RepoMirrors
/
mpv
mirror of https://github.com/mpv-player/mpv
1
0
Fork 0
Code Issues Releases Wiki Activity
mpv/osdep/atomic.h

94 lines
3.5 KiB
C
Raw Normal View History

core: add a spsc ringbuffer implementation Currently every single AO was implementing it's own ringbuffer, many times with slightly different semantics. This is an attempt to fix the problem. I stole some good ideas from ao_portaudio's ringbuffer and went from there. The main difference is this one stores wpos and rpos which are absolute positions in an "infinite" buffer. To find the actual position for writing / reading just apply modulo size. The producer only modifies wpos while the consumer only modifies rpos. This makes it pretty easy to reason about and make the operations thread safe by using barriers (thread safety is guaranteed only in the Single-Producer/Single- Consumer case). Also adapted ao_coreaudio to use this ringbuffer.
2013-05-23 19:23:32 +00:00
/*
* This file is part of mpv.
* Copyright (c) 2013 Stefano Pigozzi <stefano.pigozzi@gmail.com>
*
Relicense some non-MPlayer source files to LGPL 2.1 or later This covers source files which were added in mplayer2 and mpv times only, and where all code is covered by LGPL relicensing agreements. There are probably more files to which this applies, but I'm being conservative here. A file named ao_sdl.c exists in MPlayer too, but the mpv one is a complete rewrite, and was added some time after the original ao_sdl.c was removed. The same applies to vo_sdl.c, for which the SDL2 API is radically different in addition (MPlayer supports SDL 1.2 only). common.c contains only code written by me. But common.h is a strange case: although it originally was named mp_common.h and exists in MPlayer too, by now it contains only definitions written by uau and me. The exceptions are the CONTROL_ defines - thus not changing the license of common.h yet. codec_tags.c contained once large tables generated from MPlayer's codecs.conf, but all of these tables were removed. From demux_playlist.c I'm removing a code fragment from someone who was not asked; this probably could be done later (see commit 15dccc37). misc.c is a bit complicated to reason about (it was split off mplayer.c and thus contains random functions out of this file), but actually all functions have been added post-MPlayer. Except get_relative_time(), which was written by uau, but looks similar to 3 different versions of something similar in each of the Unix/win32/OSX timer source files. I'm not sure what that means in regards to copyright, so I've just moved it into another still-GPL source file for now. screenshot.c once had some minor parts of MPlayer's vf_screenshot.c, but they're all gone.
2016-01-19 17:36:06 +00:00
* mpv is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
core: add a spsc ringbuffer implementation Currently every single AO was implementing it's own ringbuffer, many times with slightly different semantics. This is an attempt to fix the problem. I stole some good ideas from ao_portaudio's ringbuffer and went from there. The main difference is this one stores wpos and rpos which are absolute positions in an "infinite" buffer. To find the actual position for writing / reading just apply modulo size. The producer only modifies wpos while the consumer only modifies rpos. This makes it pretty easy to reason about and make the operations thread safe by using barriers (thread safety is guaranteed only in the Single-Producer/Single- Consumer case). Also adapted ao_coreaudio to use this ringbuffer.
2013-05-23 19:23:32 +00:00
*
* mpv is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
Relicense some non-MPlayer source files to LGPL 2.1 or later This covers source files which were added in mplayer2 and mpv times only, and where all code is covered by LGPL relicensing agreements. There are probably more files to which this applies, but I'm being conservative here. A file named ao_sdl.c exists in MPlayer too, but the mpv one is a complete rewrite, and was added some time after the original ao_sdl.c was removed. The same applies to vo_sdl.c, for which the SDL2 API is radically different in addition (MPlayer supports SDL 1.2 only). common.c contains only code written by me. But common.h is a strange case: although it originally was named mp_common.h and exists in MPlayer too, by now it contains only definitions written by uau and me. The exceptions are the CONTROL_ defines - thus not changing the license of common.h yet. codec_tags.c contained once large tables generated from MPlayer's codecs.conf, but all of these tables were removed. From demux_playlist.c I'm removing a code fragment from someone who was not asked; this probably could be done later (see commit 15dccc37). misc.c is a bit complicated to reason about (it was split off mplayer.c and thus contains random functions out of this file), but actually all functions have been added post-MPlayer. Except get_relative_time(), which was written by uau, but looks similar to 3 different versions of something similar in each of the Unix/win32/OSX timer source files. I'm not sure what that means in regards to copyright, so I've just moved it into another still-GPL source file for now. screenshot.c once had some minor parts of MPlayer's vf_screenshot.c, but they're all gone.
