All pull requests

Pull requests

• Allow CEnum 1.0

  #3397·maleadt opened Oct 15, 2026Member·

  ·

• Poll an event capturing the context's work when synchronizing the device

  performance
  #3396·maleadt opened Oct 15, 2026Member·

  ·
  1

• cuFFT: Support FFTs on non-contiguous CuArray views via buffering (Approach 1)

  #3395·huiyuxie opened Oct 15, 2026Contributor·

  ·
  1

  5

• Compute indices in 32 bits when launch-checked facts allow it

  #3389·maleadt opened Oct 13, 2026Member·

  ·
  2

• [release-6.2] Port to LLVM.jl 10 (GPUCompiler 1.23.1), release 6.2.3

  #3381·vchuravy opened Oct 11, 2026Member·

  ·
  1

• KA: round prefer_blocks workgroup sizes to warp multiples

  #3373·vchuravy opened Oct 10, 2026Member·

  ·
  2

  

• Specialize kernels on how their array arguments alias

  #3372·vchuravy opened Oct 9, 2026Member·

  ·

• Enzyme: forward rule for CuArray(::DataRef) must wrap the shadow DataRef

  #3371·vchuravy opened Oct 9, 2026Member·

  ·
  

• Slim down when we run core and how much of it is run

  #3358·kshyatt opened Oct 7, 2026Member·

  ·
  1

• Keep kernel launches type-stable with index-typed device arrays

  #3294·maleadt opened Sep 24, 2026Member·4/4·

  ·
  2

• Add an index type parameter to CuDeviceArray

  #3293·maleadt opened Sep 24, 2026Member·3/4·

  ·
  2

• Add hostcalls for device-to-host services

  #3243·maleadt opened Aug 20, 2026Member·

  ·
  2

• Try enabling fwd broadcating Enzyme test

  #3214·kshyatt opened Jul 24, 2026Member·

  ·
  5

• Implement unsafe_wrap! for in-place re-wrapping of non-owning CuArrays

  #3194·ameligrana opened Jul 12, 2026·

  ·
  1

• Remove some allocations from unsafe_wrap(::Type{CuArray{T,N,M}}, p::P…

  #3193·ameligrana opened Jul 12, 2026·

  ·
  3

• Expose directed rounding for Float64 WMMA tensor cores

  #3143·orkolorko opened May 16, 2026Contributor·4 tasks·

  ·
  2

• Use Republic.jl to forward subpackage APIs

  #3103·AntonOresten opened Apr 15, 2026Contributor·

  ·
  2

• CUPTI Profiler Host, Range Profiler, and PM Sampling APIs

  #3059·gbaraldi opened Mar 23, 2026Member·1 of 2 tasks·

  ·
  3

• Add direct sparse gpu constructors for COO, CSC, CSR and BSR matrices

  #3022·nHackel opened Jan 23, 2026·

  ·
  1

• Add AnyCuDeviceArray variations and CuScalar

  cuda arrayspeculative
  #2849·moukle opened Aug 22, 2025·

  ·
  4

• fixes the kron implementation for sparse + diagonal matrix

  #2804·tam724 opened Jun 27, 2025Contributor·

  ·
  23

• add fastmath flag

  #2732·vchuravy opened Apr 9, 2025Member·

  ·
  2

• Try fast linear indexes for KA

  enhancementneeds changesperformance
  #2612·vchuravy opened Jan 9, 2025Member·

  ·
  8

• Allow disabling the linking of libdevice in CUDACompilerParams

  enhancementneeds changesspeculative
  #2611·gbaraldi opened Jan 8, 2025Member·

  ·
  7

• make CUDA randn work with Zygote

  enhancementneeds changes
  #2581·bgctw opened Dec 9, 2024·

  ·
  5