Wolfram Sang [Tue, 6 Mar 2018 13:48:54 +0000 (14:48 +0100)]
Fix #928 by catching errors if config_list() fails
MainBar::config_list() may fail so catch errors and don't segfault.
Discovered using an OLS with unknown firmware not supporting metadata.
Signed-off-by: Wolfram Sang <wsa@the-dreams.de>
Soeren Apel [Mon, 5 Mar 2018 22:45:46 +0000 (23:45 +0100)]
Fix #1132 by passing segment IDs, not segment instances
Passing segment instances fails because this creates a race condition.
When a long conversion is taking place, the SignalBase::samples_added
signal is called often but since it's in a separate thread, the calls
are queued and aren't executed immediately. Now if the conversion is
restarted - for example as a result of a changed conversion threshold -
then the segments holding the converted data are destroyed, rendering
the pointers submitted as parameters to samples_added invalid.
Once the signal queue is processed, those invalid pointers will be
accessed and PV segfaults.
Since the signal queue can neither be emptied nor flushed, this
leaves only two sensible choices:
1) Signal samples_added less often, thereby reducing the chance of
signals being queued
2) Supply the segment ID instead of the segment instance as that's
essentially the only thing we currently care about - in fact, the
only user of samples_added (ViewBase::on_samples_added) uses the
instance to query only this
As #1 is only a band-aid and not a waterproof solution, I chose
to go with #2.
Soeren Apel [Sat, 3 Mar 2018 23:25:07 +0000 (00:25 +0100)]
Session: Fix trigger handling
Soeren Apel [Fri, 23 Feb 2018 23:18:27 +0000 (00:18 +0100)]
GlobalSettings: Remove unneeded include/using
Soeren Apel [Fri, 23 Feb 2018 23:13:13 +0000 (00:13 +0100)]
Session: Add error messages
Soeren Apel [Wed, 21 Feb 2018 17:46:29 +0000 (18:46 +0100)]
Tiny whitespace fix
Soeren Apel [Sat, 10 Feb 2018 09:52:16 +0000 (10:52 +0100)]
Fix #1038 by catching invalid cursor ranges
Uwe Hermann [Sat, 10 Feb 2018 21:01:26 +0000 (22:01 +0100)]
MinGW: Fix a compile error due to a missing #include.
In file included from [...]/pv/globalsettings.cpp:20:0:
[...]/pv/globalsettings.hpp:31:12: error: 'std::function' has not been declared
using std::function;
^
test/CMakeFiles/pulseview-test.dir/build.make:88: recipe for target 'test/CMakeFiles/pulseview-test.dir/__/pv/globalsettings.cpp.obj' failed
make[2]: *** [test/CMakeFiles/pulseview-test.dir/__/pv/globalsettings.cpp.obj] Error 1
make[2]: *** Waiting for unfinished jobs....
[ 5%] Building CXX object CMakeFiles/pulseview.dir/pv/globalsettings.cpp.obj
In file included from [...]/pv/globalsettings.cpp:20:0:
[...]/pv/globalsettings.hpp:31:12: error: 'std::function' has not been declared
using std::function;
^
Add #include <functional> everywhere where std::function is used. Only
the occurence in globalsettings.hpp was causing an actual build failure
on MinGW, though.
Martin Errenst [Sat, 3 Feb 2018 23:57:07 +0000 (00:57 +0100)]
fix PulseView compilation with ENABLE_DECODE set to OFF
Gerhard Sittig [Sun, 4 Feb 2018 13:54:32 +0000 (14:54 +0100)]
MainBar: fixup file extension filter in "Import File" dialog
The previous implementation used a "*" filter when a file of any other
format than srzip got imported. This happened to be a duplicate of
"All files", and ignored the list of filename extensions provided by
the input formats.
This change does respect the input format's file extensions, and copes
with the lack of such a list (raw binary), as well as lists that have
one (most formats), or multiple extensions (raw analog).
This fixes bug #1039.
Another byproduct of the change is that extensions and their decoration
(separators, parentheses) move outside of tr() calls. These technical
details shall not concern translators, and translations for human
languages shall not break the filter mechanism.
This implementation might be "too complex, computationally expensive".
But it works, and fixes an issue, and the code path executes seldom and
waits for user interaction anyway. Cost reduction can get applied later.
