removal-of-tags-from-small-saas-runners-on-linux - (backported from commit a0e399e5c3edd06cfc78fe41f8959f4c556045be) + Co-authored-by: Paul Aurich <paul@darkrain42.org> -commit 94ac8729f3 +commit 687213a9f4 Author: John Thacker <johnthacker@gmail.com> -Date: Sun May 5 13:59:27 2024 +0000 - - MATE: Handle fields that are in different data sources - - In addition to the start and end offset locations, store a pointer to - the data source tvb in each mate_range. The start and end offsets - are only relevant within a data source. - - If a field has a data source different from one of the protocol, - transport protocol, or payload ranges, search in the tree for the - ancestor nodes of the field, and see if an ancestor is located within - one of the ranges. - - In order to workaround #17877 (non-visible items can't change length - after being added to the tree, which affects most protocols), set - the tree as visible similar to done with a number of Lua postdissectors - that need all fields. Unfortunately this is overkill that hurts - performance. - - Fix #19619 - - (cherry picked from commit 4ec5c7f78f9af35b2cebe92c4dd7eab033a1aa82) - - Co-authored-by: John Thacker <johnthacker@gmail.com> - -commit 48c0364590 -Author: Gerald Combs <gerald@wireshark.org> -Date: Sun May 5 09:25:22 2024 +0000 - - [Automatic update for 2024-05-05] +Date: Tue Jun 25 18:46:42 2024 +0000 - Update manuf, services enterprise numbers, translations, and other items. + WiSUN: correct offset logic errors -commit cb267b4e52 -Author: Martin Nyhus <martin@nyhus.dev> -Date: Wed May 1 17:17:06 2024 +0200 + Change local variable `offset` from a `guint8` to a `guint` in two + functions with loops, to prevent a potential infinite loop from + overflowing the variable. - ZigBee TLV: fix infinite loop in deeply nested packets + Change `dissect_wisun_nrie()` to return the number of bytes it + dissected, since that's how `dissect_wisun_hie()` seemingly expects it + to behave. It previously returned the new offset at the end of what it + dissected -- which is how several `dissect_...()` functions in this + dissector are expected to behave, but this is not one of them. (This + inconsistency adds friction when maintaining this dissector and should + be addressed at some point.) - When analyzing a packet that hits the recursion limit returning the - remaining length of the buffer would allow backtracking, setting up an - infinite loop if the packet is constructed correctly and the code would - fail by hitting the too many items check. + (cherry picked from commit 6931eb67e7a3fd7a2ad115f3523fb0a4778c885f) - Since dissect_zbee_tlvs doesn't know the length of the value it is - dissecting it can't pretend to have consumed all of it. Most of the - callers of this function eventually check for bytes that weren't - consumed so returning offset (i.e. no bytes consumed) makes the value - get treated as unknown bytes. + Co-authored-by: David Perry <boolean263@protonmail.com> - Input to fuzzshark_zbee_beacon that triggered this: - fb 03 49 ff 49 eb 49 ff 49 ff 49 eb 49 ff +commit dbad9729ec +Author: Peter Wu <peter@lekensteyn.nl> +Date: Tue Jun 25 10:33:01 2024 +0000 - (cherry picked from commit 9ab952b9641bb07d796499487e8d8691d52a6902) + Qt: fix unreadable syntax filter in dark mode on KDE with Qt 6 -commit 37c53eab86 -Author: John Thacker <johnthacker@gmail.com> -Date: Thu May 2 23:29:30 2024 +0000 + KDE Plasma 6.1.0 with the Breeze Dark theme on Qt 6.7.1 and Windows 11 + with a high contrast theme have an unreadable display filter field. It + displays white text on a bright green or red background, see #18246. - Qt: Fix an typo of row vs count in columnListModel + This was caused by an invalid assumption that `colorScheme()` only + returns two values, Light and Black. It can also return Unknown which + does not imply Light. Make sure to fallback and check the actual colors. - The size of the internal store is the number of model rows (which - are the packet list columns), not the number of model columns (which - are the parameter types for the packet list columns). This caused data - not to appear for the later parameter types when the number of packet - list columns were small. More of a difference in the current branch - because width and alignment are also shown in preferences, so it - has effects at the default number of columns. + Link: https://gitlab.com/wireshark/wireshark/-/issues/18246#note_1958637216 + Link: https://code.qt.io/cgit/qt/qtbase.git/tree/src/gui/platform/unix/qgenericunixthemes.cpp?h=6.7.2#n1132 + Link: https://code.qt.io/cgit/qt/qtbase.git/tree/src/plugins/platforms/windows/qwindowstheme.cpp?h=6.7.2#n549 + Fixes: v4.1.1rc0-163-g0573bbaed6 ("Qt: Dark mode updates") - Related to #19821 (but not sufficient.) + (cherry picked from commit 333adde36f5d614176c3a44c7d3357af67da451c) - (cherry picked from commit 704fe87abd1e83a3451ba4abb2001ef7349d5c71) + Co-authored-by: Peter Wu <peter@lekensteyn.nl> - Co-authored-by: John Thacker <johnthacker@gmail.com> +commit 0cb98e630b +Author: Guy Harris <gharris@sonic.net> +Date: Tue Jun 25 04:24:18 2024 +0000 -commit a338e8117c -Author: John Thacker <johnthacker@gmail.com> -Date: Wed May 1 11:41:24 2024 +0000 + prefs: Get rid of line continuations for gui.debounce.timer description - LAPD: Set source address + Due to the use of line continuations here, the preference description written + into the wireshark preferences file ends up with excessive whitespace. - The destination and source address are both being set to the - destination value. This causes the normal all source Flow Graph - (as opposed to the one created by the VoIP Calls Dialog) to have - empty lines. + Before: + ``` + % tshark -G defaultprefs | grep -B2 debounce + # How long to wait (in milliseconds) before processing computationally intensive user input. If you type quickly, consider lowering the value for a 'snappier' experience. If you type slowly, consider increasing the value to avoid performance issues. This is currently used to delay searches in View -> Internals -> Supported Protocols and Preferences -> Advanced menu. + # A decimal number + #gui.debounce.timer: 400 + % + ``` - Set the dl_src and dl_dst addresses using the AT_STRINGZ address - instead of col_set_str so that it works for both the Resolved and - Unresolved columns, and set the src and dst addresses as shallow - copies of those. + After: + ``` + % ./run/tshark -G defaultprefs | grep -B2 debounce + # How long to wait (in milliseconds) before processing computationally intensive user input. If you type quickly, consider lowering the value for a 'snappier' experience. If you type slowly, consider increasing the value to avoid performance issues.This is currently used to delay searches in View -> Internals -> Supported Protocols and Preferences -> Advanced menu. + # A decimal number + #gui.debounce.timer: 400 + % + ``` - Fixup 7ac827fd74cf7b1eac80d54b5ce0001f15a587e3 + Fixes: 9e1905f88d ("Preferences: Support configuring debounce timers") - (cherry picked from commit 1da57b5ad536acd1d138d363825bc7aad7ad1f4b) + (cherry picked from commit fe98af0e300c83d7b87391628e3c1a4176ab6c4a) - Co-authored-by: John Thacker <johnthacker@gmail.com> + Co-authored-by: Paul Aurich <paul@darkrain42.org> -commit 8b916c603e +commit e081baf9ec Author: John Thacker <johnthacker@gmail.com> -Date: Tue Apr 30 23:20:17 2024 +0000 +Date: Mon Jun 24 19:30:50 2024 +0000 - Qt: Fix ProgressFrame location in non-VoIP Flow Graph + Qt: Write gui.fileopen_remembered_dir to recent with native separators - ProgressFrame::addToButtonBox examines the button box layout to - determine where to add itself. Changing the visibility of the - RTP player button redoes the layout and upsets the positioning - of the frame. This leaves the ProgressFrame dangling in the far - left of the ButtonBox, on top of the Help button. + QFileInfo().absolutePath() always uses '/' as the directory separator. + When writing the path to the recent file, use native separators (i.e. + '\' on Windows.) This affects things more when using the native Win32 + API (so 4.2 and earlier), but also makes the recent file