- 2607d3b Windows topology fix for AMD ThreadRipper by Jan Wassenberg · 2 hours ago upstream/master
- f8719ef Pre-release Windows fixes: missing capture, f64->u64 cast by Jan Wassenberg · 7 hours ago
- 4ae1990 1.4.0 release candidate by Jan Wassenberg · 27 hours ago
- 89e7b28 MSVC DLLEXPORT fixes for skeleton_test/thread_pool, fixes #2798 by Jan Wassenberg · 28 hours ago
- 8f3011a Fix incorrect macro layering, move sanitizer macros to detect_compiler_arch by Jan Wassenberg · 29 hours ago
- 9f73ae5 RVV fix for count_value_test + use concrete types by Jan Wassenberg · 34 hours ago
- 4605f16 RVV fix for Reorder/OrderedDemote2To: GCC 15 scalar codegen bug by Jan Wassenberg · 2 days ago
- d60946e Add printing of columns headers in profiler by Krzysztof Rymski · 2 days ago
- c34a36f Merge pull request #2994 from KimBioInfoStudio:freebsd-futex-support by Copybara-Service · 5 days ago
- b0b7c23 Add native futex support for FreeBSD via _umtx_op by Kim Yang · 6 days ago
- 294297a use large(r) test runners by Jan Wassenberg · 6 days ago
- 1590e08 Pre-release test fixes+improvements by Jan Wassenberg · 6 days ago
- 619d519 print bazel version before build by Jan Wassenberg · 6 days ago
- 5dbd0de Merge pull request #2990 from google:dependabot/github_actions/step-security/harden-runner-2.18.0 by Copybara-Service · 7 days ago
- 10c73dd Fix GCC12 build: HWY_IF_CONSTEXPR. Also add FillBytes by Jan Wassenberg · 7 days ago
- 8d797c1 fix include path, fixes #2985, thanks @stefson by Jan Wassenberg · 7 days ago
- b87a875 Bump step-security/harden-runner from 2.17.0 to 2.18.0 by dependabot[bot] · 7 days ago
- a02e5fa minor cleanup (license header, includes, comments) by Jan Wassenberg · 8 days ago
- 069f7e1 Add simple and advanced array sum tutorials to Highway by Krzysztof Rymski · 8 days ago
- a158161 Merge pull request #2982 from google:dependabot/github_actions/actions/cache-5.0.5 by Copybara-Service · 8 days ago
- cef904b Bump actions/cache from 5.0.4 to 5.0.5 by dependabot[bot] · 8 days ago
- a9daa4e Merge pull request #2981 from bkmgit:patch-1 by Copybara-Service · 8 days ago
- a27624e Add information about NumKong SIMD accelerated math library by Benson Muite · 8 days ago
- 0b05d29 Optimize FastLog10 by preabsorbing the constant into FastLog coeffeceint which essentially saves 1 Mul instruction by Nikhil Dev Goyal · 9 days ago
- 154cc0a Optimize FastLog2 by removing the redundant Mul() instruction at the end by preabsorbing multiplication of kInvLn2 directly into the implementation of FastLog by Nikhil Dev Goyal · 9 days ago
- 664837a Merge pull request #2966 from RaviTriv:count/CountIf by Copybara-Service · 9 days ago
- 1b1a6a4 add runtime check for SVE by Ravi · 9 days ago
- 3d46456 move k1 inside if by Ravi · 9 days ago
- 54ef9d7 update CountIf by Ravi · 9 days ago
- 1e86a3f widen before to prevent overflow by Ravi · 9 days ago
- 79a8d47 update formatting by Ravi · 9 days ago
- 07e0b98 add lane guard by Ravi · 9 days ago
- 35b719c refactor for masked accumulation by Ravi · 9 days ago
- e95a89a update build files by Ravi · 9 days ago
- e689d4b use lamba for CountIf by Ravi · 9 days ago
- b025d7b implement count by Ravi · 9 days ago
- ea8672c Fix f164ebb - GCC requires + before target attr by Jan Wassenberg · 9 days ago
- 7deaa38 Merge pull request #2975 from google:dependabot/github_actions/step-security/harden-runner-2.17.0 by Copybara-Service · 9 days ago
- f164ebb SVE2_128 requires I8MM/BF16. Fixes #2973 by Jan Wassenberg · 12 days ago
- 0094585 Bump step-security/harden-runner from 2.16.1 to 2.17.0 by dependabot[bot] · 12 days ago
- 9c734ef Fix SVE HWY_NATIVE_DOT_BF16 - used before defined by Jan Wassenberg · 13 days ago
- 3b9166f Attempted workaround for GCC-15 RVV vnclipu mis-optimization by Jan Wassenberg · 13 days ago
- 4dc9284 Fix GTest target name display for older GTest by Jan Wassenberg · 2 weeks ago
- 22fc355 Merge pull request #2964 from mohammadmseet-hue:fix/integer-overflow-checks by Copybara-Service · 2 weeks ago
- 9970542 Release testing updates by Jan Wassenberg · 2 weeks ago
- e525cbc Merge pull request #2963 from kleisauke:add-missing-header by Copybara-Service · 2 weeks ago
- bac2b25 Merge pull request #2968 from JamieMagee:enable-rvv-runtime-dispatch by Copybara-Service · 2 weeks ago
- 068d596 Merge pull request #2961 from google:dependabot/github_actions/step-security/harden-runner-2.16.1 by Copybara-Service · 2 weeks ago
