summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeLines
* Fix "REFER declined" deadlockFrédéric Brière2019-12-27-1/+1
| | | | | | | | | This occurs when the user declines an incoming call transfer request, due to the following path: - recvd_refer_permission() acquires a read lock over lines_mtx - move_releasing_lines_to_background() is called - A write lock over lines_mtx is acquired
* Fix "Transfer with consultation without Replaces" deadlockFrédéric Brière2019-12-27-1/+1
| | | | | | | | | | | This occurs when performing a consult transfer if the target does not support the Replaces extension, due to the following path: - refer() acquires a read lock over lines_mtx - refer_consultation() is called - A write lock over lines_mtx is acquired Closes #118
* Fix "NOTIFY received when line is idle" deadlockFrédéric Brière2019-12-27-1/+1
| | | | | | | | | | This can easily be triggered by starting a blind transfer, and hanging up before the other end has accepted/rejected it, due to the following path: - recvd_notify() acquires a read lock on lines_mtx - cleanup_dead_lines() is called - A write lock on lines_mtx is acquired
* Fix "INVITE with Replaces header" deadlockFrédéric Brière2019-12-27-1/+1
| | | | | | | | | This occurs on reception of an INVITE with a Replaces header, due to the following path: - recvd_invite() acquires a read lock on lines_mtx - recvd_initial_invite() is called - A write lock over lines_mtx is acquired
* Introduce read-write-update locks and guards to prevent deadlocksFrédéric Brière2019-12-27-13/+199
| | | | | | | | | | | | | | | | | | | | This converts t_rwmutex and t_rwmutex_guard into a read-write-update[*] lock and guard, in an attempt to circumvent the various deadlocks that were introduced with the addition of lines_mtx in 38bb6b7. [*] For more details, see https://stackoverflow.com/a/18785300 and http://lkml.iu.edu/hypermail/linux/kernel/0004.3/0117.html. Note that this is not a real fix; this would require analyzing and refactoring phone.cpp, which is well beyond my abilities. This is at best a workaround that appears to conveniently dodge all the deadlocks I've encountered so far. (It would have been more proper to introduce a separate class for this purpose, but this would have required modifying over 80 lines just to change one type for another. As phone_users_mtx is the only other instance of this class, the impact of subverting t_rwmutex directly is minimal.)
* Create t_rwmutex_guard base classFrédéric Brière2019-12-26-7/+19
| | | | (Doing this ahead of time to simplify the next commit a bit.)
* Move t_rwmutex_reader/writer definitions into mutex.cppFrédéric Brière2019-10-02-16/+32
| | | | | These classes are about to get more complex, so let's move them ahead of time into mutex.cpp.
* Make `t_phone::mutex_3way` recursive to avoid a deadlockFrédéric Brière2019-10-02-1/+1
| | | | | | | | | | | | | `t_phone::mutex_3way` can be locked twice when hanging up a conference call: - `t_phone::cleanup_3way_state()` acquires a lock - `t_audio_session::stop_3way()` is called - `t_audio_session::get_peer_3way()` is called - `t_phone::get_3way_peer_line()` is called - which acquires another lock Making that mutex recursive is a simple way to work around this issue.
* Remove duplicate mutex lock in `t_phone::end_call()`Frédéric Brière2019-10-02-2/+0
| | | | | A write lock on `lines_mtx` has already been acquired at the beginning of `end_call()`.
* Change the address of the manual until we have our own up and runningLubos Dolezel2019-06-17-1/+1
|
* Merge pull request #140 from f0k/patch-1Luboš Doležel2019-04-08-1/+1
|\ | | | | Case insensivity in WWW-Authenticate header
| * Case insensivity in WWW-Authenticate headerJan Schlüter2019-04-08-1/+1
|/
* Twinkle 1.10.2Lubos Dolezel2019-02-14-2/+8
|
* Merge pull request #136 from palinek/new_alsa_fixLuboš Doležel2018-11-29-5/+3
|\ | | | | audio: Fix parameter setting failure with alsa v1.1.7
| * audio: Fix parameter setting failure with alsa v1.1.7Palo Kisa2018-11-29-5/+3
|/
* Merge pull request #134 from lpmi-13/typofixLuboš Doležel2018-11-09-1/+1
|\ | | | | fix simple typo
| * fix simple typolpmi-132018-11-08-1/+1
|/
* Merge pull request #132 from glixx/patch-1Luboš Doležel2018-10-11-87/+324
|\ | | | | Update Russian translation
| * Update Russian translationglixx2018-10-11-87/+324
|/
* Merge pull request #125 from mkubecek/tmp/qt-5.11Luboš Doležel2018-06-06-1/+8
|\ | | | | fix Qt 5.11 build (issue #124)
| * Include <QRegExpValidator> explicitlyMichal Kubecek2018-06-06-1/+8
|/ | | | | | | Since Qt 5.11, generated ui_getprofilename.h no longer includes QHeaderView which breaks the chain that included qvalidator.h in getprofilename.cpp. As it feels rather fragile to rely on such indirect includes, let's include <QRegExpValidator> explicitly in each file using QRegExpValidator class.