and certain + error and warning pop-ups look normal to Windows users. - SequenceDialog only determines whether it's a VoIP Calls dialog or - a non-VoIP Flow Graph when initially instantiated (there's no call - to disable it), so add the VoIP parameter to the constructor and - set the visibility appropriately before adding the ProgressFrame - instead of having the VoIP Calls Dialog call an extra public method. + Fix #19891 - (cherry picked from commit a95dec6ef9612d80598547a346825639e16c4e51) + (cherry picked from commit b14ceeb6b6116b8858056f9322954f880b79b9ae) Co-authored-by: John Thacker <johnthacker@gmail.com> -commit a862720a7e +commit ac51b76cad Author: John Thacker <johnthacker@gmail.com> -Date: Tue Apr 30 22:41:28 2024 +0000 - - Qt: Use afterLayout signal for Flow Graph y axis labels +Date: Mon Jun 24 13:52:32 2024 +0000 - For the Sequence Dialog (Flow Graph / VOIP Calls), we place - text labels for the two y-axes (Time and Comment) above the - axes, which is not a normally supported position, so we - manually place it. The position of the labels depends on the - axis rectangle sizes and positions. + idl2wrs: Update generator to match recent commit - QCustomPlot 2.1.0 added a new QCustomPlot::afterLayout signal - that is extremely helpful for situations like this. It is - called immediately before the draw step in a replot, or before - printing happens, and can put the labels in the correct position - after any change. + Update wireshark_gen.py for the change made in commit + fd64746f44899d6685b959507c85867a03c54537 so it produces the same result. - This is a little more reliable than our previous solution, which - sometimes left the labels out of position, needing to be fixed - by manually pushing the Reset button. + We still need to regenerate the dissectors after the changes in + commit 32c2ab79a778f88f0f4b852034feada0345fe099. - (cherry picked from commit 84fe646b78d62dcc5540d7a2950a06b0f86de0e7) + (cherry picked from commit 7a69039fbb52bd55e86aa1100cf254fd3d07524e) Co-authored-by: John Thacker <johnthacker@gmail.com> -commit 51a925ce40 -Author: John Thacker <johnthacker@gmail.com> -Date: Mon Apr 29 01:45:43 2024 +0000 - - ptp: multiple tlv_pathsequence in announce message - - (cherry picked from commit 7ef8501c347d4336a1cb62b4b312f8c93b92efff) - - 32b82d47 Update file packet-ptp.c - 746c18cf ptp: multiple tlv_pathsequence in announce message - - Co-authored-by: Chuck Craft <bubbasnmp@gmail.com> - -commit 42e3b84e31 -Author: John Thacker <johnthacker@gmail.com> -Date: Sun Apr 28 21:16:17 2024 +0000 - - GitLab CI: Fix the Documentation job - - Remove a duplicate `cd`. Use the same pattern for the guides when - zipping and copying. +commit 730fa7a38b +Author: Gerald Combs <gerald@wireshark.org> +Date: Sun Jun 23 09:26:15 2024 +0000 - (cherry picked from commit 4f8f52dea0db04dfefaefcba47e8ba16b2d29e2e) + [Automatic update for 2024-06-23] - Co-authored-by: Gerald Combs <gerald@wireshark.org> + Update manuf, services enterprise numbers, translations, and other items. -commit d705e804ed +commit 11b71ab3f7 Author: John Thacker <johnthacker@gmail.com> -Date: Sun Apr 28 18:40:52 2024 +0000 +Date: Wed Jun 19 22:13:49 2024 +0000 - ipars: Fix truncation warning + radius: Switch back to the previous FILE if an $INCLUDE doesn't open - MAX_EOM_MSG_SIZE is defined to be too small for the value placed in - for an unknown type. + Save the previous FILE pointer so we can switch back to it if a + file name listed in an $INCLUDE doesn't open, instead of leaving + it null, so that any subsequent read attempts (or attempts to close + it at the end) don't crash. - epan/dissectors/packet-ipars.c:130:29: error: 'snprintf' will always be truncated; specified size is 16, but format string expands to at least 24 [-Werror,-Wformat-truncation] - 130 | default: snprintf(eom_msg, MAX_EOM_MSG_SIZE, "Unknown EOM type (0x%2.2X)", ia); break; + Fix #6466. (The other parts of the issue have been fixed.) - (cherry picked from commit 4320031c97b3528bc96ab5d366bab8b7fdb2a243) + (cherry picked from commit ef37a3e41e32f96d19d684e0873b51a31069184b) Co-authored-by: John Thacker <johnthacker@gmail.com> -commit f2b9b7b582 +commit cb890416db Author: John Thacker <johnthacker@gmail.com> -Date: Sun Apr 28 17:38:21 2024 +0000 - - c-ares: Suppress deprecation warnings - - c-ares 1.28.0 emits deprecation warnings by default for certain - functions. We will try to move away from those functions, but - suppress the warnings by default. +Date: Wed Jun 19 12:52:09 2024 -0400 - It's possible for recent versions of c-ares to be built without - thread safety and threading support, and it can only be checked - at runtime. We will probably have to fallback to using deprecated - functions on recent versions of c-ares if threading is not available - even if and when we switch to using event threads, so disable the - warning. + release notes: Mention fix to TCP Stream Graph tcptrace - Ping #19763 + Related to #19846 - (cherry picked from commit e1a6557d34ff45073e3e75923bf3f6521ebc2b16) - - Co-authored-by: John Thacker <johnthacker@gmail.com> - -commit 4dadd9b029 -Author: Dan <DanOfTheRoses@gmail.com> -Date: Fri Apr 26 21:49:40 2024 -0500 - - wslua: Fix integer validation signedness - - The C Standard (any revision) 6.3.1.4 Real floating and integer, - in a footnote says: - "The remaindering operation performed when a value of integer type is - converted to unsigned type need not be performed when a value of real - floating type is converted to unsigned type." - - Since a Lua_Number is a float type, converting a (possibly negative) - float to an unsigned type and then to a signed type is undefined - behvavior, and doesn't necessarily produce the same result as converting - the float to a signed integer type directly. On x86 and x86-64 it - has the same result, but it's permissible to instead first cast the float - to the nearest unsigned integer value (i.e. 0), so that the end result - is 0 instead on negative. - - On the ARM architecture this is what seems to happen: - - https://embeddeduse.com/2013/08/25/casting-a-negative-float-to-an-unsigned-int/ - https://github.com/mavlink/mavlink/issues/2073#issuecomment-2080103820 - - (Compilers can warn about this, but it's perhaps difficult to get the - only the right subset of conversion warnings.) - - Therefore, when adding a signed integer field, we must use - wslua_checkgint32. - - (backported from commit 15392c324d5eaefcaa298cdee09cd5b40b12e09c) - -commit af5546f31d +commit 2c0e45eb92 Author: Gerald Combs <gerald@wireshark.org> -Date: Sun Apr 28 09:25:05 2024 +0000 +Date: Sun Jun 16 09:26:25 2024 +0000 - [Automatic update for 2024-04-28] + [Automatic update for 2024-06-16] Update manuf, services enterprise numbers, translations, and other items. -commit e63cbbb720 -Author: John Thacker <johnthacker@gmail.com> -Date: Fri Apr 26 13:15:27 2024 +0000 - - 5co-rap: Use ITEM_LABEL_LENGTH for snprintf in CF_FUNCs - - At least one case cases a format truncation warning in recent - versions of clang. +commit c1885eb7a5 +Author: Guy Harris <gharris@sonic.net> +Date: Sat Jun 15 07:13:39 2024 +0000 - (cherry picked from commit f1dd3bf4f2c95571e854d3e35536021ba9aa053d) + log3gpp: Fix read of uninitialized data. - Co-authored-by: John Thacker <johnthacker@gmail.com> + Clang static analysis reports that subsecond_decimals_buf[1] through [3] might + be used uninitialized. This is because the array is being NUL-terminated, but + the NUL-termination is not relevant (and is in fact erroneous) when the array + is being directly evaluated one character at a time and assumed to be only + digits. -commit 43225b77d8 -Author: John Thacker <johnthacker@gmail.com> -Date: Fri Apr 26 11:35:25 2024 +0000 + The NUL-termination can be removed (and the array size reduced by one) and + instead we directly initialize to '0' any characters not read from the source + line. - 5co-legacy: Use ITEM_LABEL_LENGTH for snprintf in CF_FUNCs + (cherry picked from commit 6ead3e26572ebc5be814206f4a5a833866eb4f32) - In particular, one case leads to a format truncation warning on - recent versions of clang. + Co-authored-by: Darius