Soeren Apel [Thu, 8 Feb 2018 15:32:19 +0000 (16:32 +0100)]
Fix #1089 by updating the signal labels and group labels
Group names currently don't change, so this doesn't do much
but they may change in the future, so it's useful to do anyway.
Cenkron [Mon, 15 Jan 2018 02:04:17 +0000 (20:04 -0600)]
Ruler tick legends don't partly disappear when scrolled off the ends
Soeren Apel [Sun, 21 Jan 2018 13:28:27 +0000 (14:28 +0100)]
Rework the callback mechanism for the global settings
Up to now, registered callbacks could not be unregistered
because std::function does not permit comparing for equality.
Using an interface class removes the need for std::function,
making the mechanism a little less elegant but at least fully
functional.
Soeren Apel [Fri, 19 Jan 2018 20:46:21 +0000 (21:46 +0100)]
View: Fully integrate the "zero trigger time" setting
Soeren Apel [Wed, 7 Feb 2018 09:51:30 +0000 (10:51 +0100)]
Session/View: Save triggers in a list and use it
Soeren Apel [Wed, 24 Jan 2018 08:35:23 +0000 (09:35 +0100)]
Session: Clean up header a little
Soeren Apel [Tue, 6 Feb 2018 18:28:37 +0000 (19:28 +0100)]
Adjust unit tests
Cenkron [Wed, 10 Jan 2018 15:10:58 +0000 (09:10 -0600)]
Change the number of minor ticks to be either 4 or 5
Depends on the current tick interval size, in the sense of the 1, 2, 5
sequence. It was originally always 4.
Cenkron [Wed, 10 Jan 2018 02:28:11 +0000 (20:28 -0600)]
Provide a settings checkbox for showing zero at the trigger
Gerhard Sittig [Sun, 28 Jan 2018 20:02:54 +0000 (21:02 +0100)]
DecodeSignal: only pass non-zero sample rate to decoders
As reported in bug #1118 not all input sources provide a samplerate, and
decoder instances may not cope with a rate spec of 0. Only pass non-zero
sample rates to the decoder stack. This improves robustness in addition
to the specific fix for #1118 in the decoders' implementations.
Gerhard Sittig [Wed, 24 Jan 2018 20:22:43 +0000 (21:22 +0100)]
doc: Update pulseview(1) manpage, add --no-scan option (-D)
Gerhard Sittig [Wed, 24 Jan 2018 20:11:50 +0000 (21:11 +0100)]
main: introduce -D cmdline option, don't auto-scan for devices
Add a -D command line option which skips auto-detection of devices upon
startup. This can speedup program startup for setups with known devices,
and allows to skip the scan for troubled drivers which might break the
startup phase.
This -D option can be combined with -d specs, either presence or absence
of -d is acceptable when -D is specified. Users still can interactively
scan for and use devices after program startup.
This resolves bug #1116.
Cenkron [Fri, 12 Jan 2018 18:43:27 +0000 (12:43 -0600)]
win32: Use -mwindows only for non-Debug targets.
Direct cmake to apply the -mwindows linker switch based on a cmake
command line argument rather than by using a patch file. The command
line argument is -DCMAKE_BUILD_TYPE=xxxxx and applies the linker switch
only to WIN32 builds that are not Debug.
Gerhard Sittig [Tue, 2 Jan 2018 18:51:22 +0000 (19:51 +0100)]
DecodeSignal: Void a session pointer after destroy() call
Fixup an error path when starting a created decoder session fails.
Gerhard Sittig [Tue, 2 Jan 2018 21:13:06 +0000 (22:13 +0100)]
main.cpp: check log level range before using its value
(Silently) ignore invalid -l loglevel specs. This avoids a crash
reported in bug #1071.
Soeren Apel [Sun, 7 Jan 2018 20:48:54 +0000 (21:48 +0100)]
Fix unit tests
Soeren Apel [Sun, 7 Jan 2018 20:09:55 +0000 (21:09 +0100)]
Add segment display mode UI controls and some related changes
Soeren Apel [Sun, 7 Jan 2018 20:00:47 +0000 (21:00 +0100)]
Add segment display mode icons
Soeren Apel [Sun, 7 Jan 2018 19:15:19 +0000 (20:15 +0100)]
Session: Make get_segment_count() consider all segments
Also changes the type for the getters to uint32_t.
