| Commit message (Expand) | Author | Age | Files | Lines |
|---|
| * | ENH, SIMD: Implment intrinsic for mask division | Sayed Adel | 2023-01-29 | 1 | -0/+13 |
| * | ENH, SIMD: Implment intrinsic for FMA multiply add(odd) and subtract(even) | Sayed Adel | 2023-01-29 | 1 | -0/+4 |
| * | ENH, SIMD: Add special intrinsics for better non-contiguous/partial memory ac... | Sayed Adel | 2023-01-29 | 1 | -10/+307 |
| * | ENH: Implement intrinsics for shuffle over 128-bit lane and unzip | Sayed Adel | 2023-01-29 | 1 | -0/+152 |
| * | ENH, SIMD: Discard non-signaling comparison intrinsics | Sayed Adel | 2022-12-15 | 2 | -10/+1 |
| * | ENH, SIMD: Add ordered comparison intrinsics guarantees non-signaling | Sayed Adel | 2022-12-14 | 1 | -0/+10 |
| * | SIMD: Add new intrinsics to check true cross all vector lanes | Sayed Adel | 2022-09-19 | 2 | -40/+101 |
| * | ENH, SIMD: Implement new universal intrinsic for max/min operations | Sayed Adel | 2022-09-17 | 3 | -73/+172 |
| * | BUG, SIMD: Fix C++ AVX512/qsort on old gcc compilers | Sayed Adel | 2022-08-19 | 1 | -12/+12 |
| * | Merge pull request #20913 from seiko2plus/zsystem_sup | Matti Picus | 2022-06-12 | 1 | -0/+2 |
| |\ |
|
| | * | ENH, SIMD: Extend universal intrinsics to support IBMZ | Sayed Adel | 2022-06-04 | 1 | -0/+2 |
| * | | BUG: switch _CMP_NEQ_OQ to _CMP_NEQ_UQ for npyv_cmpneq_f[32,64] | Rafael CF Sousa | 2022-06-07 | 1 | -2/+2 |
| |/ |
|
| * | SIMD, ENH: Add universal intrinsic andc8 and use it to remove ifneq | Rafael CF Sousa | 2022-05-29 | 2 | -8/+11 |
| * | SIMD, ENH: Use logical bitwise to implement comparison functions (bool_) | Rafael CF Sousa | 2022-05-23 | 1 | -0/+12 |
| * | SIMD: Use universal intrinsics to implement comparison functions | Rafael Cardoso Fernandes Sousa | 2022-05-23 | 2 | -0/+54 |
| * | Make common/simd/avx512/memory.h compatible with C++ | serge-sans-paille | 2022-02-21 | 1 | -2/+4 |
| * | road-to-cxx: x86-qsort.dispatch.c.src -> x86-qsort.dispatch.cpp | serge-sans-paille | 2022-02-21 | 1 | -2/+2 |
| * | Merge pull request #20363 from seiko2plus/svml2npyv/tanh_f32 | Matti Picus | 2022-02-10 | 1 | -0/+29 |
| |\ |
|
| | * | SIMD: Add new universal intrinsics for lookup table | Sayed Adel | 2022-01-19 | 1 | -0/+29 |
| * | | Merge pull request #20133 from r-devulap/avx512-sort | Matti Picus | 2022-02-07 | 1 | -0/+72 |
| |\ \
| |/
|/| |
|
| | * | MAINT: rename reduce_min/max functions to not clash with local def in loops_m... | Raghuveer Devulapalli | 2022-02-02 | 1 | -12/+12 |
| | * | MAINT: Use cvt instruction instead of extract | Raghuveer Devulapalli | 2022-02-02 | 1 | -4/+4 |
| | * | SIMD: Add intrinsics emulation for reduce_min and reduce_max instructions | Raghuveer Devulapalli | 2022-02-02 | 1 | -0/+72 |
| * | | SIMD: add universal intrinsic for round to nearest | Sayed Adel | 2021-12-19 | 1 | -2/+6 |
| * | | ENH, SIMD: add new universal intrinsics for floor | HowJMay | 2021-12-12 | 1 | -0/+4 |
| |/ |
|
| * | ENH, SIMD: add new universal intrinsics for trunc | HowJMay | 2021-11-16 | 1 | -0/+4 |
| * | ENH: Implement SIMD for ceil | HowJMay | 2021-11-01 | 1 | -0/+4 |
| * | ENH: Add conversion for 512bit intrin to 256bit intrin | HowJMay | 2021-11-01 | 1 | -1/+21 |
| * | fix up | Sayed Adel | 2021-10-28 | 2 | -7/+7 |
| * | BUG, SIMD: Workaround broadcasting SIMD 64-bit integers on MSVC 32-bit mode | Sayed Adel | 2021-10-28 | 1 | -2/+30 |
| * | BUG, SIMD: Fix unexpected result of uint8 division on X86 | Sayed Adel | 2021-05-19 | 1 | -9/+10 |
| * | DOC: Add comment for ifdef macro guard | HowJmay | 2021-05-16 | 1 | -1/+1 |
| * | ENH: Add max/min value comparison for integers | HowJmay | 2021-05-10 | 1 | -0/+32 |
| * | ENH: Add min values comparison for floating point | HowJmay | 2021-05-02 | 1 | -0/+17 |
| * | Merge pull request #18863 from HowJMay/simd-max | Sayed Adel | 2021-04-30 | 1 | -0/+17 |
| |\ |
|
| | * | ENH: Add max values comparison for floating point | HowJmay | 2021-05-01 | 1 | -0/+17 |
| * | | fix: Fix the typo "implment" | HowJmay | 2021-04-28 | 1 | -3/+3 |
| |/ |
|
| * | SIMD: add NPYV fast integer division intrinsics for AVX512 | Sayed Adel | 2021-03-08 | 1 | -2/+222 |
| * | use more accurate words. | Qiyu8 | 2021-01-27 | 1 | -1/+1 |
| * | improve sumup intriniscs. | Qiyu8 | 2021-01-25 | 1 | -18/+18 |
| * | treat u8/u16 as overflow protection. | Qiyu8 | 2021-01-22 | 1 | -17/+24 |
| * | Optimize the avx512 sum_u8/sum_u16. | Qiyu8 | 2021-01-21 | 1 | -24/+28 |
| * | Add new intrinsics sum_u8/u16/u64. | Qiyu8 | 2021-01-20 | 1 | -0/+25 |
| * | BUG, MAINT: improve avx512 mask logical operations | Sayed Adel | 2021-01-05 | 2 | -47/+55 |
| * | Merge branch 'master' of github.com:numpy/numpy into countnz | Qiyu8 | 2020-12-23 | 4 | -0/+127 |
| |\ |
|
| | * | ENH, SIMD: Add new NPYV intrinsics pack(1) | Sayed Adel | 2020-12-22 | 4 | -0/+127 |
| * | | use max value to detect deviation | Qiyu8 | 2020-12-14 | 1 | -1/+1 |
| * | | remove extra input | Qiyu8 | 2020-12-10 | 1 | -2/+2 |
| * | | use current intrinsics instead of create new one. | Qiyu8 | 2020-12-10 | 2 | -17/+9 |
| * | | Add non-AVX512DQ intrinsics | Qiyu8 | 2020-12-10 | 1 | -0/+8 |
