All pull requests

Pull requests

• add LSX implementation for vqdmulhq_s16

  #1455·by jinboson was merged Sep 29, 2026Contributor·

  ·

• sparc float16: use a portable version to avoid compilation errors

  #1454·by brad0 was merged Sep 30, 2026Contributor·

  ·
  2

• build(deps): bump kjanat/actionlint from 1.16.1 to 1.17.0

  dependenciesgithub_actions
  #1453·by dependabot[bot] was merged Sep 24, 2026Bot·

  ·

• wasm: use vqtbl1q_s8 for swizzle on A64V8 for SIMD and relaxed SIMD

  #1450·by jawj was merged Sep 14, 2026Contributor·

  ·
  1

• build(deps): bump hendrikmuhs/ccache-action from 1.2.23 to 1.2.24

  dependenciesgithub_actions
  #1449·by dependabot[bot] was merged Sep 10, 2026Bot·

  ·
  1

• arm/neon: add LSX implementations

  #1447·by jinboson was merged Sep 16, 2026Contributor·

  ·
  8

• Enhance simde/x86/gfni.h

  #1429·by fo40225 was merged Aug 26, 2026Contributor·

  ·
  6

• math: use fpclassify for isinf on PowerPC to avoid QEMU xststdcsp bug

  #1421·by fo40225 was merged Aug 3, 2026Contributor·

  ·

• x86/sse: fix conditional compilation for Loongarch LSX

  #1420·by elysia-best was merged Aug 20, 2026Contributor·

  ·
  1

  4

• x86/avx512/dpbf16: avoid GCC's broken 512-bit masked intrinsics

  #1419·by fo40225 was merged Jul 28, 2026Contributor·

  ·
  2

  7

• x86/gfni: fix AltiVec paths on big-endian POWER

  #1418·by fo40225 was closed Aug 16, 2026Contributor·Draft·

  ·
  4

• x86/clmul: use VEX-encoded 256-bit VPCLMULQDQ on CPUs without AVX-512

  #1417·by fo40225 was merged Aug 5, 2026Contributor·

  ·

• x86: add cvtepu8_epi32 intrinsics

  #1416·by fo40225 was merged Jul 31, 2026Contributor·

  ·

• Fix POWER7 build: correct AltiVec version guards for POWER8-only intrinsics

  #1414·by fo40225 was merged Aug 5, 2026Contributor·

  ·
  3

• x86/avx512: Add implementations and tests for functions in cmpgt.h, cvt.h, extract.h, loadu.h, sllv.h

  #1402·by woobble was merged Apr 28, 2026Contributor·

  ·
  3

• math: extend non-glibc guard to Clang and add exp10 guard

  #1400·by kjg0724 was merged Mar 30, 2026Contributor·

  ·
  2

  5

• math: guard __builtin_roundeven against non-glibc platforms

  #1398·by kjg0724 was merged Mar 27, 2026Contributor·

  ·
  1

  5

• build(deps): bump EndBug/add-and-commit from 9 to 10

  dependenciesgithub_actions
  #1397·by dependabot[bot] was merged Mar 23, 2026Bot·

  ·

• x86 avx512: add more reduce functions

  #1395·by AdrianRiedl was merged Mar 13, 2026Contributor·

  ·
  7

• x86 avx: fix cvtt bounds

  #1393·by TheNumbat was merged Mar 4, 2026Contributor·

  ·
  3

• Add various SSE4.2 functions

  #1386·by mrexodia was closed Feb 13, 2026·

  ·
  1

  15

• New packaging script

  #1383·ahoarau opened Feb 3, 2026Contributor·

  ·
  2

• x86 f16: implement rounding support

  #1380·by TheNumbat was merged Feb 10, 2026Contributor·

  ·
  14

• test: Avoid -ffinite-math-only on floating point comparisons

  #1373·iv-m opened Jan 21, 2026Contributor·

  ·
  4

• CMake support

  #1372·by ahoarau was merged Jan 24, 2026Contributor·

  ·
  1

  4