Davis <darius-wireshark@free-range.com.au> - (cherry picked from commit 2056120ddce063850acb5cfd207dce724093c83e) +commit f30a10ae2f +Author: Niels Widger <niels@qacafe.com> +Date: Fri Jun 14 12:33:51 2024 +0000 - Co-authored-by: John Thacker <johnthacker@gmail.com> + capinfos: Fix printing packet comment headers with -TM -commit 21d3008eb0 -Author: Anders Broman <a.broman58@gmail.com> -Date: Sun Apr 21 19:42:46 2024 +0200 + Commit 45772de9 added support for printing packet comments to + `capinfos` with new flags `-p`/`-P` as well as including them in the + machine-readable `-M` and table report machine-readable `-TM` outputs. + Unfortunately, the commit contained a bug when `-TM` was specified + causing none of the `Packet %d Comment` headers for packet comments + being printed. This made the output decidedly less machine-readable. + This commit fixes this bug, ensuring that `-TM` output includes the + correct headers for captures containing packet comments. - PER: When showing internal bitfields handle integer values. + Previously, in `print_stats_table_header` we used `g_malloc0` to + allocate a buffer `buf` large enough to hold the `Packet %d Comment` + header for any packet with a frame number from 1 to 2^64. + Unfortunately, since this buffer was initialized with `NULL` bytes, + using `strlen(buf)` as the `n` argument to `snprintf` when writing the + `Packet %d Comment` header to `buf` always led to it writing 0 bytes + to `buf`. This is turn led `print_stats_table_header_label` to always + get an empty string as its argument, meaning no `Packet %d Comment` + header was printed. The result of this bug was that the actual packet + comment values were printed, but without the accompanying `Packet %d + Comment` header. - (cherry picked from commit f0462182da63155ffde38d51da7dbedc9fd2d0a6) + Below is an example of the previous, broken behavior: -commit ae4f2529db -Author: Niels Widger <niels@qacafe.com> -Date: Tue Apr 23 12:27:43 2024 +0000 + $ capinfos -TM comment-test.pcapng + File name File type File encapsulation File time precision Packet size limit Packet size limit min (inferred) Packet size limit max (inferred) Number of packets File size (bytes) Data size (bytes) Capture duration (seconds) Start time End time Data byte rate (bytes/sec) Data bit rate (bits/sec) Average packet size (bytes) Average packet rate (packets/sec) SHA256 SHA1 Strict time order Capture hardware Capture oper-sys Capture application Capture comment + /root/comment-test.pcapng pcapng ether microseconds (not set) n/a n/a 19 3944 2990 2.660439 2022-12-24 17:31:05.387079 2022-12-24 17:31:08.047518 1123.87 8991.00 157.37 7.14 e3ab39da3be87894a58e622dd08505b94f37bc5e478099037dcd03333a87c8b3 5afe04e6f4d5f485aad8b38ee42c54ee4fd6256e True Editcap (Wireshark) 3.6.19 (Git commit c38a01f797b1) Frame two comment Frame 6 has a comment Frame 10 with _markdown_ **markdown**\r\n\r\n## Headers too\r\n\r\n[and a link!](https://www.google.com) Frame 14 comment multiline\n\nHello there! - sharkd: Fix SIP follow filter returned in "frame" command + and here is an example of the new, fixed behavior: - `sip_follow_conv_filter` uses its `edt` argument to determine the - `sip.Call-ID == "<call-id>"` value to return as the conversation - filter. If `edt` is `NULL`, a fallback filter of `sip.Call-ID` is - returned. + $ capinfos -TM comment-test.pcapng + File name File type File encapsulation File time precision Packet size limit Packet size limit min (inferred) Packet size limit max (inferred) Number of packets File size (bytes) Data size (bytes) Capture duration (seconds) Start time End time Data byte rate (bytes/sec) Data bit rate (bits/sec) Average packet size (bytes) Average packet rate (packets/sec) SHA256 SHA1 Strict time order Capture hardware Capture oper-sys Capture application Capture comment Packet 2 Comment Packet 6 Comment Packet 10 Comment Packet 14 Comment + /home/niels/Downloads/comment-test.pcapng pcapng ether microseconds (not set) n/a n/a 19 3944 2990 2.660439 2022-12-24 12:31:05.387079 2022-12-24 12:31:08.047518 1123.87 8991.00 157.37 7.14 e3ab39da3be87894a58e622dd08505b94f37bc5e478099037dcd03333a87c8b3 5afe04e6f4d5f485aad8b38ee42c54ee4fd6256e True Editcap (Wireshark) 3.6.19 (Git commit c38a01f797b1) Frame two comment Frame 6 has a comment Frame 10 with _markdown_ **markdown**\r\n\r\n## Headers too\r\n\r\n[and a link!](https://www.google.com) Frame 14 comment multiline\n\nHello there! - `sharkd`'s `frame` method returns the SIP filter returned by - `sip_follow_conv_filter` for SIP conversations in a `filter` field of - its `fol` object array. These fields are emitted by - `sharkd_follower_visit_layers_cb`, which uses `get_follow_conv_func` - to retrieve `sip_follow_conv_filter`. Previously, - `sharkd_follower_visit_layers_cb` always passed `NULL` as the `edt` - argument to the function returned by `get_follow_conv_func`, therefore - `sip_follow_conv_filter` always saw a `edt=NULL` and was not able to - output a specific `sip.Call-ID` filter, even if one was otherwise be - available. + (cherry picked from commit 5a94fd5022d9a6100090fb0f22de99801cfc9101) - This commit updates `sharkd_follower_visit_layers_cb` to instead pass - the `edt` argument that is available as the `edt` argument to its - caller, `sharkd_session_process_frame_cb`. + Co-authored-by: Niels Widger <niels@qacafe.com> - (cherry picked from commit 9778cc82207520547e22c39f11ca3c1ac52c8aea) +commit 83b9ca9263 +Author: Gerald Combs <gerald@wireshark.org> +Date: Sun Jun 9 09:25:58 2024 +0000 - Co-authored-by: Niels Widger <niels@qacafe.com> + [Automatic update for 2024-06-09] -commit f8fd960cb8 -Author: Balint Reczey <balint@balintreczey.hu> -Date: Sat Apr 20 18:57:17 2024 +0200 + Update manuf, services enterprise numbers, translations, and other items. - wslua: Fix support for Lua 5.1 and 5.2 on 32bit +commit 6efb981fb7 +Author: Guy Harris <gharris@sonic.net> +Date: Thu Jun 6 17:28:53 2024 +0000 - Add macros to use lua_pushnumber for Lua 5.1/5.2 and lua_pushinteger for Lua 5.3/5.4 - based on change proposal by Stig Bjørlykke. + mergecap: prevent appending to an input file - Ping #10881 + Prevent the infinite loop described in #19402 by refusing to operate if + asked to append to an output file that is also one of the inputs. Uses + `files_identical()` from wsutil. - Follow up for commit 8cac5932f79b87deda26480769a266ff344036e6. + (cherry picked from commit 3b63098fb1e2b7f8fe350989bb973d8323464639) - Co-authored-by: Stig Bjørlykke <stig@bjorlykke.org> + Co-authored-by: David Perry <boolean263@protonmail.com> -commit 91e4bbf5cb +commit 52084eab98 Author: John Thacker <johnthacker@gmail.com> -Date: Fri Apr 26 01:26:04 2024 +0000 +Date: Thu Jun 6 01:08:10 2024 +0000 - Qt: Fix flow graph scroll direction via keyboard + Qt: Allow the Learn links to word wrap - The y (time) axis is range reversed. QCP reversed ranges still - have the mathematically smaller element in lower and the larger - in upper, so reverse the direction of movement in panAxes. + Over the course of the 4.0 series the Sharkfest, Discord, and Donation + links were added to the Welcome page. They are now the limiting factor + in the application page width. For some language translations they + are even longer. - Also fix the keyboard right arrow allowing the graph to be scrolled - to right farther than the QScrollBar allows, into extra empty space. + Some of the "Learn" links have multiple words, like "Questions and + Answers," or "Wireshark Discord." Allow the multiword link titles + to word wrap. - Fix #12932 + Fix #19861 - (cherry picked from commit cc4c931b4a3a3d481aeb9ea320e5fb5e31ca0f70) + (cherry picked from commit 389151f569600f5340845879f7d3a5b783b6ea57) Co-authored-by: John Thacker <johnthacker@gmail.com> -commit 4e70db8cf0 -Author: Yaniv Michael Kaul <yaniv.kaul@scylladb.com> -Date: Thu Apr 25 16:46:21 2024 +0300 - - Release-4.2: CQL: take into account NULL and unset values - - From https://gitlab.com/wireshark/wireshark/-/commit/291822a25cfaf54a3556c19d5d47af7efb1bdf48 to stable branch. - - Signed-off-by: Yaniv Kaul <mykaul@gmail.com> - -commit a4d779fbbb -Author: