2
0
mirror of https://github.com/boostorg/thread.git synced 2026-01-26 19:12:11 +00:00
Commit Graph

2437 Commits

Author SHA1 Message Date
Austin Beer
82e2c82e95 Renaming timespec to platform 2017-10-02 09:23:54 -06:00
Vicente J. Botet Escriba
3479524845 trace elapsed time. 2017-10-02 13:18:33 +02:00
Vicente J. Botet Escriba
178e0661ab fix bad clock:: use, shoudl be chrono::steady_clock. 2017-10-02 08:59:41 +02:00
Vicente J. Botet Escriba
b94c36f960 move to 650 ms to see if the test pass. 2017-10-02 08:41:39 +02:00
Vicente J. Botet Escriba
9f7e37c892 Merge branch 'develop' into feature/timespec_clocks 2017-10-01 13:28:57 +02:00
Vicente J. Botet Escriba
e7d6055ed3 Merge pull request #167 from austin-beer/feature/timespec_clocks_pr12
Fixes and Cleanup
2017-10-01 10:15:31 +02:00
Vicente J. Botet Escriba
75d5ebbea5 uncomment dependency from test jamfile to threadapi to see if we get a gree CI on windows. 2017-09-30 13:44:28 +02:00
Vicente J. Botet Escriba
9969ec8504 remove the #warning 2017-09-30 12:18:30 +02:00
Vicente J. Botet Escriba
995033cf20 Make sync_timed_queue more general adding a TimePoint parameter and fix the inteface of all the time related functions _for/_until to take generic chrono::timepoints and chrono::durations. 2017-09-30 12:15:06 +02:00
Vicente J. Botet Escriba
d5da7a4dda disable ex_lambda_future test for msvc-11.0 and add a warning to signal this doesn't work. 2017-09-30 10:03:17 +02:00
Austin Beer
659d309206 Re-enabled tests on Windows that had been temporarily disabled. 2017-09-29 15:53:21 -06:00
Austin Beer
847dcbf237 Removed unnecessary explicit qualifiers. 2017-09-29 15:28:04 -06:00
Austin Beer
35d25851ce * Reverted some of my Windows updates to get a condition variable test on Windows to pass again. Further analysis is needed to determine why the updates didn't work.
* Increased the range of one test so it passes on Windows.
2017-09-29 14:52:42 -06:00
Austin Beer
5ffca91e60 * Hopefully fixed the build failure on Mac OS.
* Refactored to use timespec as the internal storage with pthread and intmax_t with Windows.
2017-09-29 10:22:01 -06:00
Austin Beer
51e6e654e3 Re-disabled the three move_ctor_upgrade_lock tests on Windows that are still failing. 2017-09-29 02:17:58 -06:00
Vicente J. Botet Escriba
71bc9c855d comment dependencies to threadapi feature on test/Jamfile to show that there is no impact. 2017-09-29 08:43:30 +02:00
Vicente J. Botet Escriba
f3492f8da0 Add runs for msvc-9.0, msvc-10.0 and msvc-11.0. 2017-09-29 08:41:25 +02:00
Austin Beer
3c25ad09ce Attempting to fix build errors on Windows. 2017-09-29 00:34:47 -06:00
Austin Beer
a2e037bb54 Fixes and Cleanup
- Fixed build failures on Windows. The timespec struct is not supported by older versions of Visual Studio. I changed the internal representation inside of the *_timespec_timepoint classes to a boost::intmax_t representing the number of nanoseconds since the epoch.
- Fixed some functions that wouldn't execute at all if they were provided a negative time duration or an absolute time that was in the past. From what I understand, they should instead execute once and then return immediately.
- Moved pthread/timespec.hpp to detail/timespec.hpp.
- Deleted detail/internal_clock.hpp and moved the seven relevant lines into detail/timespec.hpp. This keeps all of the internal clock declarations in one place.
- Renamed thread_detail::internal_clock_t to detail::internal_chrono_clock to be consistent with and yet clearly differentiated from detail::internal_timespec_clock.
- Removed "using namespace chrono" to eliminate ambiguious namespace resolution when referencing detail::internal_chrono_clock.
- Re-enabled a few tests on Windows that had previously been disabled. I want to see whether or not they still need to be disabled.
2017-09-28 22:27:37 -06:00
Vicente J. Botet Escriba
a4a21b1c12 Merge pull request #166 from austin-beer/feature/timespec_clocks_pr11
Switching the Windows condition_variable and interruptible_wait functions over to using internal_timespec_clock.
2017-09-29 01:56:58 +02:00
Vicente J. Botet Escriba
e9ce83b399 Merge pull request #160 from karzhenkov/fix-threadapi-detection
Auto-detection of <threadapi> based on <target-os>
2017-09-28 23:50:38 +02:00
Austin Beer
c2b207faa4 Switching the Windows condition_variable and interruptible_wait functions over to using internal_timespec_clock. 2017-09-28 13:24:21 -06:00
Alexander Karzhenkov
c3a007e88b <threadapi> feature definition moved to module "threadapi.jam" 2017-09-28 20:18:43 +05:00
Vicente J. Botet Escriba
68abb96818 Merge pull request #165 from austin-beer/feature/timespec_clocks_pr10
Added support for getting time on Windows to the timespec classes.