2016-01-19 17:36:06 +00:00
* GNU Lesser General Public License for more details.
core: add a spsc ringbuffer implementation Currently every single AO was implementing it's own ringbuffer, many times with slightly different semantics. This is an attempt to fix the problem. I stole some good ideas from ao_portaudio's ringbuffer and went from there. The main difference is this one stores wpos and rpos which are absolute positions in an "infinite" buffer. To find the actual position for writing / reading just apply modulo size. The producer only modifies wpos while the consumer only modifies rpos. This makes it pretty easy to reason about and make the operations thread safe by using barriers (thread safety is guaranteed only in the Single-Producer/Single- Consumer case). Also adapted ao_coreaudio to use this ringbuffer.
2013-05-23 19:23:32 +00:00
*
Relicense some non-MPlayer source files to LGPL 2.1 or later This covers source files which were added in mplayer2 and mpv times only, and where all code is covered by LGPL relicensing agreements. There are probably more files to which this applies, but I'm being conservative here. A file named ao_sdl.c exists in MPlayer too, but the mpv one is a complete rewrite, and was added some time after the original ao_sdl.c was removed. The same applies to vo_sdl.c, for which the SDL2 API is radically different in addition (MPlayer supports SDL 1.2 only). common.c contains only code written by me. But common.h is a strange case: although it originally was named mp_common.h and exists in MPlayer too, by now it contains only definitions written by uau and me. The exceptions are the CONTROL_ defines - thus not changing the license of common.h yet. codec_tags.c contained once large tables generated from MPlayer's codecs.conf, but all of these tables were removed. From demux_playlist.c I'm removing a code fragment from someone who was not asked; this probably could be done later (see commit 15dccc37). misc.c is a bit complicated to reason about (it was split off mplayer.c and thus contains random functions out of this file), but actually all functions have been added post-MPlayer. Except get_relative_time(), which was written by uau, but looks similar to 3 different versions of something similar in each of the Unix/win32/OSX timer source files. I'm not sure what that means in regards to copyright, so I've just moved it into another still-GPL source file for now. screenshot.c once had some minor parts of MPlayer's vf_screenshot.c, but they're all gone.
2016-01-19 17:36:06 +00:00
* You should have received a copy of the GNU Lesser General Public
* License along with mpv. If not, see <http://www.gnu.org/licenses/>.
core: add a spsc ringbuffer implementation Currently every single AO was implementing it's own ringbuffer, many times with slightly different semantics. This is an attempt to fix the problem. I stole some good ideas from ao_portaudio's ringbuffer and went from there. The main difference is this one stores wpos and rpos which are absolute positions in an "infinite" buffer. To find the actual position for writing / reading just apply modulo size. The producer only modifies wpos while the consumer only modifies rpos. This makes it pretty easy to reason about and make the operations thread safe by using barriers (thread safety is guaranteed only in the Single-Producer/Single- Consumer case). Also adapted ao_coreaudio to use this ringbuffer.
2013-05-23 19:23:32 +00:00
*/
osdep: rename atomics.h to atomic.h The standard header is stdatomic.h, so the extra "s" freaks me out every time I look at it.