Yaniv Michael Kaul <yaniv.kaul@scylladb.com> -Date: Thu Apr 25 16:28:19 2024 +0300 - - Release-4.2: CQL: fix show paging state first - - Brings to release 4.2 a single fix from https://gitlab.com/wireshark/wireshark/-/commit/161b9809b17c394e025453b5accdd9733eee3ee7 : - If Has_more_pages flag is set, then paging_state field should be the first to apper in the metadata, before the (optional) global_table_spec - - Signed-off-by: Yaniv Kaul <mykaul@gmail.com> - -commit c193bcea08 +commit d64dcbe0d6 Author: John Thacker <johnthacker@gmail.com> -Date: Thu Apr 25 15:17:47 2024 +0000 +Date: Thu Jun 6 00:18:46 2024 +0000 - Qt: Fix Window scaling graph axis labels + E212: Fix a duplicate entry in mnc value_string - Commit 259a3e508f37bb3c9b6d18cdacf7a18aa6dbd8f8 added a graph - of the congestion window to the Window Scale graph, but didn't - update the axes now that the window has two graphs. + (cherry picked from commit 19405895f6c9602002b9eed0a3884765018b019f) - Fix #17425. Fix #15016 + ac6ab8cf E212: Fix a duplicate entry in mnc value_string - (cherry picked from commit 259bbbff2a5257319ff69aa32cc377fc501eb022) + Co-authored-by: Martin Mathieson <martin.r.mathieson@googlemail.com> - Co-authored-by: John Thacker <johnthacker@gmail.com> - -commit 47199b51b9 +commit eca87127fd Author: John Thacker <johnthacker@gmail.com> -Date: Thu Apr 25 14:39:26 2024 +0000 - - TLS: SM3 and SM2 are not in TLS (Signature|Hash)Algorithm registries +Date: Mon Jun 3 20:05:25 2024 -0400 - These two value strings are only for displaying the contents of the - two octets contained in a signature_algorithms extension as a TLS 1.2 - SignatureAndHashAlgorithm. (As opposed to displaying it as a TLS 1.3 - SignatureScheme where the two octets do not necessarily separately - indicate signature and hash.) + Qt: Fix byte view text wrapping with Qt <= 5.10 - SM3 and SM2 are not assigned in those registries, are not supported in - TLS 1.2 (RFC 8998), and if they were would not receive the values chosen - here. + There's at least two concepts that might be described as the width + of a string of characters; the width of a box that surrounds the + existing characters, and the width that measures where to place + the next characer. The former has always been obtained from + QFontMetrics::boundingRect(text).width(), but prior to Qt 5.11 + QFontMetrics::width(text) gave the latter. This led to many people + using the latter when they usually wanted the former. - Add a comment explaining why even though RFC 8422 registered ED25519 - and ED448 to the TLS 1.2 SignatureAlgorithm and HashAlgorithm registries - we don't add it to the value strings for those. (It would take more - complicated processing to avoid confusion, and for little gain.) + Thus in Qt 5.11 QFontMetrics::width() was deprecated for + QFontMetrics::horizontalAdvance(). - Fix #19801 + In our case, we really want horizontalAdvance, because we're manually + drawing characters. It makes a difference because for non ASCII + characters we use the UTF-8 middle dot character, which has a lot + of necessary blank space before and after it where other characters + shouldn't be drawn. With some fonts and scalings using the width + of the bounding rectangle makes our lines too short and the end of + the ASCII text wraps, when there's lots of non ASCII / middle dots. - (cherry picked from commit ea5ed37426c9489b66ea32cec818d68c584a5a85) + So on Qt 5.10 and earlier, use the width() function that is + equivalent to horizontalAdvance (een though it is marked obsolete + for later versions.) - Co-authored-by: John Thacker <johnthacker@gmail.com> + Fix #18096 -commit dc29e276dd +commit eeab1dbeaf Author: John Thacker <johnthacker@gmail.com> -Date: Thu Apr 25 11:04:41 2024 +0000 - - E212: update MNC information +Date: Mon Jun 3 13:03:28 2024 +0000 - Update MNC information from the 15 November 2013 release of - Mobile Network Codes (MNC) for the international identification - plan for public networks and subscriptions - (According to Recommendation ITU-T E.212 (09/2016)) + Qt: Make it possible to clear Comments from all interfaces - https://www.itu.int/pub/T-SP-E.212B-2023 - https://www.itu.int/pub/T-SP-OB.1280 + We only add an entry for an interface if it has a non-empty Comment + (aka user description). That means we need to make sure that the + map has a default empty QStringList for that pref so we can set it + to the empty string; otherwise, if some interfaces have Comments, but + the user tries to the clear the comments from all interfaces, it won't + work. The map won't have an entry for that pref, so it won't get + changed to the empty string. - Fix #19800 + The same is true for the Monitor Mode setting. - (cherry picked from commit a2eaaa4e8804189ef50406fb6b2783d492367949) + (cherry picked from commit f622eac21f24b450076fb85bac3b40f58eca2ecc) Co-authored-by: John Thacker <johnthacker@gmail.com> -commit 22a9b5ab5d -Author: Gian Lorenzo Meocci <gianlorenzo.meocci@athonet.com> -Date: Wed Apr 24 09:37:25 2024 +0200 - - fix pfcp outer header desc - -commit 52d644b084 -Author: Niels Widger <niels@qacafe.com> -Date: Wed Apr 24 12:26:07 2024 +0000 - - sharkd: Improve sub-stream support in "frame" and "follow" methods - - Add new optional `sub_stream` input argument to `sharkd`'s `follow` - method which, if given, specifies the sub-stream index number to - follow (e.g. for HTTP/2 or QUIC streams). If no `sub_stream` argument - is given, all sub-streams are followed (by using the special - `SUBSTREAM_UNUSED` sub-stream index value). - - Add new `followers` output field to `sharkd`'s `frame` method which - contains an array of objects each containing `protocol`, `filter`, - `stream` and (for HTTP/2 or QUIC streams) `sub_stream` fields. A new - `followers` array was added to the response instead of adding the - stream and sub-stream indices to the existing `fol` field due to - backwards-compatibility issues. The `fol` field is an array of string - arrays, each subarray containing the same values as the `protocol` and - `filter` fields in the new `followers` array, i.e.: - - "fol": [["TCP", "tcp.stream eq 0"], ["TLS", "tcp.stream eq 0"]] - - which makes adding new, possibly optional, fields difficult. The - `fol` field should be considered deprecated, any new fields should be - added to `followers`. The new `followers` field is an array of - objects, making adding new fields down the road simple: - - "followers": [{"protocol": "TCP", "filter": "tcp.stream eq 0"}, {"protocol": "TLS", "filter": "tcp.stream eq 0"}] - - Updated existing and added new `sharkd` unit tests to validate - behavior of new fields/arguments. - - (cherry picked from commit 289a159042e3b81c2c135031dd4da9590aea848e) - - Co-authored-by: Niels Widger <niels@qacafe.com> - -commit 15b77707dd -Author: John Thacker <johnthacker@gmail.com> -Date: Wed Apr 24 22:41:03 2024 +0000 - - github/workflow: fix macos - - For macos-latest runner on Github using macos14 on arm there is only Python - >= 3.11 available. - - We have to install pytest and dmgbuild manually +commit e128c2804e +Author: Gerald Combs <gerald@wireshark.org> +Date: Sun Jun 2 09:26:34 2024 +0000 - (cherry picked from commit f186f44a711b1ba831173238c49a8e280697ae1e) + [Automatic update for 2024-06-02] - Co-authored-by: Uli Heilmeier <uh@heilmeier.eu> + Update manuf, services enterprise numbers, translations, and other items. -commit 06a5e5c445 +commit c15c3f9b18 Author: John Thacker <johnthacker@gmail.com> -Date: Wed Apr 24 10:55:41 2024 +0000 +Date: Sat Jun 1 12:23:13 2024 +0000 - SSH: Handle "curve25519-sha256@libssh.org" KEX name + Qt: Fix leak on CaptureFileDialog preview of file with errors - As RFC 8731 notes, 'The "curve25519-sha256" key exchange method is - identical to the "curve25519-sha256@libssh.org" key exchange method - created by Aris Adamantiadis and implemented in libssh and OpenSSH.' + Remember to call wtap_close when the preview stats have an error + to avoid leaking the file. - Handle the other name in ssh_kex_hash_type. It is already accounted - for in other functions. - - Fix #19240 - - (cherry picked from commit 