| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | Fix psimd_qfma_f32 for FMA-enabled x86 processorsupstream-master | Marat Dukhan | 2020-05-17 | 1 | -2/+2 |
| | | |||||
| * | clang-cl compatibility | Marat Dukhan | 2020-04-16 | 1 | -2/+2 |
| | | |||||
| * | Default to not use QFMA instructions on WAsm SIMD | Marat Dukhan | 2020-02-21 | 1 | -1/+1 |
| | | |||||
| * | psimd_andnotmask_f32 function | Marat Dukhan | 2019-12-26 | 1 | -0/+5 |
| | | |||||
| * | Fix bug in psimd_abs_f32 | Marat Dukhan | 2019-12-26 | 1 | -1/+1 |
| | | |||||
| * | psimd_div_f32 function | Marat Dukhan | 2019-12-06 | 1 | -0/+4 |
| | | |||||
| * | Fix definition of psimd_signblend_u* | Marat Dukhan | 2019-09-30 | 1 | -3/+3 |
| | | |||||
| * | Require signed mask in blend of unsigned SIMD vectors | Marat Dukhan | 2019-09-30 | 1 | -7/+7 |
| | | | | | | The result of comparison of unsigned SIMD vectors is a signed SIMD vector. Using it as a mask with previous specification was incompatible with gcc. | ||||
| * | psimd_qfma_f32 function | Marat Dukhan | 2019-09-27 | 1 | -0/+15 |
| | | |||||
| * | psimd_loadN_s32 and psimd_loadN_u32 functions | Marat Dukhan | 2019-08-29 | 1 | -0/+36 |
| | | |||||
| * | Optimized blends and FP32 min/max for WAsm SIMD128 | Marat Dukhan | 2019-08-29 | 1 | -0/+18 |
| | | |||||
| * | psimd_load_splat_* functions | Marat Dukhan | 2019-08-29 | 1 | -0/+28 |
| | | | | | New psimd_load_splat_[u8/s8/u16/s16/u32/s32/f32] functions load a single value and duplicate it to all lanes | ||||
| * | psimd_storeN_s32 and psimd_storeN_u32 functions | Marat Dukhan | 2019-08-29 | 1 | -0/+42 |
| | | |||||
| * | Include SSE/AVX intrinsics headers | Marat Dukhan | 2018-09-06 | 1 | -2/+30 |
| | | |||||
| * | Add concat even/odd functions for 8-bit types | Marat Dukhan | 2018-09-06 | 1 | -0/+40 |
| | | |||||
| * | Add psimd_s32->psimd_f32 conversion function | Marat Dukhan | 2018-09-06 | 1 | -0/+12 |
| | | |||||
| * | Recognize both __ARM_NEON__ and __ARM_NEON as NEON macros | Marat Dukhan | 2018-09-06 | 1 | -21/+21 |
| | | |||||
| * | Splat vector lane | Marat Dukhan | 2017-04-15 | 1 | -0/+35 |
| | | |||||
| * | Vector reversal functions | Marat Dukhan | 2017-04-14 | 1 | -0/+59 |
| | | |||||
| * | New API for vector shuffles + additional shuffle operations | Marat Dukhan | 2017-04-14 | 1 | -21/+187 |
| | | |||||
| * | Blend optimizations for ARM NEON | Marat Dukhan | 2017-03-23 | 1 | -7/+35 |
| | | |||||
| * | Special implementation for ARM NEON | Marat Dukhan | 2017-03-23 | 1 | -17/+89 |
| | | |||||
| * | Make header compatible with recent Apple Clang | Marat Dukhan | 2017-03-22 | 1 | -2/+2 |
| | | |||||
| * | Include headers for [u]intN_t and size_t | Marat Dukhan | 2017-03-01 | 1 | -1/+5 |
| | | |||||
| * | F32 load/store for partial vector and/or stride | Marat Dukhan | 2017-02-26 | 1 | -0/+130 |
| | | |||||
| * | And-with-mask intrinsic | Marat Dukhan | 2017-02-22 | 1 | -0/+5 |
| | | |||||
| * | Signblend intrinsic | Marat Dukhan | 2017-02-22 | 1 | -0/+30 |
| | | |||||
| * | Swap intrinsic | Marat Dukhan | 2017-02-22 | 1 | -0/+50 |
| | | |||||
| * | Fix declaration of PSIMD_INTRINSIC | Marat Dukhan | 2017-02-22 | 1 | -2/+2 |
| | | |||||
| * | F32 reduce and all-reduce operations | Marat Dukhan | 2017-02-22 | 1 | -0/+66 |
| | | |||||
| * | Vector-pair data types and casts | Marat Dukhan | 2017-02-17 | 1 | -4/+64 |
| | | |||||
| * | Public release | Marat Dukhan | 2017-02-17 | 1 | -0/+451 |
 |
index : platform_external_psimd | |
| Unnamed repository; edit this file 'description' to name the repository. |
| aboutsummaryrefslogtreecommitdiffstats |