2016-09-07 09:26:25 +00:00
#ifndef MP_ATOMIC_H
#define MP_ATOMIC_H
core: add a spsc ringbuffer implementation Currently every single AO was implementing it's own ringbuffer, many times with slightly different semantics. This is an attempt to fix the problem. I stole some good ideas from ao_portaudio's ringbuffer and went from there. The main difference is this one stores wpos and rpos which are absolute positions in an "infinite" buffer. To find the actual position for writing / reading just apply modulo size. The producer only modifies wpos while the consumer only modifies rpos. This makes it pretty easy to reason about and make the operations thread safe by using barriers (thread safety is guaranteed only in the Single-Producer/Single- Consumer case). Also adapted ao_coreaudio to use this ringbuffer.
2013-05-23 19:23:32 +00:00
atomics: switch to C11 stdatomic.h In my opinion, we shouldn't use atomics at all, but ok. This switches the mpv code to use C11 stdatomic.h, and for compilers that don't support stdatomic.h yet, we emulate the subset used by mpv using the builtins commonly provided by gcc and clang. This supersedes an earlier similar attempt by Kovensky. That attempt unfortunately relied on a big copypasted freebsd header (which also depended on much more highly compiler-specific functionality, defined reserved symbols, etc.), so it had to be NIH'ed. Some issues: - C11 says default initialization of atomics "produces a valid state", but it's not sure whether the stored value is really 0. But we rely on this. - I'm pretty sure our use of the __atomic... builtins is/was incorrect. We don't use atomic load/store intrinsics, and access stuff directly. - Our wrapper actually does stricter typechecking than the stdatomic.h implementation by gcc 4.9. We make the atomic types incompatible with normal types by wrapping them into structs. (The FreeBSD wrapper does the same.) - I couldn't test on MinGW.
2014-05-20 23:04:47 +00:00
#include <inttypes.h>
compat: use __atomic operations instead of __sync, when present Fixes #434 Fixes #437
2013-12-30 15:33:50 +00:00
#include "config.h"
atomics: switch to C11 stdatomic.h In my opinion, we shouldn't use atomics at all, but ok. This switches the mpv code to use C11 stdatomic.h, and for compilers that don't support stdatomic.h yet, we emulate the subset used by mpv using the builtins commonly provided by gcc and clang. This supersedes an earlier similar attempt by Kovensky. That attempt unfortunately relied on a big copypasted freebsd header (which also depended on much more highly compiler-specific functionality, defined reserved symbols, etc.), so it had to be NIH'ed. Some issues: - C11 says default initialization of atomics "produces a valid state", but it's not sure whether the stored value is really 0. But we rely on this. - I'm pretty sure our use of the __atomic... builtins is/was incorrect. We don't use atomic load/store intrinsics, and access stuff directly. - Our wrapper actually does stricter typechecking than the stdatomic.h implementation by gcc 4.9. We make the atomic types incompatible with normal types by wrapping them into structs. (The FreeBSD wrapper does the same.) - I couldn't test on MinGW.
2014-05-20 23:04:47 +00:00
#if HAVE_STDATOMIC
#include <stdatomic.h>
ao: simplify hack for float atomics stdatomic.h defines no atomic_float typedef. We can't just use _Atomic unconditionally, because we support compilers without C11 atomics. So just create a custom atomic_float typedef in the wrapper, which uses _Atomic in the C11 code path.
2017-11-30 00:14:33 +00:00
typedef _Atomic float mp_atomic_float;
atomics: switch to C11 stdatomic.h In my opinion, we shouldn't use atomics at all, but ok. This switches the mpv code to use C11 stdatomic.h, and for compilers that don't support stdatomic.h yet, we emulate the subset used by mpv using the builtins commonly provided by gcc and clang. This supersedes an earlier similar attempt by Kovensky. That attempt unfortunately relied on a big copypasted freebsd header (which also depended on much more highly compiler-specific functionality, defined reserved symbols, etc.), so it had to be NIH'ed. Some issues: - C11 says default initialization of atomics "produces a valid state", but it's not sure whether the stored value is really 0. But we rely on this. - I'm pretty sure our use of the __atomic... builtins is/was incorrect. We don't use atomic load/store intrinsics, and access stuff directly. - Our wrapper actually does stricter typechecking than the stdatomic.h implementation by gcc 4.9. We make the atomic types incompatible with normal types by wrapping them into structs. (The FreeBSD wrapper does the same.) - I couldn't test on MinGW.
