google/highway issues and pull requests

#1926 - f16 fixes: casts inside SVE ops

Pull Request - State: closed - Opened by copybara-service[bot] 10 months ago

#1925 - Bump github/codeql-action from 3.22.12 to 3.23.0

Pull Request - State: closed - Opened by dependabot[bot] 10 months ago
Labels: ready to pull, dependencies

#1924 - warning fixes (missing includes)

Pull Request - State: closed - Opened by copybara-service[bot] 10 months ago

#1923 - f16 warning fixes (cast); add missing includes

Pull Request - State: closed - Opened by copybara-service[bot] 10 months ago

#1922 - warning fixes (lint types, casts)

Pull Request - State: closed - Opened by copybara-service[bot] 10 months ago

#1921 - add missing DLLEXPORT, fixes #1918, thanks @dbermond

Pull Request - State: closed - Opened by copybara-service[bot] 10 months ago

#1920 - Added integer Div/Mod operations

Pull Request - State: closed - Opened by johnplatts 10 months ago - 3 comments
Labels: ready to pull

#1919 - Made fixes to SCALAR, EMU128, NEON, and PPC float to int conversions

Pull Request - State: closed - Opened by johnplatts 10 months ago
Labels: ready to pull

#1918 - Build failure in tests

Issue - State: closed - Opened by dbermond 10 months ago - 2 comments

#1917 - Bump github/codeql-action from 3.22.11 to 3.22.12

Pull Request - State: closed - Opened by dependabot[bot] 10 months ago
Labels: ready to pull, dependencies

#1916 - More f16 fixes. Revamp tests/Iota casts

Pull Request - State: closed - Opened by copybara-service[bot] 10 months ago

#1915 - fix implementation-defined cast

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1914 - add parallel flags to speed up multiarch builder

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1913 - Test #537 failure upon building Highway-1.0.4 with GCCcore-12.3.0 on x86_64

Issue - State: closed - Opened by monfortm 11 months ago - 7 comments

#1912 - Support for AVX_VNNI as an extension to the AVX2 target

Issue - State: closed - Opened by ag01 11 months ago - 3 comments

#1911 - Made fixes to PreventElision on PPC

Pull Request - State: closed - Opened by johnplatts 11 months ago
Labels: ready to pull

#1910 - f16/ppc build fixes: missing ConvertScalarTo template arg, copy

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1909 - F16/GCC build/test fixes:

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1908 - Fixed Clang compiler warning in TestSumsOfAdjQuadAbsDiff

Pull Request - State: closed - Opened by johnplatts 11 months ago
Labels: ready to pull

#1907 - GCC AVX2/AVX3 compilation fixes for F16 Load/Store

Pull Request - State: closed - Opened by johnplatts 11 months ago
Labels: ready to pull

#1906 - Added HWY_Z14 and HWY_Z15 targets

Pull Request - State: closed - Opened by johnplatts 11 months ago - 1 comment
Labels: ready to pull

#1905 - add Foreach and simplify Transform* to use LoadN

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1904 - warning fixes (cast/sign conversion)

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1903 - OrderedDemote2To() f64->f32 ?

Issue - State: open - Opened by Pflugshaupt 11 months ago - 3 comments

#1902 - warning fixes (cast)

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1901 - reduce non-test vqsort compile time

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1900 - Bump github/codeql-action from 2.22.10 to 3.22.11

Pull Request - State: closed - Opened by dependabot[bot] 11 months ago
Labels: ready to pull, dependencies

#1899 - Fix compilation errors with HWY_SCALAR target

Pull Request - State: closed - Opened by johnplatts 11 months ago
Labels: ready to pull

#1898 - Compilation error with GCC 10 and older 'call to non-constexpr function 'float hwy::F32FromF16'

Issue - State: closed - Opened by ag01 11 months ago - 1 comment

#1897 - Bump github/codeql-action from 2.22.8 to 2.22.10

Pull Request - State: closed - Opened by dependabot[bot] 11 months ago
Labels: ready to pull, dependencies

#1896 - Major revamp of vqsort Partition, fixes UB

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1895 - Adding support for IBM/Z14 and onwards

Issue - State: closed - Opened by seiko2plus 11 months ago - 5 comments

#1894 - Bump github/codeql-action from 2.22.8 to 2.22.9

Pull Request - State: closed - Opened by dependabot[bot] 11 months ago - 1 comment
Labels: dependencies

#1893 - Enabled truncating AlignedNDArrays without copying memory.

