swscale/x86/swscale: Remove obsolete and harmful MMX(EXT) functions

x64 always has MMX, MMXEXT, SSE and SSE2 and this means
that some functions for MMX, MMXEXT, SSE and 3dnow are always
overridden by other functions (unless one e.g. explicitly
disables SSE2). So given that the only systems that
benefit from these functions are truely ancient 32bit x86s
they are removed.

Moreover, some of the removed code was buggy/not bitexact
and lead to failures involving the f32le and f32be versions of
gray, gbrp and gbrap on x86-32 when SSE2 was not disabled.
See e.g.
https://fate.ffmpeg.org/report.cgi?time=20220609221253&slot=x86_32-debian-kfreebsd-gcc-4.4-cpuflags-mmx

Notice that yuv2yuvX_mmx is not removed, because it is used
by SSE3 and AVX2 as fallback in case of unaligned data and
also for tail processing. I don't know why yuv2yuvX_mmxext
isn't being used for this; an earlier version [1] of
554c2bc7086f49ef5a6a989ad6bc4bc11807eb6f used it, but
the version that was eventually applied does not.

[1]: https://ffmpeg.org/pipermail/ffmpeg-devel/2020-November/272124.html

Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>

1 files changed, 3 insertions, 27 deletions

diff --git a/libswscale/x86/output.asm b/libswscale/x86/output.asm
index 1e498fddf6..84e94baaf6 100644
--- a/libswscale/x86/output.asm
+++ b/libswscale/x86/output.asm
@@ -312,11 +312,9 @@ cglobal yuv2planeX_%1, %3, 8, %2, filter, fltsize, src, dst, w, dither, offset
 %endif ; %1 == 8/9/10/16
 %endmacro
 
-%if ARCH_X86_32
+%if ARCH_X86_32 && HAVE_ALIGNED_STACK == 0
 INIT_MMX mmxext
 yuv2planeX_fn  8,  0, 7
-yuv2planeX_fn  9,  0, 5
-yuv2planeX_fn 10,  0, 5
 %endif
 
 INIT_XMM sse2
@@ -407,19 +405,11 @@ cglobal yuv2plane1_%1, %3, %3, %2, src, dst, w, dither, offset
     movq            m3, [ditherq]        ; dither
     test       offsetd, offsetd
     jz              .no_rot
-%if mmsize == 16
     punpcklqdq      m3, m3
-%endif ; mmsize == 16
     PALIGNR         m3, m3, 3, m2
 .no_rot:
-%if mmsize == 8
-    mova            m2, m3
-    punpckhbw       m3, m4               ; byte->word
-    punpcklbw       m2, m4               ; byte->word
-%else
     punpcklbw       m3, m4
     mova            m2, m3
-%endif
 %elif %1 == 9
     pxor            m4, m4
     mova            m3, [pw_512]
@@ -431,36 +421,22 @@ cglobal yuv2plane1_%1, %3, %3, %2, src, dst, w, dither, offset
 %else ; %1 == 16
 %if cpuflag(sse4) ; sse4/avx
     mova            m4, [pd_4]
-%else ; mmx/sse2
+%else ; sse2
     mova            m4, [pd_4min0x40000]
     mova            m5, [minshort]
-%endif ; mmx/sse2/sse4/avx
+%endif ; sse2/sse4/avx
 %endif ; %1 == ..
 
     ; actual pixel scaling
-%if mmsize == 8
-    yuv2plane1_mainloop %1, a
-%else ; mmsize == 16
     test          dstq, 15
     jnz .unaligned
     yuv2plane1_mainloop %1, a
     REP_RET
 .unaligned:
     yuv2plane1_mainloop %1, u
-%endif ; mmsize == 8/16
     REP_RET
 %endmacro
 
-%if ARCH_X86_32
-INIT_MMX mmx
-yuv2plane1_fn  8, 0, 5
-yuv2plane1_fn 16, 0, 3
-
-INIT_MMX mmxext
-yuv2plane1_fn  9, 0, 3
-yuv2plane1_fn 10, 0, 3
-%endif
-
 INIT_XMM sse2
 yuv2plane1_fn  8, 5, 5
 yuv2plane1_fn  9, 5, 3