All pull requests

Pull requests

• fix(cuda_std): stop dynamic_shared_mem overlapping static shared memory

  #417·shubhamkapoor01 opened Oct 25, 2026·3 tasks done·

  ·
  1

• CI: run cust's tests on a simulated GPU

  #415·saqibkh opened Oct 2, 2026·

  ·

• feat(cudnn): add batch normalization forward/backward

  #405·Rachit2323 opened Jul 28, 2026·4 tasks done·

  ·

• fix(nvvm): keep address-space-qualified pointers out of the generic space

  #404·nh13 opened Jul 22, 2026Contributor·

  ·

• fix(cuda_std): pin glam to "0.30" to match the rest of the workspace

  #398·samlaf opened Jun 27, 2026·

  ·
  1

• docs(getting_started): standardize on CUDA thread/kernel terminology

  #397·samlaf opened Jun 26, 2026·

  ·

• feat(cufft): cuFFT integration

  #395·CaseySanchez opened Jun 5, 2026·4 tasks done·

  ·

• feat(godbolt): add Compiler Explorer integration for rust-cuda kernels

  #394·g-ampo opened May 6, 2026·1 of 3 tasks·

  ·
  1

  22

• fix: KernelInvocation::from_raw uses grid dimensions for block dimensions

  #393·39ali opened May 4, 2026Contributor·1 of 3 tasks·

  ·

• fix ArrayFormat naming and mapping

  #392·39ali opened May 4, 2026Contributor·1 of 3 tasks·

  ·

• fix(abi): preserve array alignment and apply 16-byte workaround generically

  #388·Snehal-Reddy opened May 1, 2026Contributor·3 tasks done·

  ·
  1

  2

• add async copy opcodes

  #384·39ali opened Apr 27, 2026Contributor·1 of 3 tasks·

  ·

• fix warp assert on pow of 2

  #382·39ali opened Apr 26, 2026Contributor·1 of 3 tasks·

  ·
  2

• SystemAtomic scope should be system not device

  #381·39ali opened Apr 26, 2026Contributor·1 of 3 tasks·

  ·

• atomicInc/atomicDec with bounds

  #380·39ali opened Apr 26, 2026Contributor·1 of 3 tasks·

  ·

• fix(build): add CUDA 13.2 support

  #368·CharryWu opened Mar 26, 2026Contributor·1 of 2 tasks·

  ·
  4

• User controllable and automatic constant memory placement

  #363·PMQ9 opened Feb 19, 2026Contributor·

  ·

• Allow adjusting loop unroll aggressiveness

  #350·niklebedenko opened Jan 14, 2026Contributor·

  ·
  4

• examples: Add dot product of two vectors example

  #345·karpenkoag opened Dec 22, 2025·

  ·
  1

• feat: Add global fast_math flag to CudaBuilder

  #331·koreaygj opened Dec 1, 2025·

  ·
  9

• WIP 'opaque' pointers, no atomics yet

  #323·FractalFir opened Nov 26, 2025Collaborator·

  ·
  3

• Fix non-release kernel builds via CudaBuilder

  #322·LegNeato opened Nov 26, 2025Contributor·

  ·
  1

  7

• Update to nightly-2025-08-11.

  #291·nnethercote opened Oct 17, 2025Collaborator·

  ·
  8

• Add modal CI CUDA image

  #285·LegNeato opened Oct 1, 2025Contributor·

  ·

• Fixed a bug where self-referential types would cause infinite recursion and stack overflows

  #277·FractalFir opened Sep 21, 2025Collaborator·

  ·
  3