Modifications to RTT graph:
- change x-axis to time (s) rather than sequence number
[ avoids sequence number wraparound ambiguity, plus
easier to correlate RTT changes to tcptrace graph ]
- change RTT computation to properly handle acks to GSO packets
- change RTT computation to take SACK blocks into account
Bug fixes:
- eliminate potential memory leak if some packets are unacked
- ensure RTT graph is shown if TCPGraph window is opened to it directly
Change-Id: I2bdcab97399ebde0f15c78fa19c882529a814580
Reviewed-on: https://code.wireshark.org/review/19662
Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Call to plugin_if_get_ws_info when no packets are being displayed
causes an Access Violation in plugin_if_mainwindow_get_ws_info
as the pointer in cf->current_frame is NULL.
Bug: 12973
Change-Id: I33e8fe6f3894963a6f1f1ec9aa9169f85e1e3b18
Reviewed-on: https://code.wireshark.org/review/19659
Reviewed-by: Roland Knall <rknall@gmail.com>
Refactored so that all handling of ranges in struct preference
can be internal to prefs.c
Change-Id: I68577909f9c07b23a16ab3443a523355d4645314
Reviewed-on: https://code.wireshark.org/review/19577
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
This is mostly to address memory leaks in range preferences (the biggest
user of range functionality) on shutdown.
Now range preferences must use epan scoped memory when referencing
internal preference structures to keep consistency.
Change-Id: Idc644f59b5b42fa1d46891542b53ff13ea754157
Reviewed-on: https://code.wireshark.org/review/19387
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Reviewed-by: Michael Mann <mmann78@netscape.net>
It's not used at all in the Qt UI, and the point at which it's used in
the GTK+ UI doesn't involve reading configuration files - between it and
RA_INTERFACES, some mostly quick command-line argument processing is
done.
Change-Id: Ib4de97c8b6dba658a6e9d8f596b551f7c45739d2
Reviewed-on: https://code.wireshark.org/review/19551
Reviewed-by: Guy Harris <guy@alum.mit.edu>
The GTK+ version, at least, crashes if there are more "doing XXX..."
items put up than the calculated count, so, now that we're putting up
items for extcap binaries, we need to count the extcap binaries.
Clean up some stuff we found doing this (indentation, a _U_ on something
that's used).
Change-Id: I1f88042b64ce4b9ae352de37689677c694e3770b
Reviewed-on: https://code.wireshark.org/review/19549
Reviewed-by: Guy Harris <guy@alum.mit.edu>
We save a list of dissectors that are disabled through the Enabled Protocols
dialog. This is because we assume dissectors are enabled by default.
For dissectors that are disabled by default, we have no way to keep them
enabled through the Enabled Protocols dialog. A dissector that defaults
to being disabled has to be reset to enabled each time Wireshark is launched.
Add a list similar to the disabled list for enabling dissectors that are
disabled by default.
This mostly applies to post-dissectors.
Change-Id: I31a8d97a9fdbc472fe2a8666384e0f8786bb8e9f
Reviewed-on: https://code.wireshark.org/review/19405
Petri-Dish: Michael Mann <mmann78@netscape.net>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
As per coverity multiple CIDs.
Change-Id: I30b5a5e45de2fc9b3c563c56151bd33483c754cd
Reviewed-on: https://code.wireshark.org/review/19379
Reviewed-by: Roland Knall <rknall@gmail.com>
Petri-Dish: Roland Knall <rknall@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Add WinSparkle can_shutdown and shutdown_request callbacks which are
called prior to running the installer. Reject updates when we have
unsaved information. Add notes about possible improvements.
Ping-Bug: 9687
Ping-Bug: 12989
Change-Id: Ia126244b311417aa3105ea8136f186adc2745445
Reviewed-on: https://code.wireshark.org/review/19244
Reviewed-by: Gerald Combs <gerald@wireshark.org>
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
Those sizes are limited by the packet sizes we support, and we only
support a maximum packet size of 2^32.