2014-05-20 23:04:47 +00:00
#else
// Emulate the parts of C11 stdatomic.h needed by mpv.
atomic: remove __atomic builtin usage Using these was a temporary solution while some compilers implemented the underlying atomic mechanisms, but not the C11 language parts (or that's what I guess). Not really useful for us anymore. Also, there is the slight risk of having subtly incorrect semantics by using potentially changing compiler internals and such.
2017-02-13 05:45:40 +00:00
typedef struct { unsigned long v; } atomic_ulong;
typedef struct { int v; } atomic_int;
typedef struct { unsigned int v; } atomic_uint;
typedef struct { _Bool v; } atomic_bool;
typedef struct { long long v; } atomic_llong;
typedef struct { uint_least32_t v; } atomic_uint_least32_t;
typedef struct { unsigned long long v; } atomic_ullong;
atomics: switch to C11 stdatomic.h In my opinion, we shouldn't use atomics at all, but ok. This switches the mpv code to use C11 stdatomic.h, and for compilers that don't support stdatomic.h yet, we emulate the subset used by mpv using the builtins commonly provided by gcc and clang. This supersedes an earlier similar attempt by Kovensky. That attempt unfortunately relied on a big copypasted freebsd header (which also depended on much more highly compiler-specific functionality, defined reserved symbols, etc.), so it had to be NIH'ed. Some issues: - C11 says default initialization of atomics "produces a valid state", but it's not sure whether the stored value is really 0. But we rely on this. - I'm pretty sure our use of the __atomic... builtins is/was incorrect. We don't use atomic load/store intrinsics, and access stuff directly. - Our wrapper actually does stricter typechecking than the stdatomic.h implementation by gcc 4.9. We make the atomic types incompatible with normal types by wrapping them into structs. (The FreeBSD wrapper does the same.) - I couldn't test on MinGW.
2014-05-20 23:04:47 +00:00
ao: simplify hack for float atomics stdatomic.h defines no atomic_float typedef. We can't just use _Atomic unconditionally, because we support compilers without C11 atomics. So just create a custom atomic_float typedef in the wrapper, which uses _Atomic in the C11 code path.
2017-11-30 00:14:33 +00:00
typedef struct { float v; } mp_atomic_float;
atomics: switch to C11 stdatomic.h In my opinion, we shouldn't use atomics at all, but ok. This switches the mpv code to use C11 stdatomic.h, and for compilers that don't support stdatomic.h yet, we emulate the subset used by mpv using the builtins commonly provided by gcc and clang. This supersedes an earlier similar attempt by Kovensky. That attempt unfortunately relied on a big copypasted freebsd header (which also depended on much more highly compiler-specific functionality, defined reserved symbols, etc.), so it had to be NIH'ed. Some issues: - C11 says default initialization of atomics "produces a valid state", but it's not sure whether the stored value is really 0. But we rely on this. - I'm pretty sure our use of the __atomic... builtins is/was incorrect. We don't use atomic load/store intrinsics, and access stuff directly. - Our wrapper actually does stricter typechecking than the stdatomic.h implementation by gcc 4.9. We make the atomic types incompatible with normal types by wrapping them into structs. (The FreeBSD wrapper does the same.) - I couldn't test on MinGW.
