1
0
mirror of https://github.com/f4exb/sdrangel.git synced 2024-11-27 02:09:14 -05:00
Commit Graph

2634 Commits

Author SHA1 Message Date
f4exb
625f0eb72c File record default file name fix (3): web API 2018-05-09 18:59:39 +02:00
f4exb
775b2a270d File record default file name fix (2) 2018-05-09 17:39:48 +02:00
f4exb
1ee75f127d SDRDaemon source: fixed UDP socket readyRead signal connection (removed queued connection flag) 2018-05-09 09:57:26 +02:00
f4exb
0981d04904 File record default file name fix (1) 2018-05-08 11:03:09 +02:00
f4exb
b32330780b Level meters: fixed text position and make sure appropriate font is used 2018-05-07 03:38:57 +02:00
f4exb
97677075b1 AM demod: fixed delayed squelch 2018-05-06 22:56:24 +02:00
f4exb
b9a19577c7 SSB demod: squelch: change ramp up/down constant to half of the averaging time 2018-05-06 14:50:13 +02:00
f4exb
56c0aaedcd Mag AGC: corrected step calculation. Added method to combine step up and down smoothing 2018-05-06 02:39:39 +02:00
f4exb
e0d3c77fc3 Sample source plugins: use liberation font 2018-05-06 00:55:42 +02:00
f4exb
8281058b24 Sample sink plugins: use liberation font 2018-05-06 00:40:09 +02:00
f4exb
bcc3476e9d Channel Tx plugins: use liberation font 2018-05-06 00:30:43 +02:00
f4exb
c43ecc8afe Channel Rx plugins: use liberation font 2018-05-05 18:12:05 +02:00
f4exb
bec196953e Corrected default font size 2018-05-05 15:21:49 +02:00
f4exb
5ce49a387d Changed font for upward compatibiilty (test) 2018-05-05 15:10:19 +02:00
f4exb
359af254f0 SSB modulator: fixed issue #167: typo in remove Tx registration method 2018-05-03 00:28:27 +02:00
f4exb
1db2da3b60 NFM demod: fixed squelch 2018-05-02 01:22:34 +02:00
f4exb
c7db2a3ca7 Bumped plugins version 2018-05-02 00:29:18 +02:00
f4exb
058f3d5af8 RTLSDR: optimized decimator 2018-05-02 00:24:50 +02:00
f4exb
19c32b4354 Removed 24/16 bit differentiation on Decimator instantiation where possible 2018-05-01 23:57:12 +02:00
f4exb
d735025c6c With global adoption of even/odd decimators the accu type must be 32 not 64 bits 2018-04-29 10:38:25 +02:00
f4exb
8d565f8187 SSB mod: fixed span slider aspect and bulky low/high cut slider group 2018-04-27 07:09:37 +02:00
f4exb
34ff36926e SSB demod: fixed span slider aspect 2018-04-27 06:59:36 +02:00
f4exb
45aa323cf6 SSB demod: make the filter sliders less bulky with only one scale on the high cutoff 2018-04-27 06:41:48 +02:00
f4exb
b03e9c59cb Benchmarking: implemented decimator float to int test 2018-04-25 01:44:54 +02:00
f4exb
2efa7ab594 DSD demod: allow use of audio rates that are integer multiples of 8k other than 48k (x2,3,4,5) 2018-04-23 01:04:47 +02:00
f4exb
486468afe0 Updated Debian changelog and DSD demod doc 2018-04-22 18:49:58 +02:00
f4exb
06cd90e354 AM demod: use buffered delay squelch 2018-04-22 10:23:11 +02:00
f4exb
7e6267f41c SSB demod: use delay line to squeeze squelch tail 2018-04-22 09:37:34 +02:00
f4exb
4670cbaa33 Merge branch 'master' into dev 2018-04-22 07:48:13 +02:00
f4exb
e206d65283
Merge pull request #162 from jigpu/spelling-fixes
Spelling fixes
2018-04-22 07:45:20 +02:00
f4exb
a099b05fb6
Merge pull request #164 from jigpu/memleak-fixes
Memleak fixes
2018-04-22 07:40:58 +02:00
Jason Gerecke
1eaae0de30 BFM demod: Delete the SpectrumVis when destroying BFMDemodGUI
Take care of a pair of memory leaks that occur when the BFM demod GUI
is closed.