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1892 - Add top level permission to multiarch

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1891 - update gcc bf16 fix, wrong macro and missing ToByte. Fixes #1888, thanks @stefson

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1890 - fix arm8: sign conversion, use LimitsMax instead of ~0

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1889 - f16 build fix: ConvertStaticTo instead of static_cast

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1888 - compile is broken on aarch64 with gcc: hwy/ops/arm_sve-inl.h:59:7: error: no type named ‘type’ in ‘struct hwy::N_SVE::DFromV_t<__SVBfloat16_t>

Issue - State: closed - Opened by stefson 11 months ago - 6 comments

#1887 - Renamed hwy::AlignedNDArray::data_size to ::memory_size, to conform to ::memory_shape.

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1885 - replace x86 LoadDup128 usages with Dup128VecFromValues. Fixes #1870, thanks @ag01

Pull Request - State: closed - Opened by copybara-service[bot] 11 months ago

#1884 - Made enhancements to BitCastScalar and fixed F16 compare operators

Pull Request - State: closed - Opened by johnplatts 11 months ago - 3 comments
Labels: ready to pull

#1881 - `Error: unknown architectural extension 'sve2-aes'` when compiling on Graviton3

Issue - State: closed - Opened by fbarbari 11 months ago - 10 comments

#1873 - Bump github/codeql-action from 2.22.6 to 2.22.7

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: ready to pull, dependencies

#1872 - Bump step-security/harden-runner from 2.6.0 to 2.6.1

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: ready to pull, dependencies

#1871 - define HWCAP2_SVE2 and HWCAP2_SVEAES if not defined by system headers

Pull Request - State: open - Opened by copybara-service[bot] 12 months ago - 1 comment

#1870 - Potential sub-optimal AVX512 codegen for DupEven/DupOdd for 16-bit integer types

Issue - State: closed - Opened by ag01 12 months ago - 3 comments

#1869 - Lint fixes - include guard and includes

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1868 - Bump github/codeql-action from 2.22.5 to 2.22.6

Pull Request - State: closed - Opened by dependabot[bot] 12 months ago
Labels: ready to pull, dependencies

#1867 - fix matvec bf16 threshold, use smaller of MatT,VecT

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1866 - avoid UB from casting [b]f16 to u16, enable SVE f16

Pull Request - State: open - Opened by copybara-service[bot] 12 months ago

#1865 - add opt-out for bf16, move defines closer together, require clang18

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1864 - LLVM 18: error: always_inline function '_mm512_loadu_ps' requires target feature 'evex512', but would be inlined into function 'LoadU' that is compiled without support for 'evex512'

Issue - State: closed - Opened by LinuxUserGD 12 months ago - 4 comments

#1863 - inline sum

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1862 - enable clang16 Arm/PPC runtime dispatch if opted-in. Refs #1782

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1861 - avoid compile errors from HWY_IF outside hwy::HWY_NAMESPACE

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1860 - add recommendation that top-level functions avoid vector args

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1859 - GCC workaround for sign conversion in AddWithWraparound

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1858 - improve SVE PromoteUpperTo codegen (use native instruction)

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1857 - compare scalar and vector results for MulAddLoop

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1856 - Defined a masking model, and implemented a downscaling energy calculation.

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1855 - Added a copy assignment from initializer list to hwy::Span to simplify writing readable tests.