2017-09-28 08:17:23 +02:00
Austin Beer
29c3b48da0 Switching over to using the existing GetTickCount64() code for reporting monotonic time on Windows. This code is less precise than QueryPerformanceCounter(), but it's more reliable on older versions of Windows (e.g. XP and Server 2003). 2017-09-27 15:40:50 -06:00
Alexander Karzhenkov
9c1c372067 <threadapi> is now optional; detection rules moved to "threadapi.jam" 2017-09-28 01:25:17 +05:00
Austin Beer
13713f3b9c Added support for getting time on Windows to the timespec classes. 2017-09-27 10:52:53 -06:00
Vicente J. Botet Escriba
776d42f6cb Merge branch 'develop' into feature/timespec_clocks 2017-09-26 08:01:28 +02:00
Vicente J. Botet Escriba
b20ea9ebb1 fix strike characted. 2017-09-26 08:00:44 +02:00
Vicente J. Botet Escriba
46c1ae6f96 Merge branch 'develop' into feature/timespec_clocks 2017-09-26 05:55:40 +02:00
Vicente J. Botet Escriba
f02f23eff5 Add STACK_SIZE_PARAM_IS_A_RESERVATION when setting the stack size. 2017-09-26 05:54:53 +02:00
Vicente J. Botet Escriba
fb0317ec7d Merge pull request #164 from austin-beer/feature/timespec_clocks_pr8
Fixed compile issue on Windows
2017-09-25 19:25:38 +02:00
Austin Beer
417a2d8a46 Fixed compile issue on Windows. 2017-09-25 11:15:09 -06:00
Vicente J. Botet Escriba
ce7f2b09cc try to use again pthread/shared_mutex.hpp on windows. Comment not used functions in timespec. 2017-09-25 12:52:25 +02:00
Vicente J. Botet Escriba
e6226a95ba try to fix future wait on posix duration. 2017-09-25 08:39:45 +02:00
Vicente J. Botet Escriba
aa221d02b5 try to fix issues on linu-mono mutex operation on system_time. 2017-09-25 00:18:12 +02:00
Vicente J. Botet Escriba
5a7d8eca6f Merge pull request #163 from DanielaE/feature/qoi-warning
fix or silence (msvc) compiler warnings about constant conditional ex…
2017-09-24 21:04:34 +02:00
Vicente J. Botet Escriba
de4ee72554 disable some failing unique_lock/upgrade_lock tests on windows. 2017-09-24 20:21:43 +02:00
Vicente J. Botet Escriba
ad5ea75ddd disable tsome failing unique_lock/upgrade_lock tests on windows. 2017-09-24 19:38:34 +02:00
Vicente J. Botet Escriba
dd1b5282b2 try to fix issues with windows and shared_mutex. 2017-09-24 17:29:27 +02:00
Vicente J. Botet Escriba
45321f0570 try to fix issues with new pthread/shared_mutex.hpp on windows by disabling the use of the pthread version that now depends on timespec. 2017-09-24 16:31:42 +02:00
Daniela Engert
4e83359537 fix or silence (msvc) compiler warnings about constant conditional expressions, unused parameters, narrowing, unreachable code, use of 'this' in base member initializations, and some other minor stuff
Signed-off-by: Daniela Engert <dani@ngrt.de>
2017-09-24 14:02:17 +02:00
Vicente J. Botet Escriba
bafc9f43a5 try to fix DateTime time related issues on pthread/shared_mutex.hpp. 2017-09-24 08:12:02 +02:00
Alexander Karzhenkov
ab3da91e67 Auto-detection logic for <threadapi> added to "test/Jamfile.v2" 2017-09-23 21:52:48 +05:00
Alexander Karzhenkov
6ee1c88869 Try fix regression for native_handle 2017-09-23 15:24:53 +05:00
Vicente J. Botet Escriba
b98d022d40 try to fix DatTime related functions for mutex/recursive_mutex. 2017-09-23 08:42:02 +02:00
Vicente J. Botet Escriba
f2d6d99d54 Merge pull request #161 from austin-beer/feature/timespec_clocks_pr7
Finished fixing the condition_variable[_any] timed_wait() functions.
2017-09-23 07:00:11 +02:00
Austin Beer
aeb54ddbe5 Finished fixing the condition_variable[_any] timed_wait() functions. 2017-09-22 14:22:06 -06:00
Alexander Karzhenkov
2fb41edf1a Default <threadapi> is based on <target-os> instead of host system name
* <threadapi> has an additional value "native" which is the default.
  If build request doesn't specify other value, it is replaced
  either with "pthread" or "win32" depending on <target-os>.
* <tag> modifies name of generated library only if resulting <threadapi>
  value differs from one that would have been chosen to replace "native"
  according to <target-os>.
2017-09-22 10:51:07 +05:00
Vicente J. Botet Escriba
4d385bedd3 try to fix condition_variable::timed_wait for pthread. 2017-09-22 04:18:55 +02:00