This gets rid of the variable-length scratch buffer by filtering 16
pixels at a time and writing directly to the destination. The extra
loads this requires to load the source values are compensated by not
doing a round-trip to memory before shifting.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Instead of using an evil VLA, fall back to C version when edge
emulation is needed. MPEG4 GMC is a rarely used fringe feature
so the speed loss is an acceptable cost for safer code.
Signed-off-by: Mans Rullgard <mans@mansr.com>
Discontinuous transmission is an addition to VAD/VBR operation, that
allows to greatly reduce bitrate for silent chunks or stationary
noises.
Signed-off-by: Martin Storsjö <martin@martin.st>
* qatar/master:
h264: don't touch H264Context->ref_count[] during MB decoding
x86: get_cpu_flags: add necessary ifdefs around function body
x86: Drop CPU detection intrinsics
x86: Add YASM implementations of cpuid and xgetbv from x264
Conflicts:
configure
libavcodec/h264_cabac.c
libavcodec/h264_cavlc.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '65d12900432ac880d764edbbd36818431484a76e':
configure: add --enable-lto option
x86: cpu: Break out test for cpuid capabilities into separate function
x86: ff_get_cpu_flags_x86(): Avoid a pointless variable indirection
build: Factor out mpegaudio dependencies to CONFIG_MPEGAUDIO
segment: Add comments about calls that only are relevant for some muxers
segment: Add an option for omitting the first header and final trailer
Conflicts:
configure
libavcodec/Makefile
libavformat/segment.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'a854362b40f0e458db5a1fb0d2612a5702ee0ace':
segment: Flush buffered data before finishing a segment
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'f7b240434c015056bc6319ddbdb8483757cc13e2':
segment: Set the resend_headers flag for each segment
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '378a6315b7c48195ffd94e6aa9aa6d663d42b35e':
segment: Add an option for disabling writing of a header/trailer to each segment
Conflicts:
libavformat/segment.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit 'eb447d515956b3ce182d9750083131735f00324c':
segment: Free and reinit the muxer before calling avformat_write_header
Conflicts:
libavformat/segment.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '73871dc96ff78053b9dcd0eb259b7f5a5308ec87':
segment: Use the public av_write_header/av_write_trailer functions
Conflicts:
libavformat/segment.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
* commit '0edae4e6286096023cdd6adea74722fa06029867':
segment: Properly create new AVStreams for the chained muxer
segment: Add a missing space
Conflicts:
libavformat/segment.c
Merged-by: Michael Niedermayer <michaelni@gmx.at>
The variable is copied to subsequent threads at the same time, so this
may cause wrong ref_count[] values to be copied to subsequent threads.
This bug was found using TSAN and Helgrind.
Original patch by Ronald, adapted with a local_ref_count by Clément,
following the suggestion of Michael Niedermayer.
Signed-off-by: Clément Bœsch <clement.boesch@smartjog.com>
Some HLS servers return 403 when the Range header is present. Disabling http
seekability probing prevents the header from being added.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
Add an tri-state (seek, non seek, automatic detection) option to HTTP to control seekability (default: automatic).
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
This only adds a subset of the frame rates that can be stored, a full list
would be more than 500 entries. Such full list could (and should) be added
if it has a usecase but the number of users of even the now added rates
seem rather small.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
The variable is copied to subsequent threads at the same time, so this
may cause wrong ref_count[] values to be copied to subsequent threads.
This bug was found using TSAN.
Signed-off-by: Luca Barbato <lu_zero@gentoo.org>
Discontinuous transmission is an addition to VAD/VBR operation, that
allows to greatly reduce bitrate for silent chunks or stationary
noises.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
The value used in allocation is based on a estimate of the
maximum size of the spectral coefficients multiplied with 2
and rounded up. The exact or a tighter limit should be
found and used instead. But this issue shouldnt be left
open until someone works on that.
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
ff_get_cpu_flags_x86() requires cpuid(), which is conditionally defined
elsewhere in the file. Surrounding the function body with ifdefs allows
building even when cpuid is not defined. An empty cpuflags mask is
returned in this case.
Now that there is CPU detection in YASM, there will always be one of
inline or external assembly enabled, which obviates the need to fall
back on CPU detection through compiler intrinsics.
