fdo-mirrors/mesa
1
0
Fork 0
mirror of https://gitlab.freedesktop.org/mesa/mesa.git synced 2025-12-29 10:10:09 +01:00
Code Issues Projects Releases Packages Wiki Activity Actions 2

swr/rast: _mm*_undefined_* implementations for gcc<4.9

Browse source 
Define these in terms of setzero for ancient gcc versions which don't
have the undefined intrinsics.

Cc: mesa-stable@lists.freedesktop.org

Reviewed-by: Bruce Cherniak <bruce.cherniak@intel.com>
(cherry picked from commit f0a22956be)

Squashed with:

swr: modifications to allow gcc-4.8 compilation

Code unconditionally used avx2 intrinsics in the avx compilation.

The simd intrinsics library we used has diverged significantly between
branch and master; the non-“undefined intrinsics” portion is specific
to the branch.

This complements:
f0a22956be “swr/rast: _mm*_undefined_* implementations for gcc<4.9”

And makes this branch equivalent with the additional master patch:
d50ef7332c “swr/rast: don't use _mm256_fmsub_ps in AVX code”
This commit is contained in:
Tim Rowley 2017-07-05 13:26:02 -05:00 • committed by Andres Gomez
parent e1842e8acb
commit 7bd9eccbfe
2 changed files with 8 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
src/gallium/drivers/swr/rasterizer/common/os.h
View file

@ -171,6 +171,12 @@ void _mm256_storeu2_m128i(__m128i *hi, __m128i *lo, __m256i a)
_mm_storeu_si128((__m128i*)lo, _mm256_castsi256_si128(a));
_mm_storeu_si128((__m128i*)hi, _mm256_extractf128_si256(a, 0x1));
}
// gcc prior to 4.9 doesn't have _mm*_undefined_*
#if (__GNUC__) && (GCC_VERSION < 409000)
#define _mm_undefined_si128 _mm_setzero_si128
#define _mm256_undefined_ps _mm256_setzero_ps
#endif
#endif
inline

4
src/gallium/drivers/swr/rasterizer/common/simd16intrin.h
View file

@ -405,8 +405,8 @@ INLINE simd16mask SIMDAPI _simd16_movemask_pd(simd16scalard a)
INLINE uint64_t SIMDAPI _simd16_movemask_epi8(simd16scalari a)
{
uint32_t mask_lo = _mm256_movemask_epi8(a.lo);
uint32_t mask_hi = _mm256_movemask_epi8(a.hi);
uint32_t mask_lo = _simd_movemask_epi8(a.lo);
uint32_t mask_hi = _simd_movemask_epi8(a.hi);
return static_cast<uint64_t>(mask_lo) | (static_cast<uint64_t>(mask_hi) << 32);
}