All pull requests

Pull requests

• [STF] Waive locality domain tests if the driver is too old

  stf
  #11837·caugonnet opened Oct 3, 2026Contributor·2 tasks done·

  ·
  9

  

• [stdpar] Add none_of NVHPC smoke test

  #11836·iamAbhi-916 opened Oct 3, 2026·

  ·
  2

• Treat -0.0 and +0.0 as equal in DeviceBatchedTopK ranking

  #11835·kita-renji opened Oct 3, 2026·2 tasks done·

  ·
  1

  4

• [cudax] execution: add lane_scheduler, a "stream first" scheduler with lazy cross-stream joins

  #11833·caugonnet opened Oct 3, 2026Contributor·2 tasks done·

  ·
  2

• Prepare cuda/std headers before applying misc-include-cleaner check

  #11832·fbusato opened Oct 3, 2026Contributor·

  ·
  4

  

• [STF] Error sinks B: selectors, properties, and status-returning actions

  #11830·andralex opened Oct 3, 2026Contributor·2 tasks done·

  ·
  7

• Small fixes to tuning infrastructure documentation

  #11828·bernhardmgruber opened Oct 2, 2026Contributor·

  ·
  1

  7

• Tune the sm100 lookahead scan for int64

  #11826·bernhardmgruber opened Oct 2, 2026Contributor·

  ·
  1

• [libcu++] cuda_error: carry the exact status, its domain, and the source location

  #11825·andralex opened Oct 2, 2026Contributor·

  ·
  8

• [c.parallel]: Remove static assert preventing compiling c.parallel with nvcc 13.4

  #11823·NaderAlAwar opened Oct 2, 2026Contributor·

  ·
  3

• Add cub::detail::DeviceSetOps SetUnion, SetIntersection, and SetSymmetricDifference (keys)

  #11822·bernhardmgruber opened Oct 2, 2026Contributor·

  ·
  3

• [STF] Error sinks A: status traits and the status operand

  #11820·andralex opened Oct 2, 2026Contributor·2 tasks done·

  ·
  34

• [libcu++] Implement std::once_flag

  #11819·miscco opened Oct 2, 2026Contributor·

  ·
  5

• Add cub::DeviceReduce::ReduceNonCommutative

  #11818·Wint3rNight opened Oct 2, 2026Contributor·

  ·
  2

  5

• Fix thrust::reduce_by_key with __half and __nv_bfloat16 values

  #11817·bernhardmgruber opened Oct 2, 2026Contributor·

  ·
  1

  4

• Fix bin boundaries in i8/i16 histogram benchmarks

  #11815·gonidelis opened Oct 2, 2026Member·

  ·
  1

  6

• [CUB] Add adaptive implementation of segmented reduce

  #11813·Pennycook opened Oct 2, 2026·1 of 2 tasks·

  ·
  9

• [CUB] Modularize cub util_type

  #11812·miscco opened Oct 2, 2026Contributor·

  ·
  7

• [CUB] Migrate two-phase scan-by-key APIs to execution environments

  #11811·Seanium opened Oct 2, 2026·2 tasks done·

  ·
  1

  7

• Restore fast path for signed-byte range histograms

  #11810·gonidelis opened Oct 2, 2026Member·

  ·
  1

  14

• Retune sm107 lookahead scan tuning for i128 inputs

  #11808·gonidelis opened Oct 2, 2026Member·

  ·
  4

• [libcu++] Replace unused with [[maybe_unused]]

  #11805·miscco opened Oct 1, 2026Contributor·

  ·
  5

• Move the bulk store of the warpspeed scan into a dedicated store squad

  #11804·bernhardmgruber opened Oct 1, 2026Contributor·

  ·
  13

• [libcu++] Partially support constant_wrapper with msvc

  #11801·davebayer opened Oct 1, 2026Contributor·

  ·
  9

• [libcu++] Logical endpoint import export multiprocess test

  #11800·pciolkosz opened Oct 1, 2026Contributor·

  ·
  2