Yue Wang
79937c9495
Support timed_wait in microseconds level
...
Some Audio plugin (such as ALSA, and soon CoreAudio) already support microsecond level buffer time. However, current interval less than 1000 microseconds will cause a bug that the code treated as 0 ms.
2018-07-15 13:22:53 -07:00
Max Kellermann
65bbb0e0aa
release v0.20.17
...
-----BEGIN PGP SIGNATURE-----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=v4t8
-----END PGP SIGNATURE-----
Merge tag 'v0.20.17'
release v0.20.17
2018-02-11 13:22:43 +01:00
Max Kellermann
817656504d
thread/Util: implement system call wrapper for sched_setscheduler()
...
There is a POSIX definition for sched_setscheduler(), but Linux does
not implement that; instead of changing the process's scheduler, it
only affects one thread. This has caused some confusion among
application developers and C library developers.
While glibc implements Linux semantics, Musl has made their
sched_setscheduler() function an always-failing no-op, causing the
error message "sched_setscheduler failed: Function not implemented".
http://git.musl-libc.org/cgit/musl/commit/src/sched/sched_setscheduler.c?id=1e21e78bf7a5c24c217446d8760be7b7188711c2
Instead of relying on the C library which may be unreliable here, we
now roll our own system call wrapper.
Closes #218
2018-02-09 18:43:45 +01:00
Max Kellermann
6f00f97b66
thread/Util: rename ioprio_set() to linux_ioprio_set()
...
Juse in cas glibc gets a wrapper for the system call which would then
conflict with ours.
2018-02-09 18:43:45 +01:00
Max Kellermann
97f670658f
util/StringFormat: new utility library
2018-01-24 13:28:28 +01:00
Max Kellermann
96676f8f3b
Merge branch 'v0.20.x'
2018-01-15 09:06:31 +01:00
Max Kellermann
2eef4e6716
thread/Thread: add debug attribute "inside_handle"
...
This attribute shall be used only for IsInside() to make this safe
against a race condition described in #188 :
> There is no requirement on the implementation that the ID of the
> created thread be available before the newly created thread starts
> executing.
http://pubs.opengroup.org/onlinepubs/009695399/functions/pthread_create.html ):
This means that on some pthread implementations (e.g. Haiku), the
assert(thread.IsInside()) could fail.
Closes #188
2018-01-08 09:58:18 +01:00
Max Kellermann
d989dbfec4
thread/Thread: make IsInside() debug-only
...
This method is only used inside assert().
2018-01-08 09:56:39 +01:00
Max Kellermann
ca9fcec364
thread/Thread: fix indent
2018-01-08 09:49:08 +01:00
Max Kellermann
354104f9a9
thread/{Thread,Id}: use defaul-initialized pthread_t as "undefined" value
...
Use the "==" operator instead of pthread_equal().
This allows us to eliminate two boolean flags which are there to avoid
race conditions, and made the thing so fragile that I got tons of
(correct) thread sanitizer warnings.
2018-01-07 17:20:26 +01:00
Max Kellermann
8649ea3d6f
thread/Thread: use BoundMethod
2018-01-07 17:20:26 +01:00
Max Kellermann
752ff12c37
thread/Thread: move code to Run()
2018-01-07 17:20:26 +01:00
Max Kellermann
4a120f8090
Merge branch 'v0.20.x'
2018-01-05 10:10:17 +01:00
Max Kellermann
329f9cd9fe
thread/Util: no ioprio_set() on Android due to seccomp/SIGSYS
2018-01-04 19:33:14 +01:00
Max Kellermann
dbfc4abfd6
thread/SafeSingleton: disallow copying
2018-01-02 12:39:05 +01:00
Max Kellermann
5544375002
thread/SafeSingleton: pass parameters to constructor
2018-01-02 12:26:53 +01:00
Max Kellermann
196df1ccd5
thread/SafeSingleton: new thread-safe utility class
2018-01-02 12:10:41 +01:00
Max Kellermann
6a8c2848f6
thread/{Thread,Id}: use defaul-initialized pthread_t as "undefined" value
...
Use the "==" operator instead of pthread_equal().
This allows us to eliminate two boolean flags which are there to avoid
race conditions, and made the thing so fragile that I got tons of
(correct) thread sanitizer warnings.
2017-12-22 10:37:07 +01:00
Max Kellermann
b6af7abb1a
thread/PosixMutex: add "noexcept"
2017-12-18 23:29:08 +01:00
Max Kellermann
6246d36fe6
Merge branch 'v0.20.x'
2017-12-16 20:56:06 +01:00
Max Kellermann
dfaf08743c
*: check defined(_WIN32) instead of defined(WIN32)
...
Only _WIN32 is defined by the compiler, and WIN32 is not standardized
and may be missing.
Closes #169
2017-12-12 10:22:20 +01:00
Max Kellermann
618704f504
thread/*: add "noexcept"
2017-11-26 11:58:53 +01:00
Max Kellermann
979f1b6c39
release v0.20.9
...
