Fix crash of MSVC builds compiled with --system-libx264 and /MT (default) CRT.

Removes need for -Bsymbolic during linking.

Henrik Gramner authored 6 years ago and Anton Mitrofanov committed 5 years ago

Simplifies a lot of code and avoids having to export public asm functions.

Note that the force_align_arg_pointer function attribute is broken in clang
versions prior to 6.0.1 which may result in crashes, so make sure to either
use a newer clang version or a different compiler.

pkg-config doesn't like spaces in version string.

Some hardware decoders reject to decode streams with non-zero
constraint_set4_flag/constraint_set5_flag.

Result was never used in that case.

Bug report by Xuezhi Yan.

Bug report by Hongxu Chen.

It now includes pkg-config, so lavf can be detected.

Supported targets:
 - debian amd64
 - debian aarch64
 - windows 32 bit
 - windows 64 bit
 - macos 64bit

The tests are ran on all supported targets (via wine on windows).

The release jobs are only available on master/stable branches in
videolan/x264 repository, and must be ran manually when a developer
wishes to upload the artifacts.

This will break decoding with older versions of FFmpeg/Libav.

Henrik Gramner authored 6 years ago and Anton Mitrofanov committed 6 years ago

Warn when the following are used without the appropriate cpuflag:
 * YMM and ZMM registers
 * 'pextrw' with a memory operand
 * GPR instruction set extensions

Henrik Gramner authored 6 years ago and Anton Mitrofanov committed 6 years ago

Allows for marking symbols as having limited global scope, similar to
using 'hidden' symbol visibility on ELF.

Henrik Gramner authored 6 years ago and Anton Mitrofanov committed 6 years ago

Allows for automatic command line completion for both options and values.

Options such as --input-csp and --input-fmt will dynamically retrieve
supported values from libavformat when compiled with lavf support.

Execute 'source tools/bash-autocomplete.sh' in bash to enable.

Yusuke Nakamura authored 7 years ago and Anton Mitrofanov committed 6 years ago

Progressive High, Constrained High, and Progressive High 10.

Even in Main profile, constraint_set4_flag is now set to 1 if progressive,
and constraint_set5_flag is set to 1 if no B-slices are present.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

Around a ~1% speedup to the overall encoding for --slow.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

No overall speedup, just tidier code.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

Seems to give about a 1-2% overall speedup on --slow.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

Around a ~2% speedup to the overall encoding for --slow.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

Yet another use of xxpermdi, another 10% gain.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

About 20% faster.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

Now 4x4 is as slow as C and 4x8 is a 2% faster than before.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

And use it on the other satd > 8.

5-10% faster depending on the size.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

10% faster.

Luca Barbato authored 6 years ago and Anton Mitrofanov committed 6 years ago

4x faster than C.

Henrik Gramner authored 6 years ago and Anton Mitrofanov committed 6 years ago

Most x86-64 operating systems nowadays doesn't even allow .text relocations
in object files any more, and there is no measurable overall performance
difference from using RIP-relative addressing in x264 asm.

Enforcing PIC reduces complexity and simplifies testing.