diff options
| author | James Almer <jamrial@gmail.com> | 2021-11-15 23:15:38 -0300 |
|---|---|---|
| committer | James Almer <jamrial@gmail.com> | 2021-11-19 11:21:03 -0300 |
| commit | 67b92d68c68bc2fac682ca3671003913498bc44c (patch) | |
| tree | 2c27d184139ef19282d2a96241ca3b2e0c35bb18 /libavutil | |
| parent | b2aec70bd6bbf4620d363dfdc7d9e54f2f32f9ee (diff) | |
| download | ffmpeg-67b92d68c68bc2fac682ca3671003913498bc44c.tar.gz | |
x86/intmath: add VEX encoded versions of av_clipf() and av_clipd()
Prevents mixing inlined SSE instructions and AVX instructions when the compiler
generates the latter.
Signed-off-by: James Almer <jamrial@gmail.com>
Diffstat (limited to 'libavutil')
| -rw-r--r-- | libavutil/x86/intmath.h | 30 |
1 files changed, 30 insertions, 0 deletions
diff --git a/libavutil/x86/intmath.h b/libavutil/x86/intmath.h index 1520c25ec9..8a6b5ae261 100644 --- a/libavutil/x86/intmath.h +++ b/libavutil/x86/intmath.h @@ -134,6 +134,36 @@ static av_always_inline av_const float av_clipf_sse(float a, float amin, float a #endif /* __SSE__ */ +#if defined(__AVX__) && !defined(__INTEL_COMPILER) + +#undef av_clipd +#define av_clipd av_clipd_avx +static av_always_inline av_const double av_clipd_avx(double a, double amin, double amax) +{ +#if defined(ASSERT_LEVEL) && ASSERT_LEVEL >= 2 + if (amin > amax) abort(); +#endif + __asm__ ("vmaxsd %1, %0, %0 \n\t" + "vminsd %2, %0, %0 \n\t" + : "+&x"(a) : "xm"(amin), "xm"(amax)); + return a; +} + +#undef av_clipf +#define av_clipf av_clipf_avx +static av_always_inline av_const float av_clipf_avx(float a, float amin, float amax) +{ +#if defined(ASSERT_LEVEL) && ASSERT_LEVEL >= 2 + if (amin > amax) abort(); +#endif + __asm__ ("vmaxss %1, %0, %0 \n\t" + "vminss %2, %0, %0 \n\t" + : "+&x"(a) : "xm"(amin), "xm"(amax)); + return a; +} + +#endif /* __AVX__ */ + #endif /* __GNUC__ */ #endif /* AVUTIL_X86_INTMATH_H */ |