Soeren Apel [Fri, 5 Jan 2018 19:48:20 +0000 (20:48 +0100)]
Supply the segment ID when adding samples to optimize trace painting
Use case is as follows:
- Capture 20+ segments with ~500kS each
- Afterwards, enable conversion for a channel
Without this change, the converted logic will be repainted
20++ times because we are only told that new samples were
added but not which segment.
With this change, the logic trace is only painted when we
see that samples were added to the segment we're showing.
Soeren Apel [Fri, 5 Jan 2018 14:24:41 +0000 (15:24 +0100)]
Add segment_id to all segment classes
Soeren Apel [Wed, 3 Jan 2018 23:30:28 +0000 (00:30 +0100)]
LogicSegment: Don't use new/delete in get_unpacked_sample()
Soeren Apel [Wed, 3 Jan 2018 16:14:51 +0000 (17:14 +0100)]
LogicSegment: Limit end in get_subsampled_edges() if needed
This is necessary because there is a race condition between
adding samples and trace repainting.
Soeren Apel [Tue, 2 Jan 2018 10:07:36 +0000 (11:07 +0100)]
SignalBase: Don't terminate conversion when there's no data
Instead of terminating, we wait instead.
We do this because SignalBase::on_samples_added() somehow doesn't
reliably see that there's no conversion thread active anymore.
conversion_thread_.joinable() returns true when the thread was
already terminated for whatever reason, resulting in on_samples_added()
trying to notify a non-existant thread.
Soeren Apel [Thu, 28 Dec 2017 16:37:48 +0000 (17:37 +0100)]
Implement Trace::ShowLastCompleteSegmentOnly display mode
Soeren Apel [Wed, 27 Dec 2017 09:31:30 +0000 (10:31 +0100)]
DecodeSignal: Restructure metadata handling
Instead of handling the metadata separately from the mux/decode segments,
it's much neater to handle both together. Doing this also allows us to
remove the need for query_input_metadata() since we're taking the metadata
from the muxed logic segments.
Soeren Apel [Thu, 21 Dec 2017 22:10:36 +0000 (23:10 +0100)]
SignalBase: Implement multi-segment conversion
Soeren Apel [Tue, 19 Dec 2017 19:20:57 +0000 (20:20 +0100)]
Finalize segment decoding
Soeren Apel [Wed, 13 Dec 2017 16:19:29 +0000 (17:19 +0100)]
View: Fixes related to multi-segment display
Soeren Apel [Fri, 15 Dec 2017 14:16:41 +0000 (15:16 +0100)]
DecodeSignal: Mux all segments
Soeren Apel [Wed, 29 Nov 2017 08:32:24 +0000 (09:32 +0100)]
Prepare multi-segment protocol decoding ability
Soeren Apel [Wed, 20 Dec 2017 19:22:27 +0000 (20:22 +0100)]
DecodeSignal: Allow muxed logic data to be cached
Soeren Apel [Thu, 23 Nov 2017 17:30:26 +0000 (18:30 +0100)]
DecodeSignal: Rename some items
Soeren Apel [Tue, 21 Nov 2017 21:07:09 +0000 (22:07 +0100)]
Begin PD multisegment support
Soeren Apel [Tue, 21 Nov 2017 21:07:09 +0000 (22:07 +0100)]
Move current_segment_ to Trace
Soeren Apel [Mon, 20 Nov 2017 21:50:53 +0000 (22:50 +0100)]
Flesh out segment display mode handling
Soeren Apel [Tue, 7 Nov 2017 09:56:25 +0000 (10:56 +0100)]
Implement segment display mode handling and update notifications
Soeren Apel [Fri, 15 Dec 2017 22:56:39 +0000 (23:56 +0100)]
Don't set SR_CONF_LIMIT_FRAMES to 1
probonopd [Fri, 24 Nov 2017 20:56:17 +0000 (20:56 +0000)]
Make it pass desktop-file-validate
https://travis-ci.org/AppImage/appimage.github.io/builds/
306921880#L695
Soeren Apel [Tue, 12 Dec 2017 22:31:35 +0000 (23:31 +0100)]
DecodeSignal: Make sure bit IDs are always up-to-date
https://www.youtube.com/watch?v=mw6Bd6Hv168 shows what
happens when this isn't done.