* Merge pull request #119 from petercolberg/masterLuboš Doležel2018-03-28-1/+1
|\ | | | | Fix spelling mistake
| * Fix spelling mistakePeter Colberg2018-03-27-1/+1
|/
* Merge pull request #113 from fbriere/misc/qt-versionLuboš Doležel2018-02-11-36/+0
|\ | | | | Remove some remaining Qt 4 code
| * Remove remaining bits of Qt 4 codeFrédéric Brière2018-02-10-32/+0
| |
| * Remove some code specific to Qt 4Frédéric Brière2018-02-10-4/+0
|/
* Merge pull request #110 from fbriere/feature/travis-matrixLuboš Doležel2018-02-10-3/+20
|\ | | | | Test various compiler versions and build options on Travis CI
| * Test various compiler versions and build options on Travis CIFrédéric Brière2018-02-10-3/+20
|/
* Merge pull request #109 from fbriere/misc/travis-build-depsLuboš Doležel2018-02-10-1/+2
|\ | | | | Adjustments to the list of packages installed on Travis CI
| * Add comment explaining the need for qttools5-dev-tools on Travis CIFrédéric Brière2018-02-09-0/+1
| |
| * Remove qtscript5-dev from the list of packages installed on Travis CIFrédéric Brière2018-02-09-1/+1
| | | | | | | | | | Twinkle has never actually used QtScript, as far as I can tell. Rather, it uses embedded JavaScript in QML, which is all part of qtdeclarative.
| * Remove qtquick1-5-dev from the list of packages installed on Travis CIFrédéric Brière2018-02-09-1/+1
|/ | | | Twinkle no longer uses QtQuick 1 (since 497c609).
* Merge pull request #108 from fbriere/misc/build-depsLuboš Doležel2018-02-09-3/+14
|\ | | | | Various additions to the list of build dependencies
| * Mention that SpeexDSP is required with the main Speex libraryFrédéric Brière2018-02-09-1/+1
| |
| * Add note about ALSA often being packaged as libasoundFrédéric Brière2018-02-09-1/+1
| |
| * Add cmake, bison and flex to the list of build dependenciesFrédéric Brière2018-02-09-0/+6
| |
| * List which Qt submodules are actually requiredFrédéric Brière2018-02-09-1/+4
| | | | | | | | | | | | There are only three Qt submodules which we actually need and check for in CMakeLists, namely: base (Qt5Widgets), declarative (Qt5Quick) and tools (Qt5LinguistTools).
| * Add libmagic and libreadline to the list of build dependenciesFrédéric Brière2018-02-09-0/+2
|/
* Merge pull request #107 from fbriere/bug/bind-is-ambiguousLuboš Doležel2018-02-09-4/+4
|\ | | | | Disambiguate reference to ::bind()
| * Disambiguate reference to ::bind()Frédéric Brière2018-02-09-4/+4
|/ | | | | | There's a potential ambiguity between ::bind() and std::bind() if <functional> happens to be included beforehand (as is the case with libc++).
* Merge pull request #105 from fbriere/bug/addressbook-emitLuboš Doležel2018-02-09-1/+1
|\ | | | | bugfix: Signal name was omitted in selectLocalAddress() emit
| * bugfix: Signal name was omitted in selectLocalAddress() emitFrédéric Brière2018-02-09-1/+1
| |
* | Merge pull request #106 from fbriere/bug/cmake-libmagic-fileLuboš Doležel2018-02-09-6/+4
|\ \ | |/ |/| Don't require file(1) when checking for libmagic
| * Don't require file(1) when checking for libmagicFrédéric Brière2018-02-09-6/+4
|/ | | | | | | | | | | | | | | | | | Attempting to compile Twinkle on a system where file(1) is not available currently produces a confusing error message about libmagic not being found. The only reason to require file(1) was to obtain the libmagic version, as the MAGIC_VERSION constant was apparently only introduced in 5.13. But since we don't need any particular version, we might as well drop this requirement. Using find_program() instead of find_path() avoids picking /usr/include/file by mistake, which resulted in a (harmless) empty version string in the CMake output. (Thanks to https://bro-tracker.atlassian.net/browse/BIT-1096 for providing some of this information.)
* Merge pull request #94 from gwozniak/masterLuboš Doležel2017-11-26-0/+53
|\ | | | | PAI Header option
| * PAI Header optionGrzegorz Woźniak2017-04-13-0/+53
| |
* | Merge pull request #102 from traud/UDUBLuboš Doležel2017-11-26-2/+2
|\ \ | | | | | | User-Determined User Busy (UDUB)
| * | User-Determined User Busy (UDUB)Alexander Traud2017-11-26-2/+2
|/ / | | | | | | | | | | | | | | | | | | | | When you reject a call, either A. the forwarding rule for User-Busy kicks in, like an answering machine or another extension (of a personal secretary or a deputy). Or B. the caller hears normal ringing first, and then the busy tone. To allow case A, a SIP user-agent client (UAC) has to return the status User-Busy (486). The same should happen in case B, because the caller shall hear a busy tone on rejection. 486 is mandated for SIP clients in mobile phones by the GSMA. Furthermore, other SIP-client creators give a 486 in that case as well; I tested Acrobits, Atlinks, Counterpath, Gigaset, Grandstream, RTX, Snom, Yealink, and Vtech. Before this change, twinkle rejected a call with the SIP-Status 603. The class 6xx requires that *all* other registered phones stop to ring. Furthermore, some SIP user-agent servers (UAS) follow RFC 3398 and map a status 603 to the cause-code 21, which is mapped back to status 403 (Forbidden). Cisco and Digium Asterisk do this. For example in Asterisk, after you rejected the call in twinkle, the caller did not get 603 or 486, but 403. However, in case of 403, the original caller is allowed to re-try the call setup. For example, I have a Nokia Symbian/S60 based phone which tries via IPv4 first, then after it received the 403, that phone tries again via IPv6. Consequently, a user of twinkle had to reject the call twice when 603 was returned.
* | Fix crash in parsing pending message countLubos Dolezel2017-11-15-2/+6
| |
* | Fix DTMF key 3 being sent twice (resolves #95)Lubos Dolezel2017-04-28-1/+0
|/
generated by cgit v1.2.3 (git 2.25.1) at 2024-11-21 23:38:14 +0000