Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

fix(simd.h): Address NEON issues #4143

Merged
merged 1 commit into from
Feb 8, 2024
Merged

fix(simd.h): Address NEON issues #4143

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion src/cmake/compiler.cmake
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -303,7 +303,7 @@ set (SIMD_COMPILE_FLAGS "")
if (NOT USE_SIMD STREQUAL "")
message (STATUS "Compiling with SIMD level ${USE_SIMD}")
if (USE_SIMD STREQUAL "0")
set (SIMD_COMPILE_FLAGS ${SIMD_COMPILE_FLAGS} "-DOIIO_NO_SSE=1")
set (SIMD_COMPILE_FLAGS ${SIMD_COMPILE_FLAGS} "-DOIIO_NO_SIMD=1")
else ()
string (REPLACE "," ";" SIMD_FEATURE_LIST ${USE_SIMD})
foreach (feature ${SIMD_FEATURE_LIST})
Expand Down
13 changes: 10 additions & 3 deletions src/include/OpenImageIO/simd.h
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -77,6 +77,13 @@
// OIIO_SIMD_HAS_SIMD8 : nonzero if vfloat8, vint8, vbool8 are defined
// OIIO_SIMD_HAS_SIMD16 : nonzero if vfloat16, vint16, vbool16 are defined

#ifdef OIIO_NO_SIMD /* Request to disable all SIMD */
# define OIIO_NO_SSE 1
# define OIIO_NO_AVX 1
# define OIIO_NO_AVX2 1
# define OIIO_NO_NEON 1
#endif

#if defined(_M_ARM64) || defined(__aarch64__) || defined(__aarch64)
# ifndef __ARM_NEON__
# define __ARM_NEON__
Expand Down Expand Up @@ -4788,9 +4795,9 @@ OIIO_FORCEINLINE void vint4::store (unsigned char *values) const {
_mm_store_ss((float*)values, _mm_castsi128_ps(val8));
#elif OIIO_SIMD_NEON
vint4 clamped = m_simd & vint4(0xff);
simd_t val16 = vcombine_s16(vqmovn_s32(clamped), vdup_n_s16(0));
simd_t val8 = vcombine_u8(vqmovun_s16(val16), vdup_n_u8(0));
vst1q_lane_u32((uint32_t*)values, val8, 0);
int16x8_t val16 = vcombine_s16(vqmovn_s32(clamped), vdup_n_s16(0));
uint8x16_t val8 = vcombine_u8(vqmovun_s16(val16), vdup_n_u8(0));
vst1q_lane_u32((uint32_t*)values, vreinterpretq_u32_u8(val8), 0);
#else
SIMD_DO (values[i] = m_val[i]);
#endif
Expand Down
Loading