This squelches some compiler warnings.
Remove some casts that this renders unnecessary.
Change-Id: Id9a7bcf8c2ce30bbed7be6c0e28deb9cf38002e0
Reviewed-on: https://code.wireshark.org/review/19279
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
It's GTK+-only, so it shouldn't be in ui/ui_util.h. Get rid of the
unused Qt packet list implementation of it.
Change-Id: Ia9f8fe2209939dff5244e6948c36f29509340f68
Reviewed-on: https://code.wireshark.org/review/19226
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Just directly call the packet_list_select_ routine from the GTK+ code.
Change-Id: I9146fb968c407d6186b146a86aa34678765f7352
Reviewed-on: https://code.wireshark.org/review/19225
Reviewed-by: Guy Harris <guy@alum.mit.edu>
This function can be used by code outside ui (eg. extcap).
Ping-Bug: 13218
Change-Id: Ic11f7acebefeaf777692df044ebff9b1bc387aa3
Reviewed-on: https://code.wireshark.org/review/19178
Petri-Dish: Dario Lombardo <lomato@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
Integrate the Spandsp library for G.722 and G.726 support. Adds support
for G.722 and all eight variants of G.726.
Note: this also fixes a crash in Qt (buffer overrun, reading too much
data) caused by confusion of the larger output buffer (resample_buff)
with the smaller input buffer (decode_buff). It was not triggered before
because the sample rate was always 8k, but with the addition of the new
codecs, a different sample rate became possible (16k).
Fix also a crash which occurs when the RTP_STREAM_DEBUG macro is enabled
and the VOIP Calls dialog is opened (the begin frame, start_fd, is not
yet known and therfore a NULL dereference could occur).
Passes testing (plays normally without bad RTP timing errors) with
SampleCaptures files: sip-rtp-g722.pcap and sip-rtp-g726.pcap. Tested
with cmake (Qt), autotools (Qt and GTK+) with ASAN enabled.
Bug: 5619
Change-Id: I5661908d193927bba50901079119eeff0c04991f
Reviewed-on: https://code.wireshark.org/review/18939
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Make the init_progfile_dir() call unconditionally, even if plugins
aren't supported, as that doesn't necessarily mean nobody uses the
directory containing the executable.
Report the error the same way in all programs, and free the error string
after we're finished with it.
Make the error - and the comment before the code - reflect what
init_progfile_dir() is actually doing (the goal is to get the full
pathname of the directory *containing* the executable; that's generally
done by getting the pathname of the executable and stripping off the
name of the executable, but that's won't necessarily always be the
case). Also note for TShark that we won't be able to capture traffic,
just as we do for Wireshark (if we don't have the pathname of the
program file, we don't have a pathname to use to find dumpcap).
Have the plugin scanner just fail silently if we weren't able to get the
plugin directory path, so we don't have to worry about calling it if
init_progfile_dir() fails.
Clean up white space while we're at it.
Change-Id: I8e580c719aab6fbf74a764bf6629962394fff7c8
Reviewed-on: https://code.wireshark.org/review/19076
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Have programs that use libwiretap call that routine rather than
separately calling some or all of init_open_routines(),
wtap_register_plugin_types(), and wtap_opttypes_initialize().
Also don't have routines internal to libwiretap call those. Yes, this
means doing some initialization work when it isn't necessary, but
scattering on-demand calls throughout the code is a great way to forget
to make those calls.
Change-Id: I5828e1c5591c9d94fbb3eb0a0e54591e8fc61710
Reviewed-on: https://code.wireshark.org/review/19069
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Show codec libraries in About dialog, this should give the user a clue
of what codecs are available.
SBC is already supported, Spandsp (for G.722/G.726) is work in progress.
Change-Id: Iebc4d9c9fae619a442e06c8afc780a420aa3971b
Reviewed-on: https://code.wireshark.org/review/18978
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Not all codecs require the plugin infrastructure. For example, G.711U/A
is a built-in codec. Allow such functionality to be registered even if
plugin support is disabled.