Philipp Marek [Sat, 18 Nov 2017 12:53:24 +0000 (13:53 +0100)]
Session: Fix mismatched delete operator.
Mismatched free() / delete / delete []
at 0x4C2D31B: operator delete(void*) (vg_replace_malloc.c:576)
by 0x1C76D0: operator() (unique_ptr.h:78)
by 0x1C76D0: ~unique_ptr (unique_ptr.h:268)
by 0x1C76D0: pv::data::AnalogSegment::append_interleaved_samples(float const*, unsigned long, unsigned long) (analogsegment.cpp:78)
...
Address 0x1d123f70 is 0 bytes inside a block of size 312 alloc'd
at 0x4C2C97F: operator new[](unsigned long) (vg_replace_malloc.c:423)
by 0x1C765E: pv::data::AnalogSegment::append_interleaved_samples(float const*, unsigned long, unsigned long) (analogsegment.cpp:78)
by 0x1AFDD5: pv::Session::feed_in_analog(std::shared_ptr<sigrok::Analog>) (session.cpp:1142)
Philipp Marek [Sat, 18 Nov 2017 12:53:12 +0000 (13:53 +0100)]
.gitignore: Don't care about the doxygen-generated files.
Philipp Marek [Fri, 17 Nov 2017 10:17:36 +0000 (11:17 +0100)]
Session: Fix mismatched delete operator.
Valgrind reported this here:
Mismatched free() / delete / delete []
at 0x4C2D31B: operator delete(void*) (vg_replace_malloc.c:576)
by 0x1C76D0: operator() (unique_ptr.h:78)
by 0x1C76D0: ~unique_ptr (unique_ptr.h:268)
by 0x1C76D0: pv::data::AnalogSegment::append_interleaved_samples(float const*, unsigned long, unsigned long) (analogsegment.cpp:78)
by 0x1AFDD5: pv::Session::feed_in_analog(std::shared_ptr<sigrok::Analog>) (session.cpp:1142)
by 0x1B043C: pv::Session::data_feed_in(std::shared_ptr<sigrok::Device>, std::shared_ptr<sigrok::Packet>) (session.cpp:1187)
...
Address 0x1cbf6b90 is 0 bytes inside a block of size 940 alloc'd
at 0x4C2C97F: operator new[](unsigned long) (vg_replace_malloc.c:423)
by 0x1C765E: pv::data::AnalogSegment::append_interleaved_samples(float const*, unsigned long, unsigned long) (analogsegment.cpp:78)
by 0x1AFDD5: pv::Session::feed_in_analog(std::shared_ptr<sigrok::Analog>) (session.cpp:1142)
Soeren Apel [Sun, 29 Oct 2017 22:38:06 +0000 (23:38 +0100)]
Rework new segment notification mechanism
Soeren Apel [Fri, 27 Oct 2017 16:44:51 +0000 (18:44 +0200)]
Append new segments to the end, not the beginning of the list
Soeren Apel [Mon, 9 Oct 2017 15:47:19 +0000 (17:47 +0200)]
Session: Query the device's sample rate when needed, not sooner
Soeren Apel [Sat, 23 Sep 2017 06:33:16 +0000 (08:33 +0200)]
DecodeSignal: Remove outdated code and comment
Soeren Apel [Fri, 22 Sep 2017 16:25:06 +0000 (18:25 +0200)]
SignalBase: Make sure PDs are restarted when conv options change
Soeren Apel [Fri, 22 Sep 2017 06:27:56 +0000 (08:27 +0200)]
TraceView: Properly handle the header resizing events
Soeren Apel [Fri, 22 Sep 2017 06:24:17 +0000 (08:24 +0200)]
TraceView: Make header resize state determination reliable
Soeren Apel [Tue, 19 Sep 2017 05:50:50 +0000 (07:50 +0200)]
View: Move comment to where it belongs
Soeren Apel [Tue, 19 Sep 2017 05:48:36 +0000 (07:48 +0200)]
Adjust trace view header width when signal names change
This change is primarily needed because before, newly
created decode signals had a name assigned to them at time
of the constructor call. This changed, and now the name
is empty upon creation, breaking the previously working
header size adjustment.