- 4a43b15 Enable RVV runtime dispatch for Clang 19+ by Jamie Magee · 2 weeks ago
- bdf3cd9 Fix integer overflow and missing bounds checks in AlignedNDArray and ImageBase by mohammadmseet-hue · 3 weeks ago
- a430909 Pre-release fixes: x86 cross compiler, UNUSED by Jan Wassenberg · 3 weeks ago
- 9c8f74e Add missing minmax-inl.h header to CMake and Meson build files by Kleis Auke Wolthuizen · 3 weeks ago
- 6d2072e Bump step-security/harden-runner from 2.16.0 to 2.16.1 by dependabot[bot] · 3 weeks ago
- 7e2773c clang aarch64 OOM workaround: disable some of NEON/SVE targets by Jan Wassenberg · 3 weeks ago
- fde5200 fix clangd warning (extra overload), fixes #2957. Also update op_wishlist by Jan Wassenberg · 3 weeks ago
- eeebfb8 Merge pull request #2950 from RaviTriv:minmax by Copybara-Service · 3 weeks ago
- a1ac79c leverage if statement to prevent over/under flow errors by Ravi · 3 weeks ago
- 3fc866e 4x unroll by Ravi · 3 weeks ago
- 972d1ed update .gni by Ravi · 3 weeks ago
- 9d29a1e update wishlist by Ravi · 3 weeks ago
- 4fb15b2 update year by Ravi · 3 weeks ago
- 8b2cf53 Re-enable SVE targets. Fixes #2908 by Jan Wassenberg · 3 weeks ago
- 2c04b8e Add all remaining functions of FastMath in math_benchmark for completeness by Nikhil Dev Goyal · 3 weeks ago
- 4b97a0e Add template 'kHandleSubnormals' to FastExp and FastExp2 by Nikhil Dev Goyal · 3 weeks ago
- 374767c Fixing a bug in 'TestMath' which didnt handle the case correctly when both min and max were negative leading to no tests running for that case and ending with the max ulp error 0 if both min and max were passed negative. by Nikhil Dev Goyal · 3 weeks ago
- 888be30 Optimize Log() in math-inl.h by Nikhil Dev Goyal · 3 weeks ago
- 6bd7760 Optimize FastTanh() by Nikhil Dev Goyal · 3 weeks ago
- 6f0c329 Optimize FastLog() by Nikhil Dev Goyal · 3 weeks ago
- ed26e26 Split compare_test due to SVE OOM. Refs #2908 by Jan Wassenberg · 3 weeks ago
- c943faa update bazel build by Ravi · 3 weeks ago
- 5dc0fd3 add minmax algo by Ravi · 3 weeks ago
- 40d5812 Correct copyright year by Nikhil Dev Goyal · 4 weeks ago
- 3d8b9be Add dynamic function selection for benchmarking via command line flags by Nikhil Dev Goyal · 4 weeks ago
- 83804a4 Merge pull request #2945 from LXYan2333:master by Copybara-Service · 4 weeks ago
- 76a6701 Add `HWY_ATTR` requirements of lambda to reference by LXYan2333 · 4 weeks ago
- 9ce2d10 Add math_benchmark test file. by Nikhil Dev Goyal · 4 weeks ago
- d8642f6 Optimize FastLog() by Nikhil Dev Goyal · 4 weeks ago
- 7a1336c Tighten Type safety by Nikhil Dev Goyal · 4 weeks ago
- c827df3 Use piecewise cubic instead of piecewise pade1,1 in FastTanh by Nikhil Dev Goyal · 4 weeks ago
- a90ab8e Tighten Type safety in math-inl.h by Nikhil Dev Goyal · 4 weeks ago
- 71d8358 Add FastExp2 by Nikhil Dev Goyal · 4 weeks ago
- 8127819 Merge pull request #2936 from bkmgit:threads-header-install by Copybara-Service · 4 weeks ago
- 4ab9340 fix gcc-12 build: skip 64-bit Lookup8 for 128-bit vectors by Jan Wassenberg · 4 weeks ago
- 048283d bit_set.h header is needed when building with threads by Benson Muite · 4 weeks ago
- f93da7d Fix InterleaveUpperBlocks for SVE. Refs #2908 by Jan Wassenberg · 4 weeks ago
- 158be64 faster math_test (timeout on Arm) by Jan Wassenberg · 4 weeks ago
- 9b848fa Remove redundant debug assertion by Nikhil Dev Goyal · 4 weeks ago
- e3f9e53 Reformat by Nikhil Dev Goyal · 4 weeks ago
- 39f27a3 Use Lookup8 in FastTan, FastLog, FastTanh. by Nikhil Dev Goyal · 4 weeks ago
- a19ae4e Extend Lookup8 to >= 16 bit types by Jan Wassenberg · 4 weeks ago
- 05230e7 Add Lookup8 by Jan Wassenberg · 4 weeks ago
- 4c1444d update TwoTablesLookupLanes doc with guidance for partial vectors by Jan Wassenberg · 4 weeks ago
- fb96551 Merge pull request #2927 from bkmgit:patch-1 by Copybara-Service · 4 weeks ago
- 86cd321 Mention xSimd as an alternate library by Benson Muite · 4 weeks ago
- 0cdac86 disable SVE2 for vqsort, but not SVE2_128 by Jan Wassenberg · 5 weeks ago
- dff9d8f Fix Log1p Division Underflow on ARMv7 NEON by Nikhil Dev Goyal · 5 weeks ago
- 314f4b0 Eliminate Dead Multiply in polynomial calculation by Nikhil Dev Goyal · 5 weeks ago
- 71a3d05 Optimize FastTan, FastTanh, FastLog with paralell blend chain path for registers >= 32 by Nikhil Dev Goyal · 5 weeks ago
- c631cbc Warning fix: signed vs unsigned by Jan Wassenberg · 5 weeks ago
- 0c8a170 Refactor FastTan, FastTanh, and FastLog rational approximations. by Nikhil Dev Goyal · 5 weeks ago