Whenever av_gettime() is used to measure relative period of time,
av_gettime_relative() is prefered as it guarantee monotonic time
on supported platforms.
Signed-off-by: Olivier Langlois <olivier@trillion01.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
These functions are using the POSIX clock_gettime() function with the
CLOCK_MONOTONIC clock id. If these are not present on the targeted
platform, the new functions will fallback on using the original realtime functions
av_gettime() and av_usleep().
Monotonic support can be added on other platforms with their
equivalent native system API eventually if possible.
Whenever time is requested to measure relative time, the monotonic clock,
when available, is superior to the system realtime clock because it is
not affected by discontinuous jumps in the system time
In a future step, offering the flexibility to let the user choose between
rt and monotonic clock for avdevices packets will be investigated.
It is very easy to experience the issues that this patch attempt to address
by rewinding back in the past the system time while ffmpeg is running.
this is breaking the ffmpeg report printing (ffmepg.c:print_report()) and
the the rate emulator functionality (-re) without the patch.
Signed-off-by: Olivier Langlois <olivier@trillion01.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
* cus/stable:
ffplay: add support for toggling between multiple video filters with the w key
ffplay: fix typo in docs
ffplay: try multiple sample rates if audio open fails
cmdutils: replace usages of "#ifdef __MINGW32__" with "#ifdef _WIN32" because MSVC only defines _WIN32
ffplay: fix compilation with Visual Studio
ffplay: increase AV_SYNC_THRESHOLD_MIN to 0.04
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Less than 0.04 sec delays should not be noticable, and it helps us with 50fps
content where some timing errors can cause a frame dup where it is not really
necessary.
Signed-off-by: Marton Balint <cus@passwd.hu>
The second argument is a temp register for non-SSSE3 cases
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
it was always 1 before with ISOBMFF(cherry picked from commit fb1402b1ec78d80acd6ced76bf78d65560965c4c)
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Use default values if parsed variable is found not to
have any value. Avoids crashing at strlen for salt/user
on the auth call afterwards and needless NULL assignments
for the rest (default is already NULL for those).
Should fix Coverity Scan issues #966644 and #966645
Signed-off-by: Reynaldo H. Verdejo Pinochet <r.verdejo@sisa.samsung.com>
* commit '4d25c2bc76f0a708c2ba470b6bfeb547efe37115':
build: Support building using clang-usan and disable-optimization
Change not merged, as it has been fixed differently and this change
would prevent disabling optimizations with ubsan
See: bebce653e5
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '7c377f0e74de8ab57f6baa64f3ec197c82104764':
avisynth: set duration of audio streams
See: ae8e2d7634
Merged-by: Michael Niedermayer <michaelni@gmx.at>
Clang Undefined Behaviour Sanitizer seems to make non-constant constant
expressions, making the compilation fail when they are used within
inline assembly if no optimization is requested.
Add -O1 to the cflags to hide the misbehaviour.
Bug-Id: 683
