Refactoring mmx2/mmxext YASM code with cpuflags will force renames.
So switching to a consistent naming scheme beforehand is sensible.
The name "mmxext" is more official and widespread and also the name
of the CPU flag, as reported e.g. by the Linux kernel.

Fixes overflows for large image sizes.

Found-by: Mateusz "j00ru" Jurczyk and Gynvael Coldwind
CC: libav-stable@libav.org

Revert "swscale: update context offsets after removal of AlpMmxFilter."
(commit a95e3fa90b)
and
Revert "swscale: Remove some write-only variables related to alpha handling."
(commit 9d03cb9fc5).

They broke alpha handling - it's the evil inline asm that still uses that
variable, so it's not truely write-only.

Prevents crashes on 1-tap filter (unscaled). Also rename "bguf" argument
to "vbuf", seems that was a typo.

Fixes bug 221.

CC: libav-stable@libav.org

Also implement sse2/ssse3/avx versions.

Additional comments from Måns Rullgard have been integrated
by Reinhard Tartler.

Signed-off-by: Reinhard Tartler <siretart@tauware.de>

Also implement SSE2/AVX variants.

Also implement MMX/MMX2 versions and SSE4 versions.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>

Speed: from 3.9x to 9.6x speed improvement over C, and some small
(up to 15%) speed improvements over existing MMX code (particularly
for bigger filters).

This allows using more specific implementations for chroma/luma, e.g.
we can make assumptions on filterSize being constant, thus avoiding
that test at runtime.

Based on a somewhat similar idea in FFmpeg's swscale copy.

We would use the second half of the U plane buffer, rather than the
V plane buffer, to output the V plane pixels.

For 9/10bit, it means we don't have to upscale to 16bit before
actual scaling or pixel format conversion, and thus a performance
gain.

This means that precision is retained when scaling between sample
formats with >8 bits per component (48bit RGB, 16bit grayscale,
9/10/16bit YUV).

Remove unused variables "flags" and "dstFormat" in yuv2packed1,
merge source rows per plane for yuv2packed[12], and make every
source argument int16_t (some where invalidly set to uint16_t).
This prevents stack pollution and is part of the Great Evil Plan
to simplify swscale.

This prevents a crash when converting to NV12/21 without the bitexact
flags enabled.

Remove inline keyword for functions that are only called through
their function pointers (and thus cannot be inlined); add av_cold
keyword to init function, and use av_always_inline instead of
inline for functions that must be inlined for performance reasons.

This prevents the following compiler warnings: "warning:
initialization from incompatible pointer type". Since the
variables are only ever used in inline assembly, their type
is actually irrelevant (so the part where it was wrong did
not invoke any buggy behaviour).

This way, they look like regular code, which is easier to
understand.

Make yuv2yuvX16_c a function pointer for yuv2yuvX(), so that the
function pointer becomes bitdepth-independent.

Signed-off-by: Ronald S. Bultje <rsbultje@gmail.com>

libswscale/x86/swscale_template.c:2072: warning: unused variable ‘canMMX2BeUsed’
libswscale/x86/swscale_template.c:2145: warning: unused variable ‘canMMX2BeUsed’
libswscale/x86/swscale_template.c:2209: warning: unused variable ‘chrVPixBuf’
libswscale/x86/swscale_template.c:2237: warning: unused variable ‘chrVSrcPtr’

Preparatory step to implement support for sizes > VOFW.

Interleave macros and code so that it's easier to find the
actual code that belongs to a function. Also reindent where
appropriate and remove dead code.

This allows cutting up the function in much smaller and easier-
to-maintain chunks.

If that flag is set, swScale() already proxies the call to
yuv2rgbXinC_full(). Therefore, this flag is never set when
yuv2packed1() is called.

Instead, only set the function pointers if bitexact flag is
not set during initialization. Since a change in flags triggers
a re-init anyway, this doesn't situations where flag values
change during runtime.

Instead, set function pointers conditionally during init. This
patch also reveals a whole branch of dead assembly code that is
therefore also removed.