6966cdb73e6f5dc2e6a03136b07b8ed59c7ff65a) + (cherry picked from commit c29c1bb27ce6a4ab50866ea3b580af859d44323b) Co-authored-by: John Thacker <johnthacker@gmail.com> -commit 1a88d0c734 -Author: Christian Krump <christian.krump@br-automation.com> -Date: Tue Apr 23 10:58:35 2024 +0200 - - EPLv2: modified limitation of valid SDO segment size - - (cherry picked from commit a9f783600fc787ce891f1eda8e242793cc436fc8) - -commit 585a3dc48c -Author: Gerald Combs <gerald@wireshark.org> -Date: Tue Apr 23 13:27:56 2024 -0700 - - GitLab CI: after_script updates - - In GitLab 17 and later, after_script will run when a job is cancelled. - Update our various jobs accordingly. - - (cherry picked from commit e03a58958e065c7e2dc7e15a90978fa5d57edcf6) - - Conflicts: - .gitlab-ci.yml - -commit 0c45a5d772 +commit 825070af48 Author: John Thacker <johnthacker@gmail.com> -Date: Tue Apr 23 03:03:28 2024 +0000 +Date: Sat Jun 1 02:22:01 2024 +0000 - lldpd: move hf_lldp_tlv_enable_system_cap* to capabilities_enabled_tree + test: skip test_sharkd_req_follow_http2 test if no http2 support - (cherry picked from commit e267bfdd88a7fe190f52906ac98a3b03035c580f) + Originally reported downstream in Gentoo at https://bugs.gentoo.org/933290. - Co-authored-by: Paul Donald <newtwen+github@gmail.com> + This is the same kind of thing as https://gitlab.com/wireshark/wireshark/-/merge_requests/9029. -commit c8a21a6d58 -Author: Guy Harris <gharris@sonic.net> -Date: Tue Apr 23 02:28:05 2024 +0000 + (cherry picked from commit cc8264d68f638c72ada1e94a9913902462c1a25a) - fcfzs: fix calculation of the number of fill bytes. + Co-authored-by: Sam James <sam@gentoo.org> - If a name length is a multiple of 4, no padding is required to make what - follows be on a 4-byte boundary. Only use 4 - (len % 4) to calculate - the padding if len isn't a multiple of 4. +commit 182646016c +Author: Martin Mathieson <martin.r.mathieson@googlemail.com> +Date: Fri May 31 15:44:21 2024 +0000 - (cherry picked from commit e8b256c9a0f427b8dc2387ab02a47dc8c2f687e0) + Protobuf: don't write beyond end of array - Co-authored-by: Guy Harris <gharris@sonic.net> - -commit 0b80c5ab9e -Author: Stig Bjørlykke <stig@bjorlykke.org> -Date: Mon Apr 22 15:52:00 2024 +0200 - - icmpv6: Do not add time string when lifetime is infinity - - Do not convert the lifetime value to string when the value - is used for infinity. - - (cherry picked from commit 1a40e22a0095056cc4397e07d7b95ce74587a65b) +commit d61af59a94 +Author: John Thacker <johnthacker@gmail.com> +Date: Fri May 31 12:54:10 2024 +0000 -commit 8e515cac2e -Author: Anders Broman <a.broman58@gmail.com> -Date: Mon Apr 22 12:17:29 2024 +0000 + Qt: Fix CaptureOptionsDialog tooltip for stop files spinbox - NR-RRC: Remove version from SIB dissector table name. + The tooltip for the stop files spinbox is accidentally duplicated + from the stop number of packets spinbox. - (cherry picked from commit e40ec77006f159f816038e5295851f0caf218a7a) + (cherry picked from commit 8b9c3c620e983ec4e63cfeda2f5de449f2bd3ace) - Co-authored-by: Anders Broman <a.broman58@gmail.com> + Co-authored-by: John Thacker <johnthacker@gmail.com> -commit 98fa3eff88 +commit 2e88a7ca5f Author: Pascal Quantin <pascal@wireshark.org> -Date: Mon Apr 22 15:40:52 2024 +0200 - - NAS 5GS: fix dissection of NAS message container IE +Date: Fri May 31 10:40:32 2024 +0200 - Take into consideration the outer NAS message security header before - trying to dissect the IE. + NAS 5GS: fix dissection of extended rejected NSSAI IE - Closes #19793 - - (cherry picked from commit ab3511c5977d0dd483d1d7e3d741940a12eaacc4) + (cherry picked from commit adb9543df7a312e9309f0b54eeab62249de54d84) Conflicts: epan/dissectors/packet-nas_5gs.c -commit 6ed87c1638 +commit df91454779 Author: John Thacker <johnthacker@gmail.com> -Date: Mon Apr 22 13:14:34 2024 +0000 - - QUIC: handle connection migrations for zero length CIDs - - In Quic Connection Migrations are possible even without source - connection IDs. Currently, after connection migration Wireshark fails to - associate answers with zero length CIDs for the new address to the - original connection. - - After migration when the client sends data from the new IP - the connection data needs to be associated with the new conversation. - So when the server answers and the connection is identified by the - conversation a connection is found. - - (cherry picked from commit 95a9fc5c0fa569cc7c5026f0ef235736b99848eb) - - Co-authored-by: Max Schrötter <mail@schrc3b6.de> - -commit c77f142c0e -Author: Anders Broman <a.broman58@gmail.com> -Date: Sun Apr 21 18:34:09 2024 +0000 +Date: Wed May 29 15:06:27 2024 +0000 - NR-RRC: Add callable dissectors for SIB messages + MySQL: Handle DATETIME if the column definition is unsigned - Add callable dissectors for SIB messages. - Add dis table - Fix column display for single SIB message. + At least some captures mark DATETIMEs with the UNSIGNED flag in + the column defintion. Dissect those too. - (cherry picked from commit dfec2284f34dae9ee515539d5ce5bb9f14762e3c) + Dissect #19859. - Co-authored-by: Anders Broman <a.broman58@gmail.com> - -commit 77ccc7ece0 -Author: Guy Harris <gharris@sonic.net> -Date: Mon Apr 22 09:15:13 2024 +0000 - - tipc: fix byte count. - - Use the value used as the number of bytes in an atom to determine - whether that number is singular or plural; don't use some other - unrelated value. - - (cherry picked from commit aea66ddc3336a095e6201b633a78014bd3af3a1d) - - Co-authored-by: Guy Harris <gharris@sonic.net> - -commit 207981eb13 -Author: Guy Harris <gharris@sonic.net> -Date: Mon Apr 22 06:54:31 2024 +0000 - - tipc: fix padding after bearer instance string. - - Presumably if the length of the string is a multiple of 4, it needs zero - byts of padding to put the next item on a 4-byte boundary. This means - that (4 - (length % 4)) is not the right formula - a check for whether - there's any padding must be done first. - - (cherry picked from commit cda7c030e1eafdca22028d1d8c87e484bba37555) - - Co-authored-by: Guy Harris <gharris@sonic.net> - -commit 05e9eeef13 -Author: Gerald Combs <gerald@wireshark.org> -Date: Sun Apr 21 09:25:15 2024 +0000 - - [Automatic update for 2024-04-21] - - Update manuf, services enterprise numbers, translations, and other items. - -commit fc290d976e -Author: Guy Harris <gharris@sonic.net> -Date: Sat Apr 20 18:58:25 2024 +0000 - - pcapio: fix a check for idb_filter size. - - The filter string in the idb_filter option has a maximum length of - UINT16_MAX - 1, because the maximum size of an option bodyis UINT16_MAX, - and the first byte of the option contains the filter type. - - The code to write the filter did the right check; the code to count the - idb_filter option length checked against UINT16_MAX. Fix that. - - While we're at it, fix a comment. - - (cherry picked from commit d277ecdb3ff7d04e7f64566f21409b3f1359eb2f) - - Co-authored-by: Guy Harris <gharris@sonic.net> - -commit 32bde22d9b -Author: John Thacker <johnthacker@gmail.com> -Date: Sat Apr 20 13:04:27 2024 +0000 - - editcap, libwiretap: Don't use array of initial DSBs after freeing - - wtap_dump_close frees the passed in GArray of initial DSBs, used - by editcap for injecting DSBs from a file or list of files. - - Add functions to increment and decrement the reference count of - an array of wtap blocks. Dereference the block of initial DSBs - in wtap_dump_close() instead of freeing it. In editcap, before - closing the dump file in cases where we intend to open a new - file (e.g., with a maximum time value or a maximum packet count), - reference the block. - - Fix #19782, #19783, #19784. - - (cherry picked from commit be3550b3b138f39bebb87ac0b8490e75fc8cc847) + (cherry picked from commit 9962a87a8a02ec09a98e2f34dcb5095bf409c3c4) Co-authored-by: John Thacker <johnthacker@gmail.com> -commit b8d61cd977 -Author: Pascal Quantin <pascal@wireshark.org> -Date: Sat Apr 20 12:12:00 2024 +0000 - - GSM RP: fix dissection of SMS in 5G Nf interface - - Closes #19773 - - (cherry picked from commit 9dffc3834ea0f7439a2e2d2f55350b6d5e2596c5) - - Co-authored-by: Pascal Quantin <pascal@wireshark.org> - -commit a7661fd568 +commit cef77b8fed