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1854 - undo rollback of "Added SaturatedAbs and SaturatedNeg operations"

Pull Request - State: closed - Opened by copybara-service[bot] 12 months ago

#1853 - Copybara import of the project:

Pull Request - State: closed - Opened by copybara-service[bot] almost 1 year ago - 1 comment

#1852 - add basic unit tests of bfloat16_t raw byte format

Pull Request - State: open - Opened by copybara-service[bot] almost 1 year ago

#1851 - add binary arithmetic ops for float16_t on x86

Pull Request - State: closed - Opened by copybara-service[bot] almost 1 year ago

#1850 - Made AlignedNDArray automatically pad the innermost dimension to make sure all innermost arrays are aligned.

Pull Request - State: closed - Opened by copybara-service[bot] almost 1 year ago

#1849 - Made AlignedNDArray automatically pad the innermost dimension to make sure all innermost arrays are aligned.

Pull Request - State: closed - Opened by copybara-service[bot] almost 1 year ago

#1848 - add MaskedMin/MaxOr

Pull Request - State: closed - Opened by copybara-service[bot] almost 1 year ago

#1847 - Added HWY_DASSERTs asserting buffer alignment to all Load and Store operations that require alignment.

Pull Request - State: open - Opened by copybara-service[bot] almost 1 year ago

#1846 - Made fixes to loops in hwy/contrib/algo/copy-inl.h, find-inl.h, and transform-inl.h

Pull Request - State: closed - Opened by johnplatts about 1 year ago - 2 comments
Labels: ready to pull

#1845 - Fixed compilation error in hwy/contrib/matvec/matvec-inl.h

Pull Request - State: closed - Opened by johnplatts about 1 year ago
Labels: ready to pull

#1844 - Added SaturatedAbs and SaturatedNeg operations

Pull Request - State: closed - Opened by johnplatts about 1 year ago - 4 comments
Labels: ready to pull

#1843 - Bump github/codeql-action from 2.22.4 to 2.22.5

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago
Labels: ready to pull, dependencies

#1842 - add MatVec overload that supports both mat/vec as bf16

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1841 - Bump github/codeql-action from 2.22.3 to 2.22.4

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago
Labels: ready to pull, dependencies

#1840 - fix unneeded trailing semicolon, let Clang use __bf16 on SSE2

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1839 - fix wording

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1838 - use +E notation for printing small floats

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1837 - update list of functions supporting bf16

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1836 - Bump actions/checkout from 3 to 4

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago
Labels: ready to pull, dependencies

#1835 - Created a Span class (inspired by std::span and absl::Span), and an AlignedArray class using it.

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1834 - fixes: int/uint cast, pointer overflow, f suffix, constexpr

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1833 - clang6 + lint fixes: cast + formatting

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1832 - fix typo

Pull Request - State: closed - Opened by GoodDaisy about 1 year ago - 7 comments

#1831 - further speed up ThreadPoolTest, add cast/change to size_t. Refs #1828

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1830 - Bump github/codeql-action from 2.22.2 to 2.22.3

Pull Request - State: closed - Opened by dependabot[bot] about 1 year ago
Labels: ready to pull, dependencies

#1829 - Remove GNU extension from compilation

Pull Request - State: closed - Opened by malaterre about 1 year ago
Labels: ready to pull

#1828 - Add testing on foreign archs

Pull Request - State: closed - Opened by malaterre about 1 year ago - 5 comments
Labels: ready to pull

#1827 - arm_sve-inl.h:2819:20: error: no matching function for call to 'IfThenElse(const svbool_t&, const __SVBfloat16_t&, const __SVBfloat16_t&)'

Issue - State: closed - Opened by malaterre about 1 year ago - 2 comments

#1826 - fix sign conversion(cast). Fixes #1824, thanks @malaterre

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1825 - thread_pool fixes: add to CMake, faster test

Pull Request - State: closed - Opened by copybara-service[bot] about 1 year ago

#1824 - arm_sve-inl.h:1530:73: error: unsigned conversion from 'int' to 'uint32_t' {aka 'unsigned int'} changes value from '-16777216' to '4278190080' [-Werror=sign-conversion]

Issue - State: closed - Opened by malaterre about 1 year ago - 2 comments

GitHub / google/highway issues and pull requests