==786== 19,952 (24 direct, 19,928 indirect) bytes in 1 blocks are definitely lost in loss record 7,065 of 7,081
==786==    at 0x4C2D54F: operator new(unsigned long) (vg_replace_malloc.c:334)
==786==    by 0x54C82ED: allocate (new_allocator.h:111)
==786==    by 0x54C82ED: allocate (alloc_traits.h:436)
==786==    by 0x54C82ED: _M_get_node (stl_list.h:383)
==786==    by 0x54C82ED: _M_create_node<FFTWEngine::Plan* const&> (stl_list.h:572)
==786==    by 0x54C82ED: _M_insert<FFTWEngine::Plan* const&> (stl_list.h:1801)
==786==    by 0x54C82ED: push_back (stl_list.h:1118)
==786==    by 0x54C82ED: FFTWEngine::configure(int, bool) (fftwengine.cpp:35)
==786==    by 0x4F4590B: SpectrumVis::handleConfigure(int, int, FFTWindow::Function) (spectrumvis.cpp:206)
==786==    by 0x4F459EE: SpectrumVis::handleMessage(Message const&) (spectrumvis.cpp:170)
==786==    by 0x548346C: BasebandSampleSink::handleInputMessages() (basebandsamplesink.cpp:21)
==786==    by 0x69E1615: QMetaObject::activate(QObject*, int, int, void**) (in /usr/lib/libQt5Core.so.5.10.1)
==786==    by 0x5497D4B: MessageQueue::push(Message*, bool) (messagequeue.cpp:52)
==786==    by 0x33F9D61A: BFMDemodGUI::BFMDemodGUI(PluginAPI*, DeviceUISet*, BasebandSampleSink*, QWidget*) (bfmdemodgui.cpp:352)
==786==    by 0x33F9D985: BFMDemodGUI::create(PluginAPI*, DeviceUISet*, BasebandSampleSink*) (bfmdemodgui.cpp:50)
==786==    by 0x33FB035D: non-virtual thunk to BFMPlugin::createRxChannelGUI(DeviceUISet*, BasebandSampleSink*) (bfmplugin.cpp:57)
==786==    by 0x4F47F19: DeviceUISet::loadRxChannelSettings(Preset const*, PluginAPI*) (deviceuiset.cpp:201)
==786==    by 0x4EA51EA: MainWindow::loadPresetSettings(Preset const*, int) (mainwindow.cpp:575)
==786==    by 0x4EAC81B: MainWindow::MainWindow(qtwebapp::LoggerWithFile*, MainParser const&, QWidget*) (mainwindow.cpp:176)
==786==    by 0x10A49B: runQtApplication(int, char**, qtwebapp::LoggerWithFile*) (main.cpp:120)
==786==    by 0x109B38: main (main.cpp:131)
==786==
==786== 54,096 (112 direct, 53,984 indirect) bytes in 1 blocks are definitely lost in loss record 7,075 of 7,081
==786==    at 0x4C2D54F: operator new(unsigned long) (vg_replace_malloc.c:334)
==786==    by 0x69E9162: QObject::QObject(QObject*) (in /usr/lib/libQt5Core.so.5.10.1)
==786==    by 0x5497BEE: MessageQueue::MessageQueue(QObject*) (messagequeue.cpp:26)
==786==    by 0x54834FF: BasebandSampleSink::BasebandSampleSink() (basebandsamplesink.cpp:6)
==786==    by 0x4F45A29: SpectrumVis::SpectrumVis(float, GLSpectrum*) (spectrumvis.cpp:26)
==786==    by 0x33F9D53C: BFMDemodGUI::BFMDemodGUI(PluginAPI*, DeviceUISet*, BasebandSampleSink*, QWidget*) (bfmdemodgui.cpp:342)
==786==    by 0x33F9D985: BFMDemodGUI::create(PluginAPI*, DeviceUISet*, BasebandSampleSink*) (bfmdemodgui.cpp:50)
==786==    by 0x33FB035D: non-virtual thunk to BFMPlugin::createRxChannelGUI(DeviceUISet*, BasebandSampleSink*) (bfmplugin.cpp:57)
==786==    by 0x4F47F19: DeviceUISet::loadRxChannelSettings(Preset const*, PluginAPI*) (deviceuiset.cpp:201)
==786==    by 0x4EA51EA: MainWindow::loadPresetSettings(Preset const*, int) (mainwindow.cpp:575)
==786==    by 0x4EAC81B: MainWindow::MainWindow(qtwebapp::LoggerWithFile*, MainParser const&, QWidget*) (mainwindow.cpp:176)
==786==    by 0x10A49B: runQtApplication(int, char**, qtwebapp::LoggerWithFile*) (main.cpp:120)
==786==    by 0x109B38: main (main.cpp:131)
2018-04-21 18:56:21 -07:00
Jason Gerecke
141997475c BFM demod: RDS demod: Initialize RDSDemod array elements
The m_parms.tot_errs array is not initialized prior to its first use
in the RDSDemod::biphase function. ASAN does not pick up on this
directly, but instead reports it as follows (note that ASAN fills
memory with 0xBE and -1094795586 is 0xBEBEBEBE):