Author: John Thacker <johnthacker@gmail.com> -Date: Thu Apr 18 15:37:18 2024 +0000 - - Windows: Add include depended upon packets checkbox - - Non-Windows already has this checkbox. +Date: Wed May 29 14:22:43 2024 +0000 - The vast majority of the time users do want dependent packets - so that reassembly, etc. works. Occasionally they don't (e.g., - perhaps to exclude retransmitted segments or duplicates), but - the primary use of this checkbox is helping users understand why - extra packets are included so they don't think it's in error. + SPRT: Fix crash - We might want to get rid of the special Windows dialogs. They - use a deprecated API, but the newer API is more difficult to customize. - The Qt common dialogs are better than the GTK+ file dialogs at the - time we started to use the Windows dialogs. + SDP can setup a RTP conversation with a setup frame before the current + frame, which changes the dissection on the second pass. If in the period + in the middle there is a SPRT packet, it can be dissected differently on + the second pass, and the SPRT conversation data won't be found on the + second pass. - Follow up to dcc9cbffefe0598eadec11cbc1869943a26219c5 - Fix #19772 + Fix #19559 (at least prevent the crash. There's some more cleanup that + should happen.) - (cherry picked from commit 90d7ecf6971b89a918f61d95a9bce77d74f8cb64) + (cherry picked from commit 05f6364cbd766e8758f98c5ee2070aef27c1ffef) Co-authored-by: John Thacker <johnthacker@gmail.com> -commit bd26249852 -Author: Alexis La Goutte <alexis.lagoutte@gmail.com> -Date: Fri Apr 19 06:50:41 2024 +0000 - - Zigbee Direct: Fixed UUID mismatch with join, pjoin & leave characteristics - - (cherry picked from commit 8b802bffd12e3bbd80bb6e6e501309bb7fee8124) - - Co-authored-by: Cole Wu <chw@exegin.com> - -commit 0e5038cf89 -Author: Alexis La Goutte <alexis.lagoutte@gmail.com> -Date: Thu Apr 18 11:13:55 2024 +0000 - - ieee80211: Fix typo in display filters - - (cherry picked from commit d4da04f8ee47f12cb737601fdbf2be48e881303c) - - Co-authored-by: WFA-achuang <achuang@wi-fi.org> - -commit d4526a807b -Author: John Thacker <johnthacker@gmail.com> -Date: Thu Apr 18 11:56:00 2024 +0000 - - Qt: Fix removing ignored packets from selected range - - Correctly remove ignored packets from the selected packet - range on non-Windows, which for some reason has never done this. - The Windows dialog has. - - (cherry picked from commit 317177d1d079a1054f2b8935c30c376d67dd7eb5) - - Co-authored-by: John Thacker <johnthacker@gmail.com> - -commit 4d63d194ba +commit 9797333ed5 Author: John Thacker <johnthacker@gmail.com> -Date: Tue Apr 16 09:05:40 2024 -0400 - - pcapng: fix writing hash options - - Since compute_block_option_size() and our option writing routines - already add needed padding, we don't need to compute the size - with padding in pcapng_compute_packet_hash_option_size(). That allows - us to use it both when computing the option size for the total - block size, and in the writing routine, for consistency. - - Fix some errors where the type octet was not being included in - the bytes written for both one type of packet verdict, and for - packet hash. Also an error where the packet verdict option was - written with the queue option type. - - Fix a possible buffer overrun for the fixed length hash types - when the hash length is wrong. We have a few different options - about how to handle hashes with bogus length - omit, write what - we have, truncate (if we have extra bytes) or err. Add some - comments regarding the options; for now, choose reporting an - error. - - Fix #19766 - - (backported from commit b36749f56a4b2a9f1931aa0ab472dbf7420a75e3) - -commit 9c5c64b4cb -Author: Gerald Combs <gerald@wireshark.org> -Date: Mon Apr 15 16:06:46 2024 -0700 - - pcapng: Fix logging - - WS_LOG_DOMAIN must be defined after including config.h and before - including wslog.h. - - (cherry picked from commit 09b83c0ecee60641439b67ed5931714492ae38c1) - -commit b822d573e0 -Author: Gerald Combs <gerald@wireshark.org> -Date: Sun Apr 14 09:25:33 2024 +0000 - - [Automatic update for 2024-04-14] - - Update manuf, services enterprise numbers, translations, and other items. - -commit 2ece4fa7dc -Author: John Thacker <johnthacker@gmail.com> -Date: Fri Apr 12 07:19:12 2024 -0400 - - ieee1905: Fix bit ordering of HE-MCS set - - The Tx Rx HE MCS Support field in IEEE 1905 is reordered from - 802.11ax into big-endian order (which is already done), but that - shouldn't change which bit is considered the LSB (which is the - Max HE-MCS For 1 SS) and which is the MSB (Max HE-MCS for 8 SS). - - Fix #19737 - - (cherry picked from commit 179bf05a4d4e64427fa7e7805f1d4e134874cca8) - -commit cee6caf8ea -Author: John Thacker <johnthacker@gmail.com> -Date: Wed Apr 10 12:16:02 2024 -0400 - - Qt: Hide export dissections dialog when accepting - - When the Export Dissections Dialog is accepted but starts to - try to save the files, hide it. It will close itself after the - export is done (which deletes it), but in the meantime the - exporting from a large file can take a long time. Hiding the - dialog allows the user to cancel the save with the Progress Dialog - on the main window. Leaving the dialog visible allowed the user - to click "Cancel" on the dialog, which didn't cancel the save - (still up to the progress dialog), but did delete the dialog on - close, resulting in a crash when the export did finish. - - This is for Linux and macOS, and matches the behavior of the native - Windows dialog, which already hides itself when Save is clicked. - - (cherry picked from commit e84adf41da08c60ff329f3e91bb948e2e1e33753) - -commit 7fb337faf8 -Author: John Thacker <johnthacker@gmail.com> -Date: Wed Apr 10 10:28:45 2024 -0400 - - Qt: Fix crash when closing TimeShiftDialog before redissection done - - Commit f67eccedd9836e6ced1f57ae9889f57a5400a3d7 made it so that - the TimeShiftDialog doesn't have a nested event loop with exec. - That's good, but that means that an AutoConnection between - TimeShiftDialog::timeShifted and PacketList::applyTimeShift is a - DirectConnection, which means that TimeShiftDialog::applyTimeShift - waits for the packet list redissection to finish before the function - calls enableWidgets and returns. - - The user can close the Time Shift Dialog while the packet list - redissection is still ongoing. As the TimeShiftDialog is also now - DeleteOnClose, this means that when the redissection finishes (or - is canceled via ProgressDialog) the TimeShiftDialog will try to update - its widgets after they've been deleted, causing a crash. - - To prevent, make the connection between the TimeShiftDialog and - the PacketList a Qt::QueuedConnection. - - (cherry picked from commit ea70c16e05d0f1f0779bec6e89153a67d07d6cfa) +Date: Mon May 27 10:33:39 2024 +0000 -commit 799735b9c0 -Author: Dr. Lars Völker <lars.voelker@technica-engineering.de> -Date: Fri Mar 22 18:23:07 2024 +0100 + Qt: Only remove and delete the FunnelTextDialog's own buttons - BLF: Fixing Flags Decoding in FlexRayRcvMessage(Ex) (BUGFIX) + FunnelTextDialog has a static global map of all the buttons. When closing + a dialog, only delete the buttons from that dialog's button box, not all + the button from all the dialogs in the map. - Flags should have been in the FrameFlags byte. + Fix #18386 - (cherry picked from commit 823b49f6c338ddc0e99315e05e3e3fb5c557456c) + (cherry picked from commit d40040eac72bd3ceb51cdd1b99b9553c3f7aa48c) -commit 6288f7c592 -Author: Giovanni Musto <giovanni.musto@partner.italdesign.it> -Date: Wed Dec 13 14:26:36 2023 +0100 - - BLF: Fix BLF_OBJTYPE_CAN_FD_ERROR_64 not decoded as error - - (cherry picked from commit 7b4d667f306c4ffb6c181280bdb47d01826eb3e6) - -commit 1c7dbe91da -Author: Guy Harris <gharris@sonic.net> -Date: Sat Nov 18 14:53:33 2023 -0800 - - blf: expand a comment. - - Give more details on how the amount of padding between log container - objects is not an obvious "padd the object's size to a multiple of - <power of 2> bytes". - - [skip ci] - - (cherry picked from commit 8c22a6cc0240123750347173acef964a3909da34) - -commit 559a6f89f7 -Author: Guy Harris <gharris@sonic.net> -Date: Sat Nov 18 11:27:26 2023 -0800 - - blf: check the validity of the header length field. - - (cherry picked from commit 3e296c041f233b6605f4004bec7e4be1acd34b70) - -commit 13bcc043a4 -Author: Guy Harris <gharris@sonic.net> -Date: Sat Nov 18 01:02:04 2023 -0800 - - blf: add more error checks. - - Check for invalid lengths in headers. - - Have blf_scan_file_for_logcontainers() provide error codes and messages - on errors, and, if it fails, clean up and return an error when opening a - file. - - Always check for errors from file_seek(). - - (cherry picked from commit ae5c7ff41bdd031f7d29151501f42bfcc68d662d) - -commit 91e1ec23d7 -Author: Giovanni Musto <giovanni.musto@partner.italdesign.it> -Date: Mon Oct 23 09:30:29 2023 +0200 - - BLF: Extract FDF, BRS and ESI flags - - (cherry picked from commit c0f287ad69e05ceb8c71a5f7100cd7c59b3835ae) - -commit 058ca16704 -Author: Yingjie Deng <17322619133@163.com> -Date: Thu Apr 11 01:26:26 2024 +0800 - - fix: Hmac Missing - - The peer->mac Algorithm is not should set NULL here. The above - ssh_decryption_set_cipher_id function takes peer->enc, and set - peer->cipher to NULL in else case, the peer->cipher is the field of - struct ssh_peer_data, but not have corresponding field with peer->mac in - struct ssh_peer_data.So I just delete it in else case. - - (cherry picked from commit a2a911ea1a0adeb5a93f7cf4fd786ca0fdd0c820) - -commit a31b9964e6 -Author: Giovanni Musto <giovanni.musto@partner.italdesign.it> -Date: Tue Oct 17 09:36:17 2023 +0200 - - BLF: Fix LIN parsing - - LIN was broken because its payload was treated as a variable length field, while it is always 8 bytes - Fix DLC parsing to allow only 8 bytes instead of 15 - Fix ID parsing to take only the lower 6 bits - Parse checksum field - - (cherry picked from commit 7bf7a136766fb20132d1b9cc550f661dc70ab2d2) + Co-authored-by: John Thacker <johnthacker@gmail.com> -commit eaf7d424a7 -Author: John Thacker <johnthacker@gmail.com> -Date: Sun Apr 7 16:57:23 2024 -0400 - - Qt: Work around Qt6 dataChanged pessimization - - A change in Qt 6.0 makes QAbstractItemView::dataChanged, when called - with multiple indices, determine exactly how much of the viewport - rectangle is covered by the changed indices and only update that much - of the viewport instead of always updating the entire viewport. This - reduces CPU load when multiple indices, but not enough to span the - entire viewport, are updated at once. - ( https://codereview.qt-project.org/c/qt/qtbase/+/285280 - https://bugreports.qt.io/browse/QTBUG-58580 - ) - - Unfortunately, if the number of indices is very large (much larger - than the total number of indices that can possibly be displayed in - the viewport), computing the union of the intersecting rectangle takes - orders of magnitude longer than unconditionally updating the whole - viewport. ( https://bugreports.qt.io/browse/QTBUG-124173 ) - - When the packetListModel wants to indicate that all packets may have - changed, this makes dataChanged scale linearly with the total number of - rows/frames, whereas updating the viewport only scaled with the size of - the viewport but was unaffected by undisplayed packets. The calculation - is unnecessary because we know the entire viewport is affected. On a - sample modern workstation, measurements showed dataChanged taking 1 s - with 1.4 M packets and 9 s with 12 M packets. This signal can be issued - quite often, e.g. if IP address lookup (DNS or MaxMindIP) is enabled, it - can be issued once per second at the packet_data_timer expiration for - some period of time as idle dissection occurs when opening the file. - - Instead of issuing dataChanged, we can issue layoutAboutToBeChanged() - (in practice a no-op) and layoutChanged(), which causes the QTreeView - to clear all information about its view items (d->viewItems.clear() in - QTreeView::doItemsLayout), but without clearing the current and selected - row (unlike [begin|end]ResetModel.) This takes the same 5-8 ms in the - same test procedure as above in both the 1.4 M and 12 M packet tests. - - Improves #19486 considerably on Qt 6. - - (cherry picked from commit 5abfe91cc164d873e56c6f24a313b93b7d883a9e) - -commit e51f5bf24d +commit 7d95ca8f3d Author: Gerald Combs <gerald@wireshark.org> -Date: Sun Apr 7 09:24:53 2024 +0000 +Date: Sun May 26 09:25:08 2024 +0000 - [Automatic update for 2024-04-07] + [Automatic update for 2024-05-26] Update manuf, services enterprise numbers, translations, and other items. -commit 31188b669e -Author: John Thacker <johnthacker@gmail.com> -Date: Thu Apr 4 20:27:45 2024 -0400 - - reordercap: Handle IDBs in the middle of the file - - We can read IDBs (and other non packet blocks) in the - middle of the file, so init the dump parameters after - reading all the frames. This will move the IDBs (and NRBs - and DSBs) to the start of the new output file (which might - have to happen if they're out of order.) - - Note: Files with multiple Section Header Blocks probably still - aren't handled correctly, because the IDB number might need to be - rewritten (though we have some of that information after - commit 8ebde1309d0cc0335e32cff8c7112dc98c05d5ed) +commit ebac462611 +Author: Niels Widger <niels@qacafe.com> +Date: Thu May 23 18:15:32 2024 +0000 + + sharkd: Fix bad "stream"/"sub_stream" reponse fields in "frame" method + + In 289a1590, a new `followers` response field was added to `sharkd`'s + `frame` method with a mandatory `stream` field and an optional + `sub_stream` field. These fields contain the stream and sub-stream + index numbers for the conversation, which were obtained in + `sharkd_followers_visit_layers_cb` using `get_follow_conv_func`. This + function takes two `guint` pointers which are set by the protocol. In + reality, not all protocols (SIP for example) set the `stream` pointer + parameter passed to `get_follow_conv_func`, meaning the value of the + mandatory `stream` response field came from uninitialized memory. + This led to non-sensical `stream` response field values such as + `-2071892000` or `1726050272` being sent. + + To fix this issue, this commit makes both the `stream` and + `sub_stream` response fields optional by updating + `sharkd_followers_visit_layers_cb` to only output these fields if + `get_follow_stream_count_func` (for `stream`) or + `get_follow_sub_stream_id_func` (for `sub_stream`) do not return + `NULL`. This seems to be how `FollowStreamDialog::follow` in + `ui/qt/follow_stream_dialog.cpp` determines whether the `stream` and + `sub_stream` pointers passed to `get_follow_conv_func` can be assumed + to have been set. This also avoids the awkward need for a sentinel + value such as `SUBSTREAM_UNUSED` to determine if + `get_follow_conv_func` set `sub_stream`. + + Finally, use the more appropriate `%u` format verb instead of `%d` in + calls to `sharkd_json_value_anyf` when outputting the `stream` and + `sub_stream` fields, which both come from a `guint` and are thus + unsigned. + + (cherry picked from commit aa760445c39255d6f914cd5a48c8bacdcd1237a9) - Fix #19740 + Co-authored-by: Niels Widger <niels@qacafe.com> - (cherry picked from commit 0c637a2b5ebaf6bf24f406e5e01a39860d4a4daa) +commit 6d0aee9a39 +Author: Pascal Quantin <pascal@wireshark.org> +Date: Tue May 21 11:49:52 2024 +0000 -commit f47c873dd8 -Author: mverkleij <mail@mverkleij.nl> -Date: Fri Mar 15 14:43:29 2024 +0100 + RLC LTE: fix dissection of EUTRA DL CCCH messages - Fix reordercap -n flag generating empty output file + (cherry picked from commit 0c3bf2d77e32e08011e5567ede189b45947805ca) - (cherry picked from commit ef64dcf32879353f02057ada93c0b0add78cad3c) - (cherry picked from commit e6ad1f677ea970b85d08066135cd8ec10d23075c) + Co-authored-by: Pascal Quantin <pascal@wireshark.org> -commit b6ba652b5b +commit f29bc46988 Author: John Thacker <johnthacker@gmail.com> -Date: Thu Apr 4 07:12:45 2024 -0400 - - Qt: Ensure Copy Bytes as C String is a valid C string - - Use octal escapes padded to three digits instead of hex escapes. - An octal escape is limited to three digits by definition, whereas - hex escape sequence are not limited, which causes problems if - a hex