2014-05-20 23:04:47 +00:00
#define ATOMIC_VAR_INIT(x) \
{.v = (x)}
atomics: more correct usage of gcc/clang __atomic builtins This should be more correct. The builtins were made to directly map to C11, and the way we use them is now relatively close to how gcc implements atomics in 4.9. In particular, we make use of the load and store builtins. I'm not entirely sure why gcc didn't support stdatomic.h in 4.8 already. Maybe support for the builtins was incomplete or broken - so there's a lot of room for doubt about the correctness of this.
2014-05-20 23:27:54 +00:00
msg: use relaxed atomics for log level test This should be sufficient. If stdatomic.h is not available, we make no difference.
2015-03-02 18:09:31 +00:00
#define memory_order_relaxed 1
#define memory_order_seq_cst 2
#define atomic_load_explicit(p, e) atomic_load(p)
atomics: readd some emulation This time it's emulation that's supposed to work (not just dummied out). Unlike the previous emulation, no mpv code has to be disabled, and everything should work (albeit possibly a bit slowly). On the other hand, it's not possible to implement this kind of emulation without compiler support. We use GNU statement expressions and __typeof__ in this case. This code is inactive if stdatomic.h is available.
2016-09-06 18:13:30 +00:00
#include <pthread.h>
extern pthread_mutex_t mp_atomic_mutex;
#define atomic_load(p) \
({ __typeof__(p) p_ = (p); \
pthread_mutex_lock(&mp_atomic_mutex); \
__typeof__(p_->v) v = p_->v; \
pthread_mutex_unlock(&mp_atomic_mutex); \
v; })
#define atomic_store(p, val) \
({ __typeof__(val) val_ = (val); \
__typeof__(p) p_ = (p); \
pthread_mutex_lock(&mp_atomic_mutex); \
p_->v = val_; \
pthread_mutex_unlock(&mp_atomic_mutex); })
#define atomic_fetch_op(a, b, op) \
({ __typeof__(a) a_ = (a); \
__typeof__(b) b_ = (b); \
pthread_mutex_lock(&mp_atomic_mutex); \
__typeof__(a_->v) v = a_->v; \
a_->v = v op b_; \
pthread_mutex_unlock(&mp_atomic_mutex); \
v; })
#define atomic_fetch_add(a, b) atomic_fetch_op(a, b, +)
#define atomic_fetch_and(a, b) atomic_fetch_op(a, b, &)
#define atomic_fetch_or(a, b) atomic_fetch_op(a, b, |)
#define atomic_compare_exchange_strong(p, old, new) \
({ __typeof__(p) p_ = (p); \
__typeof__(old) old_ = (old); \
__typeof__(new) new_ = (new); \
pthread_mutex_lock(&mp_atomic_mutex); \
int res = p_->v == *old_; \
if (res) { \
p_->v = new_; \
} else { \
*old_ = p_->v; \
} \
pthread_mutex_unlock(&mp_atomic_mutex); \
res; })
atomics: switch to C11 stdatomic.h In my opinion, we shouldn't use atomics at all, but ok. This switches the mpv code to use C11 stdatomic.h, and for compilers that don't support stdatomic.h yet, we emulate the subset used by mpv using the builtins commonly provided by gcc and clang. This supersedes an earlier similar attempt by Kovensky. That attempt unfortunately relied on a big copypasted freebsd header (which also depended on much more highly compiler-specific functionality, defined reserved symbols, etc.), so it had to be NIH'ed. Some issues: - C11 says default initialization of atomics "produces a valid state", but it's not sure whether the stored value is really 0. But we rely on this. - I'm pretty sure our use of the __atomic... builtins is/was incorrect. We don't use atomic load/store intrinsics, and access stuff directly. - Our wrapper actually does stricter typechecking than the stdatomic.h implementation by gcc 4.9. We make the atomic types incompatible with normal types by wrapping them into structs. (The FreeBSD wrapper does the same.) - I couldn't test on MinGW.
2014-05-20 23:04:47 +00:00
#endif /* else HAVE_STDATOMIC */
#endif