Change-Id: I2505cc9955e7953268ec0739531278921f70a771
Reviewed-on: https://code.wireshark.org/review/18977
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
This is similar to what we have for opening a dump file - one API that
uses the file name as specified, one that creates a temporary file and
provides the file name, and one that uses the standard output.
All of those APIs handle closing the output file.
Change-Id: I56beea7be347402773460b9148ab31a8f8bc51e1
Reviewed-on: https://code.wireshark.org/review/19059
Reviewed-by: Guy Harris <guy@alum.mit.edu>
A new "--export-object <protocol>,<destdir>" option is added to tshark.
This required refactoring Export Object behavior in all GUIs to give the
export object handling to the dissector, rather than the ui layer.
Included in the refactoring was fixing some serious memory leaks in Qt
Export Object dialog, crash due to memory scope issues in GTK Export
Object dialog, and addition sorting column feature in Qt dialog (set
up by creating a widget to manage the items that were previously
leaking memory)
Bug: 9319
Ping-Bug: 13174
Change-Id: I515d7662fa1f150f672b1476716f347ec27deb9b
Reviewed-on: https://code.wireshark.org/review/18927
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Tested-by: Michael Mann <mmann78@netscape.net>
Reviewed-by: Michael Mann <mmann78@netscape.net>
Use the model from the 2.0 branch and earlier that only "tapped" the
follow data in a single location. This fixes duplicate data for
reassembled data and handles out-of-order packets.
Bug: 12855
Change-Id: I5268f13e3c08e9271acf026b859de693ad794c94
Reviewed-on: https://code.wireshark.org/review/18368
Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
Have them handle -d, -t, --disable-protocol, --disable-heuristic, and
--enable-heuristic for TShark and both flavors of Wireshark.
Change-Id: I612c276b1f9df8a2092202d23ab3d48be7857e85
Reviewed-on: https://code.wireshark.org/review/18583
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Change-Id: Ie38bdd27d9e3810e3a64b985dfd5621a3aa6d073
Reviewed-on: https://code.wireshark.org/review/18445
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Michael Mann <mmann78@netscape.net>
AC_CHECK_MEMBER() and AC_CHECK_MEMBERS() use a standard name for the
{structurename} being the name of the structure type, complete with
"struct" if a typedef wasn't used, and with all letters mapped to upper
case, and with {membername} being the name of the structure member, with
all letters mapped to upper case.
check_struct_has_member() lets you choose the name; choose the same name
that the autoconf macros use, and fix the code to check for them.
Change-Id: Ifb3cf65e7e94907ad0a2f8aacca0c21a531f0c5b
Reviewed-on: https://code.wireshark.org/review/18382
Petri-Dish: Guy Harris <guy@alum.mit.edu>
Reviewed-by: Guy Harris <guy@alum.mit.edu>
That's the time the file's inode last changed, so size changes,
permission changes, etc. affect it. It's *not* the time the file was
created; most UN*Xes don't provide that. Newer versions of FreeBSD,
NetBSD, OpenBSD, and macOS do, but other UN*Xes don't appear to.
On Windows, at least according to Microsoft's documentation, st_ctime
*is* the creation time. Hopefully that's not the result of confusion on
the part of somebody at Microsoft.
Change-Id: I20743703f6ef66e40dff9004dc91bed46af6fad0
Reviewed-on: https://code.wireshark.org/review/18378
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Pro tip: if you're modifying GTK+ code, make sure you explicitly
configure to build the GTK+ version.
Change-Id: Ia24321a5037dff67d9acf1444fa0b74f5aed089e
Reviewed-on: https://code.wireshark.org/review/18371
Reviewed-by: Guy Harris <guy@alum.mit.edu>
And some comments in the case where we're converting the result of
time() - if your machine's idea of time predates January 1, 1970,
00:00:00 UTC, it'll crash on Windows, but that's not a case where a
*file* can cause the problem due either to a bad file time stamp or bad
time stamps in the file.