escape is followed by a (non escaped) possible hex character. - - Escape double quotes and backslash. Use the simple escapes of - non printable characters too (optional, but looks nicer.) - - This possibly could go in wsutil as another ws_escape method, but - the fix needs to be backported to 4.2 so let's not introduce a - new public function yet. - - Fix #19735 +Date: Mon May 20 11:24:19 2024 +0000 - (cherry picked from commit 7130e33d30acb871538e540054ef8293912ed33a) + Qt: Switch rwin_graph axes depending on TCP Stream graph selected -commit 5a58585792 -Author: Nardi Ivan <nardi.ivan@gmail.com> -Date: Wed Jan 17 10:02:25 2024 +0100 + Fix #19846 - LUA: allow conversion from string to uint64 in any base + (cherry picked from commit 8f6822071ee44931a9594a282bc4e3649bd7153a) - This is handy when you want to specify a mask in hex format - - (cherry picked from commit 534cc08ebc9fb82b965cefdf164ae02da3827f4d) - -commit e067aa55f3 -Author: Guy Harris <gharris@sonic.net> -Date: Sun Mar 31 12:10:56 2024 -0700 - - wslua: put the example of use for treeitem:add at the right level. - - It should be a subsection of the entry for treeitem:add, which means it - must be a level *5* header. - - (cherry picked from commit 7b6e196c4117010c04d911021c16b37c454a4f19) - -commit 48288a29db -Author: Guy Harris <gharris@sonic.net> -Date: Sun Mar 31 10:51:30 2024 -0700 - - wslua: fix the documentation for Tvb:reported_length_remaining. - - Use the WSLUA_OPTARG_xxx convention for it optional offset argument, so - that the documentation generation process properly documents it. - - (cherry picked from commit 62bc594401f64317d4e2a66de726ca615cf0263b) - -commit 3911c7b7d2 -Author: John Thacker <johnthacker@gmail.com> -Date: Fri Mar 29 09:42:44 2024 -0400 + Co-authored-by: John Thacker <johnthacker@gmail.com> - editcap: Don't memmove more than allocated in the buffer +commit 710b34cda3 +Author: Jeff Dyer <jmasterfunk@gmail.com> +Date: Wed Jan 10 16:12:35 2024 +0000 - When moving from the begining with a beginning offset specified, - don't run off the end. Subtract the source memory area's full offset - from the beginning of the buffer from the capture length. + NAS 5GS: set subdissector for SMS NAS messages in HTTP2 - Fix #19724 + (cherry picked from commit 885e638a03bccdc212b51ff5517b394ab029da37) - (cherry picked from commit 7c744e7933794b09e7af4d9703194ad0b01be282) + Closes #19845 -commit 1f0bc61bbc +commit b794e7985c Author: Gerald Combs <gerald@wireshark.org> -Date: Sun Mar 31 09:24:23 2024 +0000 +Date: Sun May 19 09:25:28 2024 +0000 - [Automatic update for 2024-03-31] + [Automatic update for 2024-05-19] Update manuf, services enterprise numbers, translations, and other items. -commit e9965fe303 +commit 76ee960786 Author: John Thacker <johnthacker@gmail.com> -Date: Sat Mar 30 08:07:26 2024 -0400 - - Mongo: Ensure the offset advances +Date: Thu May 16 14:37:34 2024 +0000 - The MongoDB Wire Protocol uses _signed_ 32 bit integers for lengths. - dissect_bson_document checks for bogus values and ensures that a - non-negative (and at least 5) size is returned, but we need to make - sure to use that return value instead of trusting the value read - from the packet in dissect_op_msg_section. + tools: Update alpine-setup - Fix #19726 - - (cherry picked from commit 38c0efcee8d22d922e446888b268effc3ccf725f) - -commit 17bff1a32f -Author: Guy Harris <gharris@sonic.net> -Date: Fri Mar 29 22:31:04 2024 -0700 + Alpine Linux defaults to the BusyBox ash, in bash compatible + mode supporting `set -e -u -o pipefail` and the rest of the + setup script, so use that. Install bash just in case because + some of our other scripts might have some bashisms, I haven't + checked. - macos-setup: Add a warning about a comparomised version of XZ. - - [skip ci] - - (cherry picked from commit 75b1ffe1190de3843de3cd756a6b16200a967cd8) - -commit 8a7dc8c440 -Author: John Thacker <johnthacker@gmail.com> -Date: Thu Mar 28 19:46:56 2024 -0400 + Update to Qt6 and add opencore-amr-dev and libcap-setcap to the + optional packages, since all currently supported (and some + out of support) Alpine releases have them. Lua 5.4 will be in + another commit just to ease backporting this to 4.2 - JSON-3GPP: Handle 5GC over HTTP/1.1 (OAI), don't create HTTP/2 sessions + (cherry picked from commit 7bf71c0b3a5aa4122b2eed154119d3d59f5bc355) - http2_get_header_value currently, by calling get_http2_session, - always creates HTTP/2 conversation data on the current conversation. - This confuses the HTTP dissector if HTTP/2 is not actually present, - so guard it with proto_is_frame_protocol. - - Also, despite 3GPP TS 29.500 saying that the service based interfaces - use HTTP/2 only, OAI (and perhaps other implementations) have options - to use HTTP/1.1, so try to handle that case as well. - - Fix #19723 - - (backported from commit 8285e802241abb25c62d6fc71b17ce16175c151b) - -commit 47517b2cdd -Author: John Thacker <johnthacker@gmail.com> -Date: Tue Mar 5 07:50:09 2024 -0500 - - http2: Return header value decoded from US-ASCII - - Per the discussion in RFCs 9113 and 8187, decode field values - from US-ASCII (replacing characters outside the range, i.e. - obs-text, with UTF-8 replacement characters) before returning them - in http2_get_header_value. This is what all current dissectors - need, and reduces the chances of adding unvalided text (non UTF-8) - to the tree. - - Make a note that dissectors may need to perform additional decoding, - e.g. percent decoding or decoding according to RFC 8187. - - There may at some point be dissectors that need access to the raw - bytes as opaque data (which is discouraged by the RFCs), at which - point we could have an additional function to access that. - - Fix #19684 - - (cherry picked from commit 674e35bd62777e006fd41229c4c6583022d912f9) - -commit 9ad30bc466 -Author: Gerald Combs <gerald@wireshark.org> -Date: Fri Mar 29 10:27:10 2024 -0700 - - GQUIC+DOCSIS MAC MGMT: Recursion updates - - Make our recursion checks consistent with other dissectors. - - (cherry picked from commit 119a385ba271a51ea35579b25dc0025ff450577c) + Co-authored-by: John Thacker <johnthacker@gmail.com> -commit 4e707387de +commit 7587228433 Author: John Thacker <johnthacker@gmail.com> -Date: Thu Feb 29 19:26:23 2024 -0500 +Date: Thu May 16 14:11:19 2024 +0000 - gRPC: Add separate dissector handles for gRPC Web and Web Text + CMake: Look for the lua library with versioned suffixes - Rather than registering one gRPC dissector handle to three - flavors of gRPC and then figuring out which one got called - by inspecting pinfo->match_string, register different dissectors - with descriptions to each type, so we know which dissector was - called. + Alpine Linux puts not just the include files but also the + library in a versioned subdirectory, so look for it there too. - This especially avoids assertions when a dissector is set via - Decode As, or via the stream ID table (even though the stream - ID table won't put the gRPC dissector into the special - streaming "reassembly as you go" mode, instead falling back - to the standard HTTP/2 reassemble at the end mode.) + Fix #19841 - Also, if we have the (bidirectional) streaming mode set for - the stream ID but we don't have a streaming subdissector handle - for the (unidirectional) content type, set it to the data handle - to avoid an assertion. We could possibly try to set it to the - content type used in the other direction, but that has potential - issues. + (cherry picked from commit 59b376dce046eb4f739a70f49c20ac9da9cab794) - Fix #19679 - - (cherry picked from commit febcf46bbc4ff7e16c5cff609dc0c733b9e652a3) - -commit 196cc0d624 -Author: Dominic Pearson <dsp@technoanimal.net> -Date: Thu Mar 28 15:07:40 2024 +0100 - - cpu_info: fix missing decl in else macro block - - (cherry picked from commit 318e8d3161ac23a7f486a74500d4d2b05403a58a) + Co-authored-by: John Thacker <johnthacker@gmail.com> -commit c9f47f9ff5 +commit dd7f53165b Author: Gerald Combs <gerald@wireshark.org> -Date: Wed Mar 27 14:40:37 2024 -0700 +Date: Wed May 15 12:55:25 2024 -0700 - Version: 4.2.4 → 4.2.5 + Version: 4.2.5 → 4.2.6 [skip ci] |