-----BEGIN PGP SIGNATURE-----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=0SSp
-----END PGP SIGNATURE-----
Merge tag 'v0.20.9'
release v0.20.9
2017-06-04 12:57:05 +02:00
Max Kellermann
a057b4f6d8
*: add lost of "noexcept" specifications
2017-06-04 12:46:48 +02:00
Max Kellermann
4faef28cc5
release v0.20.7
...
-----BEGIN PGP SIGNATURE-----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=CdDy
-----END PGP SIGNATURE-----
Merge tag 'v0.20.7'
release v0.20.7
2017-05-15 23:01:49 +02:00
Max Kellermann
71f0ed8b74
*: add "noexcept" to many, many function prototypes
...
This eliminates some overhead, because the compiler doesn't need to
consider these functions throwing.
2017-05-08 14:44:49 +02:00
Max Kellermann
781487c4dd
thread/Thread: use BoundMethod
2017-02-10 22:46:09 +01:00
Max Kellermann
5ba5bc8ba1
thread/Thread: move code to Run()
2017-02-10 22:43:55 +01:00
Max Kellermann
82c66ce078
thread/Thread: Start() returns void
...
Since we switched to C++ exceptions, there is no code path which
returns false.
2017-02-10 22:41:29 +01:00
Max Kellermann
6eea56861b
AUTHORS, ...: update my email address
2017-01-17 11:54:55 +01:00
Max Kellermann
5900253b85
update copyright year
2017-01-03 20:48:59 +01:00
Max Kellermann
2e182e84c3
thread/Mutex: remove ScopeLock, use std::lock_guard directly
2017-01-03 07:11:57 +01:00
Max Kellermann
a42021655c
thread/Cond: make the integer-based timed_wait() overload private
...
Make sure this deprecated API is not used.
2017-01-03 07:07:10 +01:00
Max Kellermann
2498a2b0bd
thread/Mutex: use std::lock_guard instead of std::unique_lock
...
We need purely RAII-style, and this is std::lock_guard.
2017-01-03 06:56:24 +01:00
Max Kellermann
6fc47fbb69
thread/Mutex: make ScopeLock an alias for std::unique_lock
2016-12-29 11:53:24 +01:00
Max Kellermann
9c11184238
thread/Mutex: use std::unique_lock to implement ScopeLock
2016-12-29 11:51:25 +01:00
Max Kellermann
871ba5a488
thread/Cond: add timed_wait() overload with std::chrono support
2016-12-28 21:47:04 +01:00
Christian Körner
8bbfb5cda1
thread/PosixCond: fix timed_wait
...
pthread_cond_timedwait() in PosixCond.hxx:timed_wait(PosixMutex...) returns
EINVAL, if ts.tv_nsec >= 1E9. In this case, it returns to early.
Find attached a patch which fixes this. I chose a compare-subtraction method
to keep ts.tv_nsec below 1E9.
Another option would be
ts.tv_sec += ts.tv_nsec / 1000000000;
ts.tv_nsec %= 1000000000;
But I guess this takes more time on some ARM processors, which don't support
hardware division.
2016-09-19 11:21:21 +02:00
Max Kellermann
12091fcfb1
thread/Util: throw exception on error
2016-09-04 15:15:11 +02:00
Max Kellermann
5598826eaf
thread/Util: un-inline the functions
...
Reduce header clutter.
2016-09-04 15:11:56 +02:00
Max Kellermann
d3c7fac606
thread/Thread: throw std::system_error on error
2016-06-17 19:11:20 +02:00
Max Kellermann
fea3f6cc72
thread/Thread: use C++11 initialisers
2016-06-17 19:11:19 +02:00
Max Kellermann
eca6b9f0af
thread/Mutex: add method ScopeLock::Unlock()
2016-05-02 23:32:44 +02:00
Max Kellermann
1d67aa7bf2
update copyright year to 2016
2016-02-26 17:54:05 +01:00
Christian Halaszovich
3392cbbd91
Emit a warning if the OutputThread fails to get realtime scheduling
...
This only applies to linux systems. Here, sched_setscheduler() is
called to get realtime scheduling. With this patch, the return value
of this function is now checked and a warning / error message is
generated if it fails.
2016-02-19 17:09:44 +01:00
Max Kellermann
ac1eaff6ec
thread/Mutex: add class ScopeUnlock()
2015-12-31 12:30:53 +01:00
Max Kellermann
48d7fedbc0
thread/{Mutex,Cond}: remove obsolete comments
2015-08-25 12:52:13 +02:00
Max Kellermann
75dff64450
thread/Posix{Mutex,Cond}: use "constexpr" only with glibc
...
Apparently all other C libraries are not compatible with "constexpr".
Those which are not will get a performance penalty, but at least they
work at all.
2015-08-25 12:46:12 +02:00
Max Kellermann
ae7e1a22cb
Merge branch 'v0.19.x'
2015-01-26 20:57:29 +01:00
Max Kellermann
276a0d9500
thread/Name: include stdio.h if HAVE_PRCTL
...
Caused a build failure with uClibc because snprintf() was not
available.
2015-01-21 20:40:56 +01:00