Change-Id: I837a438e4b875dd8c4f3ec2137df7a16ee4e9498
Reviewed-on: https://code.wireshark.org/review/18369
Reviewed-by: Guy Harris <guy@alum.mit.edu>
This patch introduces new APIs to allow dissectors to have a preference for
a (TCP) port, but the underlying data is actually part of Decode As functionality.
For now the APIs are intentionally separate from the regular APIs that register a
dissector within a dissector table. It may be possible to eventually combine the
two so that all dissectors that register with a dissector table have an opportunity
to "automatically" have a preference to adjust the "table value" through the
preferences dialog.
The tcp.port dissector table was used as the guinea pig. This will eventually be
expanded to other dissector tables as well (most notably UDP ports). Some
dissectors that "shared" a TCP/UDP port preference were also converted. It also
removed the need for some preference callback functions (mostly when the callback
function was the proto_reg_handoff function) so there is cleanup around that.
Dissectors that has a port preference whose default was 0 were switched to using
the dissector_add_for_decode_as_with_preference API rather than dissector_add_uint_with_preference
Also added comments for TCP ports used that aren't IANA registered.
Change-Id: I99604f95d426ad345f4b494598d94178b886eb67
Reviewed-on: https://code.wireshark.org/review/17724
Reviewed-by: Michael Mann <mmann78@netscape.net>
Instead of checking for the boolean "FALSE", just set an empty string.
This avoids the need to check for WERROR_COMMON_FLAGS before using it.
The transformation is the same for all files, remove
"if (WERROR_COMMON_FLAGS)" and "endif()", reindent and add quotes (since
we have a string here and not a list).
Modelines have been added where missing.
Change-Id: I0ab05ae507c51fa77336d49a99a226399cc81b92
Reviewed-on: https://code.wireshark.org/review/17997
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: João Valverde <j@v6e.pt>
Reviewed-by: Dario Lombardo <lomato@gmail.com>
Tested-by: Dario Lombardo <lomato@gmail.com>
Both the QT and GTK wireless toolbar used "-1" to indicate unused
channel parameters. This was an undocumented feature that recently
stopped working. Use the now documented way (NULL) to specify that
some parameters are not used.
Bug: 12896
Change-Id: I4a93a15ba1d880592b355b7eca155632a4b92ea0
Reviewed-on: https://code.wireshark.org/review/17700
Reviewed-by: Gilbert Ramirez <gram@alumni.rice.edu>
Petri-Dish: Gilbert Ramirez <gram@alumni.rice.edu>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Anders Broman <a.broman58@gmail.com>
In commit v2.3.0rc0-117-g485bc45 (backported to v2.2.0rc0-44-g66721ca),
extcap_prefs_dynamic_vals and extcap_cleanup were added in an attempt to
address dangling pointers.
Unfortunately it is not sufficient:
- A pointer to the preference value is stored in extcap_arg and passed
to the prefs API, but this extcap_arg structure can become invalid
which result in use-after-free whenever the preference is accessed.
- On exit, a use-after-free occurs in prefs_cleanup when the preference
value is being checked.
As the preference subsystem actually manages the memory for the string
value and consumers should only provide a pointer where the value can be
stored, convert the char* field in extcap to char**. This has as
additional benefit that values are not limited to 256 bytes anymore.
extcap_cleanup is moved after epan_cleanup to ensure that prefs_cleanup
does not operate on dangling pointers.
Crash is reproducible under ASAN with: tshark -i randpkt
Ping-Bug: 12183
Change-Id: Ibf1ba1102a5633aa085dc278a12ffc05a4f4a34b
Reviewed-on: https://code.wireshark.org/review/17631
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Roland Knall <rknall@gmail.com>
A single name resolution checkbox was added to the file dialog way back
in 2000 in g0f7cf64. At that time it was needed because resolution was
synchronous and could drastically affect your load time. Since then
we've added asynchronous name resolution and more recently made it
mandatory (ge005bc8). We've also added more name resolution checkboxes
and other controls.
