Repository Issue Activity (beta)

#406 WarpMatchValue for f32/f64 uses truncating casts instead of to_bits()

#402 A `#[address_space(shared)]` static is stored with a generic `st.u32` instead of `st.shared`, corrupting memory — two identically-declared siblings lower differently

#401 `ctpop`/`ctlz`/`cttz` results are never cast back to the operand width — `u64::count_ones()` returns 0 on device

#399 Data Races & Flawed Synchronization due to Invalid Atomic Memory Scope in `SystemAtomicF32` and `SystemAtomicF64

#336 cutile-rs (cuTile Rust)? maybe?

#379 Is cuda_std still maintained?

#177 Add to godbolt

#387 Array parameters lose alignment metadata and ICE on >16 byte alignments

#226 Update rustc_llvm_wrapper to optionally support LLVM v19

#385 Support llvm-19 in containers

#197 CUDA 12.8.1 and LLVM 18.1.8

#378 cp.async ergonomics: missing address-space-aware intrinsic forces inline PTX

#376 fence_acqrel_block incorrectly uses system-level PTX scope (.sys) instead of block-level (.cta)

#219 Improve documentation

#141 Suggestion: code and commit guidelines

#364 Windows build vecadd sample fail: linking with `link.exe` failed: exit code: 1120

#204 Unable do Compile Rust-CUDA because missing CUDNN includes.

#361 Kernel arguments with `#[repr(align(16))]` cause misaligned access / `IllegalAddress`

#359 Incorrect WarpShuffleMode discriminants mapping to wrong shuffle instructions

#354 Invalid PTX generation for atomic load/store: incorrect scope suffix

#356 Inverted logic in warp_vote_any and warp_vote_all intrinsics

#218 Smart and user-controllable automatic Constant memory space placement for `static` variables

#320 Examples do not compile when release is disabled for gpu crate(s)

#262 Feature request: global "Fast Math" flag

#238 Port CUDA samples