checkasm10 with seed=511142008 failed on win32 gcc builds.

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.

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.

Clang emits aligned AVX stores for things like zeroing stack-allocated
variables when using -mavx even with -fno-tree-vectorize set which can
result in crashes if this occurs before we've realigned the stack.

Previously we only ensured that the stack was realigned before calling
assembly functions that accesses stack-allocated buffers but this is
not sufficient. Fix the issue by changing the stack realignment to
instead occur immediately in all CLI, API and thread entry points.

Martin Storsjö authored 7 years ago and Anton Mitrofanov committed 7 years ago

This version supports converting aarch64 assembly for MS armasm64.exe.

Henrik Gramner authored 7 years ago and Anton Mitrofanov committed 7 years ago

Takes advantage of opmasks to avoid having to use scalar code for the tail.

Also make some slight improvements to the checkasm test.

Vittorio Giovara authored 8 years ago and Anton Mitrofanov committed 7 years ago

Add 'i_bitdepth' to x264_param_t with the corresponding '--output-depth' CLI
option to set the bit depth at runtime.

Drop the 'x264_bit_depth' global variable. Rather than hardcoding it to an
incorrect value, it's preferable to induce a linking failure. If applications
relies on this symbol this will make it more obvious where the problem is.

Add Makefile rules that compiles modules with different bit depths. Assembly
on x86 is prefixed with the 'private_prefix' define, while all other archs
modify their function prefix internally.

Templatize the main C library, x86/x86_64 assembly, ARM assembly, AARCH64
assembly, PowerPC assembly, and MIPS assembly.

The depth and cache CLI filters heavily depend on bit depth size, so they
need to be duplicated for each value. This means having to rename these
filters, and adjust the callers to use the right version.

Unfortunately the threaded input CLI module inherits a common.h dependency
(input/frame -> common/threadpool -> common/frame -> common/common) which
is extremely complicated to address in a sensible way. Instead duplicate
the module and select the appropriate one at run time.

Each bitdepth needs different checkasm compilation rules, so split the main
checkasm target into two executables.

Uses gathers and scatters in combination with conflict detections to
vectorize the scalar part.

Also improve the checkasm test to try different mb_y values and check
for out-of-bounds writes.

Henrik Gramner authored 8 years ago and Anton Mitrofanov committed 7 years ago

These levels were added in the 2016-10 revision of the H.264 specification and
improves support for content with high resolutions and/or high frame rates.

Level 6.2 supports 8K resolution at 120 fps.

Also shrink the x264_levels array by using smaller data types.

Martin Storsjö authored 7 years ago and Henrik Gramner committed 7 years ago

Prior to this, this loop hasn't run at all. The condition has been
the same since it was introduced in 5b0cb86f.

This issue was pointed out by a clang warning.

Covers all variants: 4x4, 4x8, 4x16, 8x4, 8x8, 8x16, 16x8, and 16x16.

The functions are only ever called with pointers to fenc and fdec and the
strides are always constant so there's no point in having them as parameters.

Cover both the U and V planes in a single function call. This is more
efficient with SIMD, especially with the wider vectors provided by AVX2 and
AVX-512, even when accounting for losing the possibility of early termination.

Drop the MMX and XOP implementations, update the rest of the x86 assembly
to match the new behavior. Also enable high bit-depth in the AVX2 version.

Comment out the ARM, AARCH64, and MIPS MSA assembly for now.

Also drop the MMX version and make some slight improvements to the SSE2,
SSSE3, AVX, and AVX2 versions.

Reorder some elements in the x264_t.mb.pic struct to reduce the amount
of padding required.

Also drop the MMX implementation in favor of SSE.

Reorder some elements in the x264_mb_analysis_list_t struct to reduce the
amount of padding required.

Also drop the MMX implementation in favor of SSE.

Also make the AVX and AVX2 implementations slightly faster.

The vperm* instructions ignores unused bits, so we can pack the permutation
indices together to save cache and just use a shift to get the right values.

YMM and ZMM registers on x86 are turned off to save power when they haven't
been used for some period of time. When they are used there will be a
"warmup" period during which performance will be reduced and inconsistent
which is problematic when trying to benchmark individual functions.

Periodically issue "dummy" instructions that uses those registers to
prevent them from being powered down. The end result is more consitent
benchmark results.

AVX-512 consists of a plethora of different extensions, but in order to keep
things a bit more manageable we group together the following extensions
under a single baseline cpu flag which should cover SKL-X and future CPUs:
 * AVX-512 Foundation (F)
 * AVX-512 Conflict Detection Instructions (CD)
 * AVX-512 Byte and Word Instructions (BW)
 * AVX-512 Doubleword and Quadword Instructions (DQ)
 * AVX-512 Vector Length Extensions (VL)

On x86-64 AVX-512 provides 16 additional vector registers, prefer using
those over existing ones since it allows us to avoid using `vzeroupper`
unless more than 16 vector registers are required. They also happen to
be volatile on Windows which means that we don't need to save and restore
existing xmm register contents unless more than 22 vector registers are
required.

Also take the opportunity to drop X264_CPU_CMOV and X264_CPU_SLOW_CTZ while
we're breaking API by messing with the cpu flags since they weren't really
used for anything.

Big thanks to Intel for their support.

Simplifies writing assembly code that depends on available instructions.

LZCNT implies SSE2
BMI1 implies AVX+LZCNT
AVX2 implies BMI2

Skip printing LZCNT under CPU capabilities when BMI1 or BMI2 is available,
and don't print FMA4 when FMA3 is available.

Packed YUV is arguably more common than planar YUV when dealing with raw
4:2:2 content.

We can utilize the existing plane_copy_deinterleave() functions with some
additional minor constraints (we cannot assume any particular alignment
or overread the input buffer).

Enables assembly optimizations on x86.

Martin Storsjö authored 8 years ago and Henrik Gramner committed 7 years ago

Set up the right gas-preprocessor as assembler frontend in these cases,
using armasm as actual assembler.

Don't try to add the -mcpu -mfpu options in this case.

Check whether the compiler actually supports inline assembly.

Check for the ARMv7 features in a different way for the MSVC compiler.