Remove the name resolution checkboxes. You can just as easily change
resolution options before or after opening a file and they take up
valuable real estate.
Combine the size and packets in the Qt and Win32 dialogs and
pretty-print the size. Combine the start and elapsed times in the Qt,
Win32, and GTK+ dialogs. This lets us shrink the custom areas of the
file dialogs even further. Make the default file type combo item more
descriptive.
Change-Id: Id770adc0f284a4c7f08ee5a7db84f8435f4bf907
Reviewed-on: https://code.wireshark.org/review/17597
Tested-by: Gerald Combs <gerald@wireshark.org>
Petri-Dish: Gerald Combs <gerald@wireshark.org>
Reviewed-by: Gerald Combs <gerald@wireshark.org>
That's a less gross hack to suppress load failures due to not having
libwiretap than providing a no-op failure-message routine, as it at
least allows other code using a failure-message routine, such as
cmdarg_err() and routines that call it, to be used.
We really should put libwiretap and libwireshark plugins into separate
subdirectories of the plugin directories, and avoid even looking at
libwireshark plugins in programs that don't use libwireshark.
Change-Id: I0a6ec01ecb4e718ed36233cfaf638a317f839a73
Reviewed-on: https://code.wireshark.org/review/17506
Reviewed-by: Guy Harris <guy@alum.mit.edu>
Removed in v1.99.2rc0-399-g43f09e6, but some includes are still needed.
Basically I looked for the functions defined in the
ui/gtk/old-gtk-compat.h header file that were in files below ui/gtk/.
Then I matched it against the removed part in the above commit and added
the header back at its old place. In two other cases, the header was
only needed for Windows.
If the above commit did not reference the file, I checked which function
was in use and added the header on top, removing redundant conditionals.
Reported and tested by a user of GTK+ 2.12.12 on top of v2.0.5.
Change-Id: I649eec1e5531070f88c99d893c4920306f56d849
Reviewed-on: https://code.wireshark.org/review/17371
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
32-bit glib before 2.31.2 expand GUINT_TO_POINTER(x) as (gpointer)x.
add explicit cast since curr_layer_num is 8 bits. Fixes
v1.99.10rc0-179-g1d7bcb2.
Storing a 64-bit integer into a 32-bit pointer makes the compiler
complain. Add explicit cast. Fixes v1.99.1rc0-76-ged0b19b.
Change-Id: I75fdf17882a0f5ddce7d3b3e74b1bf80ff6cd4ae
Reviewed-on: https://code.wireshark.org/review/17417
Petri-Dish: Peter Wu <peter@lekensteyn.nl>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Peter Wu <peter@lekensteyn.nl>
According to documentation of uat_copy_cb_t, if uat->copy_cb is NULL,
memcpy() should be used.
This affected IKEv1 and IKEv2 UAT tables (possibly others).
Change-Id: I27ebdc08385d260945699f101e714d3abb288b22
Reviewed-on: https://code.wireshark.org/review/17245
Petri-Dish: Pascal Quantin <pascal.quantin@gmail.com>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: Pascal Quantin <pascal.quantin@gmail.com>
Setting our compiler warning flags in CMAKE_C_FLAGS does not allow
using different flags per target.
Allow for that possibility by setting the internal WS_WARNINGS_{C,CXX}_FLAGS
and using the COMPILE_OPTIONS property to set them.
This change is just setting mechanism and there should be no difference
in generated warnings.
The check_X_compiler_flag cmake test is changed to test each flag individually.
We need a list, not a space separated string, and the aggregate test is not
significant.
Change-Id: I59fc5cd7e130c7a5e001c598e3df3e13f83a6a25
Reviewed-on: https://code.wireshark.org/review/17150
Petri-Dish: João Valverde <j@v6e.pt>
Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org>
Reviewed-by: João Valverde <j@v6e.pt>
Kevin Hogan