All issues

Search results

[BUG] CuTe DSL: FloatNV8E5M3FNU.to(Float32) returns inf for the entire top exponent binade (codes 248-254)

? - Needs Triage

bugSomething isn't workingSomething isn't working

CuTe DSL

Status: Open.

#3602 In NVIDIA/cutlass;

· tdophung opened on Sep 8, 2026

[BUG] Example 93 GQA initialization overwrites active remote contributions

Status: Open.

#3598 In NVIDIA/cutlass;

· 1sgtpepper opened on Sep 8, 2026

[BUG] Example 93 GQA reduction races when reusing shared scratch

Status: Open.

#3597 In NVIDIA/cutlass;

· 1sgtpepper opened on Sep 8, 2026

[BUG] SM90 blockwise FP8 GEMM reuses modified B scales across M waves

Status: Open.

#3596 In NVIDIA/cutlass;

· 1sgtpepper opened on Sep 8, 2026

[BUG] CuTe DSL 4.8 dev SIGABRTs in TVM-FFI launch for SM100 ragged SDPA (4.7 passes)

Status: Open.

#3595 In NVIDIA/cutlass;

· YangXu1990uiuc opened on Sep 8, 2026

[BUG] CuTe DSL 4.8 dev regresses SM80 SDPA backward compilation (NVVM constraint 'n')

? - Needs Triage

bugSomething isn't workingSomething isn't working

CuTe DSL

Status: Open.

#3594 In NVIDIA/cutlass;

· YangXu1990uiuc opened on Sep 8, 2026

[BUG] SM100/SM103 ptr-array GEMM kernels size the per-SM tensormap workspace from the raw hw_info.sm_count, so a default KernelHardwareInfo causes out-of-bounds descriptor writes

CUTLASS C++

Status: Open.

#3585 In NVIDIA/cutlass;

· elbourne12345 opened on Sep 4, 2026

[BUG] nvcc 13.3 sm_120 local memory STL.128 unaligned

? - Needs Triage

bugSomething isn't workingSomething isn't working

CUTLASS C++

Status: Open.

#3581 In NVIDIA/cutlass;

· random25160765-collab opened on Sep 3, 2026

[BUG] CUDA_CTA_RECONFIG_ACTIVATED is not enabled for SM110a (Jetson AGX Thor)

? - Needs Triage

bugSomething isn't workingSomething isn't working

CuTe DSL

Status: Open.

#3576 In NVIDIA/cutlass;

· ptiris opened on Sep 1, 2026

[BUG] CuTeDSL >= 4.6 cannot compile NCCL GIN device ops — ptxas rejects the generated PTX (4.5.2 works)

CuTe DSL

Status: Open.

#3573 In NVIDIA/cutlass;

· kiwi3shark opened on Aug 31, 2026

Blockscaled GEMV accumulates past K when floor(K/tile) is not a multiple of the stage count: example 91 fails for batch > 1

CUTLASS C++

Status: Open.

#3568 In NVIDIA/cutlass;

· TANGBUDU opened on Aug 29, 2026

[DOC] Request: Error handling documentation for CuTe DSL exported C API

? - Needs Triage

CuTe DSL

documentationDocumentationDocumentation

Status: Open.

#3562 In NVIDIA/cutlass;

· UnpureRationalist opened on Aug 28, 2026