Soeren Apel [Sat, 9 Sep 2017 07:01:07 +0000 (09:01 +0200)]
DecodeTrace: Reset cached trace properties when decoder stack resets
Soeren Apel [Sat, 9 Sep 2017 06:42:36 +0000 (08:42 +0200)]
DecodeSignal: Break up annotation storage to allow multiple segments
Soeren Apel [Thu, 31 Aug 2017 06:24:53 +0000 (08:24 +0200)]
Implement showing only the selected segment
Soeren Apel [Sun, 10 Sep 2017 19:26:59 +0000 (21:26 +0200)]
Add segment selector UI + helpers
Uwe Hermann [Wed, 2 Aug 2017 06:14:31 +0000 (08:14 +0200)]
rowdata: Use emplace_back() for Annotation objects.
This should eliminate one unnecessary allocation per annotation.
Soeren Apel [Fri, 1 Sep 2017 06:21:10 +0000 (08:21 +0200)]
Rename DecodeSignal::segment_ to logic_mux_segment_ for clarity
Soeren Apel [Fri, 8 Sep 2017 16:14:47 +0000 (18:14 +0200)]
Fix #1024 by changing decode channel assigment to PDs
When assigning the decoder stack channels to the libsrd
instance's channels, channels that had no signal assigned
to them were still assigned anyway. This patch fixes this bug.
After doing this, another subtle bug became apparent:
The mapping between channels and bits in the data stream sent
to the PD was done via DecodeChannel->id. This is however
insufficient as the channels of the decoder stack have
IDs that may or may not match the ID needed for the data
stream. Example:
A PD has 4 channels: A, B, C and D. In PV, those channels
have the IDs 0, 1, 2 and 3. If the user only assigns A and D,
Decoder::create_decoder_inst() will use the IDs 0 and 3 as
the bit positions of those signals in the data stream sent
to libsrd. This is obviously wrong.
Hence, we now use a separate bit_id for this purpose.
Soeren Apel [Mon, 21 Aug 2017 17:58:49 +0000 (19:58 +0200)]
AnalogSignal: Prevent threshold bg from spilling beyond extents
Soeren Apel [Sat, 19 Aug 2017 21:58:36 +0000 (23:58 +0200)]
Implement "show conversion thresholds as dots" mode
Soeren Apel [Sun, 20 Aug 2017 14:18:50 +0000 (16:18 +0200)]
Convert threshold display setting to a multi-value choice
Soeren Apel [Thu, 17 Aug 2017 06:30:29 +0000 (08:30 +0200)]
AnalogSignal: Make threshold neutral area darker
Soeren Apel [Mon, 14 Aug 2017 15:52:43 +0000 (17:52 +0200)]
Session: Rework frame handling
Soeren Apel [Tue, 8 Aug 2017 17:40:11 +0000 (19:40 +0200)]
AnalogSignal: Use setting change handler for threshold display
Soeren Apel [Mon, 7 Aug 2017 19:41:11 +0000 (21:41 +0200)]
AnalogSignal: Draw analog thresholds differently
Instead of using a patterned region on top of the trace,
show the thresholds by using different background colors
for the individual areas: green/grey/red.
Soeren Apel [Sat, 5 Aug 2017 20:38:45 +0000 (22:38 +0200)]
View: Update ruler after restoring session
If we don't do this, the ruler is out-of-sync with the rest
of the view until the session is loaded and goes into stopped
state. At that point, the ruler is updated, which is too late.
Soeren Apel [Wed, 2 Aug 2017 16:42:34 +0000 (18:42 +0200)]
SignalBase: Default to dynamic conversion preset, not custom values
The reason for this change is that when you initially select a
conversion from the channel config popup dialog, the threshold
will be set to "0.0V" or "0.0V/0.0V", respectively.
This is of course not what we want and the root cause is that
when no preset is selected, NoPreset is assumed instead of
DynamicPreset. This patch changes this.
Soeren Apel [Wed, 2 Aug 2017 16:23:32 +0000 (18:23 +0200)]
Use min/max notification for autoranging of analog signals
Soeren Apel [Wed, 2 Aug 2017 07:15:18 +0000 (09:15 +0200)]
Move delayed conversion starter to SignalBase
This way, we can use the same mechanism for changing
min/max as well, preventing multiple successive starts
of the conversion algorithm.