    ./plugins/channelrx/demodbfm/rdsdemod.cpp:159:95: runtime error: signed
    integer overflow: -1094795586 + -1094795586 cannot be represented in type
    'int'

The m_parms.subcarr_bb array does not appear to be read prior to
initialization, but we initialize it to zero anyway for the sake
of good hygiene.
2018-04-21 18:54:19 -07:00
Jason Gerecke
bc4d7adce7 FileSourceGui: Prevent potential integer overflow in updateWithStreamTime
UBSan reports the following error when replaying an IQ stream:

    ./plugins/samplesource/filesource/filesourcegui.cpp:331:29: runtime
    error: signed integer overflow: 2704064 * 1000 cannot be represented
    in type 'int'

By rearranging the calculation, we can be sure that the calculation never
overflows.
2018-04-21 18:53:21 -07:00
Jason Gerecke
8cf3469c87 Fix spelling errors in documentation 2018-04-21 18:45:09 -07:00
Jason Gerecke
f6ea7b216e Fix spelling errors present in UI files 2018-04-21 18:44:54 -07:00
f4exb
b4a77e08c8 DSD demod: symmetrical attack and decay for squelch like NFM 2018-04-21 17:22:07 +02:00
f4exb
84538f1acf DSD demod: use buffered squelch to start decoding at the very beginning of the transmission regardless of squelch gate length 2018-04-21 09:56:12 +02:00
f4exb
12f5f4e30c NFM demod: buffered squelch 2018-04-21 09:23:01 +02:00
f4exb
20fbac5621 LimeSDR: updated documentation 2018-04-20 09:15:30 +02:00
f4exb
8c891a191a UDP source and sink: ensure proper click focus on line edits 2018-04-20 00:17:10 +02:00
f4exb
15f7c4d80e UDP source and sink: added dialogs to specify addresses and ports 2018-04-19 23:03:21 +02:00
f4exb
befc08f2e1 UDP source: send audio samples always on 16 bits. Options to send raw I/Q in either 16 or 24 bits regardless of sample size at compile time 2018-04-19 13:34:22 +02:00
f4exb
da362823dc UDPSink: use neutral denomination for I/Q sample size with sample size format 2018-04-19 00:52:01 +02:00
f4exb
f8251ecb50 UDPSink: fixed wrong sample sizes based on I/Q actual sample size that can now be 16 or 32 bits 2018-04-19 00:43:29 +02:00
f4exb
5c51297717 Moved FFT filter destruction at end of the mod or demod destructor 2018-04-18 22:20:47 +02:00
f4exb
aa3f12ec46 NFM demod: assymetrical fade in and fade out for the squelch 2018-04-17 09:15:02 +02:00
f4exb
31eb02ab74 LimeSDR: updated documentation with transverter shift 2018-04-17 01:31:12 +02:00
f4exb
ff88a2e1db LimeSDR output: implemented transverter shift 2018-04-17 00:55:56 +02:00
f4exb
b9587273d8 LimeSDR input: implemented transverter shift 2018-04-17 00:25:19 +02:00