Preventing this is necessary because it makes the UI
stop updating for a significant amount of time, which
we obviously don't want.
Soeren Apel [Tue, 1 Aug 2017 19:38:15 +0000 (21:38 +0200)]
Remove SignalBase::A2LChannel
Soeren Apel [Tue, 1 Aug 2017 19:35:15 +0000 (21:35 +0200)]
Use identifiers for fixed conversion preset values
Soeren Apel [Tue, 1 Aug 2017 18:58:46 +0000 (20:58 +0200)]
Restart min-max-based conversion if min or max change
Uwe Hermann [Tue, 1 Aug 2017 20:19:56 +0000 (22:19 +0200)]
Fix some random clang-tidy warnings.
Soeren Apel [Mon, 31 Jul 2017 22:09:06 +0000 (00:09 +0200)]
Trace view: Make the zero line for analog traces thicker
This is to make it stand out more and make it easily
distinguishable.
Soeren Apel [Mon, 31 Jul 2017 20:31:54 +0000 (22:31 +0200)]
Implement graphical display of A2L thresholds
Soeren Apel [Sat, 29 Jul 2017 17:44:59 +0000 (19:44 +0200)]
Typo fix: treshold -> threshold
Soeren Apel [Sat, 29 Jul 2017 17:41:31 +0000 (19:41 +0200)]
AnalogSignal: Reword "traces to show" option
Soeren Apel [Sat, 29 Jul 2017 17:33:47 +0000 (19:33 +0200)]
AnalogSignal: Use pv::data::SignalBase namespace
Soeren Apel [Mon, 31 Jul 2017 20:31:20 +0000 (22:31 +0200)]
Implement A2L presets and custom threshold handling
Soeren Apel [Tue, 25 Jul 2017 14:09:08 +0000 (16:09 +0200)]
Use getter for the conversion type instead of a local copy
Don't ask me why I used a local copy in the first place,
even I can't figure it out :)
Soeren Apel [Mon, 24 Jul 2017 06:57:51 +0000 (08:57 +0200)]
SignalBase: Minor doxygen comments
Soeren Apel [Sun, 23 Jul 2017 20:17:16 +0000 (22:17 +0200)]
Trace View: Add doxygen comments to some classes
Soeren Apel [Sun, 23 Jul 2017 07:35:25 +0000 (09:35 +0200)]
Improve hover point signaling
Two issues here:
1) it's bad style to have an event handler for an event
that was triggered in the same class
2) supplying the current hover point along with the event
is a sensible thing to do
Soeren Apel [Thu, 20 Jul 2017 06:45:06 +0000 (08:45 +0200)]
Remove signal scale handle
It will be replaced by a different mechanism.
Soeren Apel [Tue, 18 Jul 2017 18:40:33 +0000 (20:40 +0200)]
TriggerMarker: Don't return "valid" drag point coordinates
Soeren Apel [Tue, 18 Jul 2017 15:51:01 +0000 (17:51 +0200)]
Rename ViewItem::point() to ViewItem::drag_point() for clarity
Uwe Hermann [Fri, 21 Jul 2017 05:08:06 +0000 (07:08 +0200)]
INSTALL/CMakeLists.txt: Bump libsigrokdecode requirement to >= 0.6.0.
PulseView now relies on e.g. the fact that multiple libsigrokdecode
calls from different threads don't cause issues (which has not been the
case prior to libsigrokdecode 0.6.0).
Uwe Hermann [Fri, 21 Jul 2017 05:07:10 +0000 (07:07 +0200)]
CMakeLists.txt: Bump libsigrokcxx requirement to 0.6.0.
This was already documented in INSTALL, but the check wasn't updated.
Uwe Hermann [Wed, 19 Jul 2017 18:55:59 +0000 (20:55 +0200)]
Fix the build with ENABLE_DECODE=n.
pv/dialogs/settings.cpp: In member function ‘QWidget* pv::dialogs::Settings::get_decoder_settings_form(QWidget*) const’:
pv/dialogs/settings.cpp:218:1: error: no return statement in function returning non-void [-Werror=return-type]