Skip to content

Actions: NVIDIA/cutlass

Actions

Auto Label Issues

Actions

Loading...
Loading

Show workflow options

Create status badge

Loading
auto-label-issues.yml
503 workflow runs
503 workflow runs

Filter by Event

⌥ + click/return to exclude

Filter by Status

⌥ + click/return to exclude

Filter by Branch

⌥ + click/return to exclude

Filter by Actor

⌥ + click/return to exclude
[BUG] CuTe DSL: SIGSEGV compiling integer ** (int32/int64), including 2 ** 3 Auto Label Issues #569: Issue #3735 opened by JayceSu98
9s
9s
[BUG] cute.ceil_div throws internal error with int64 input Auto Label Issues #568: Issue #3734 opened by kainzhong
7s
7s
[BUG] sm120 GEMM hangs when K = 0 (grouped and non-grouped) after can_implement(), initialize() and run() return success Auto Label Issues #567: Issue #3733 opened by vera-rubin
7s
7s
[QST] SM120 FP8 MMA with FP32 accumulate, plain form against block-scaled form with unit scales Auto Label Issues #566: Issue #3732 opened by emre570
7s
7s
[BUG] Cache policy from cute.copy is dropped when the copy atom is created with invariant=True Auto Label Issues #565: Issue #3726 opened by kainzhong
10s
10s
[BUG] SM90 TMA GEMM mainloops rely on elect_one_sync() electing lane 0 (TMA issuer vs. PipelineTmaAsync is_leader) Auto Label Issues #564: Issue #3722 opened by builtin-pb
9s
9s
[BUG] SM90 cp.async + GMMA collectives: no fence.proxy.async between cp.async operand writes and wgmma reads Auto Label Issues #563: Issue #3721 opened by builtin-pb
9s
9s
[BUG] Grouped GEMM and grouped Rank2K kernels build operand iterators from a temporary Params Auto Label Issues #562: Issue #3718 opened by BwL1289
13s
13s
[BUG] CuTe DSL from_dlpack rejects float8_e4m3fn tensors (dtype_map lists Float8E5M2 only) Auto Label Issues #561: Issue #3716 opened by Vinnie6167
11s
11s
[BUG] CuTe DSL >=4.6.0: export_to_c does not apply explicit preferred_smem_carveout Auto Label Issues #560: Issue #3715 opened by naytiba
9s
9s
[BUG] [BUG] cute.arch.store doesn't work with fp8 vector Auto Label Issues #559: Issue #3712 opened by gau-nernst
9s
9s
[BUG] CuTe DSL CopyUniversalOp ignores l1c_evict_priority again (regression of #3199 since 4.6.0, loads and stores) Auto Label Issues #558: Issue #3711 opened by gau-nernst
9s
9s
[BUG] CuTe DSL >= 4.6: FlashAttention-4 SM100 backward with block-sparse masks spills registers, up to 1.8x slower than 4.5.3 (same kernel source) Auto Label Issues #557: Issue #3707 opened by janhuenermann
11s
11s
[BUG] TVM-FFI exported functions hang under multithreads Auto Label Issues #556: Issue #3706 opened by kainzhong
2m 28s
2m 28s
[BUG] CuTe DSL 4.7/4.8 regression: FlashAttention-4 SM100 forward kernel spills registers (0 -> 176 B) vs 4.5.1, ~3% slower on B300 Auto Label Issues #555: Issue #3701 opened by zpcore
10s
10s
[BUG] CuTeDSL 4.8.0 nvpu-optimization fails without a diagnostic on a sum of 10 dynamic extents Auto Label Issues #554: Issue #3699 opened by allispaul
12s
12s
[BUG][CuTe DSL] AOT TVM FFI loader does not marshal dataclass arguments Auto Label Issues #553: Issue #3698 opened by gdlg-pplx
6s
6s
[BUG] mma_atom_call for mma.sync is eliminated if C is RZ Auto Label Issues #552: Issue #3689 opened by melonedo
10s
10s
[BUG] CuTe DSL: a % b gives different results for dynamic and constant operands when the signs differ Auto Label Issues #551: Issue #3687 opened by Arthur031221
10s
10s
[BUG] CuTe DSL: cute.compile stalls for minutes when no GPU is visible Auto Label Issues #550: Issue #3681 opened by hwanseoc
6s
6s
[BUG] CuTe DSL: export_to_c fails for a --gpu-arch other than the current GPU's Auto Label Issues #549: Issue #3680 opened by hwanseoc
9s
9s
[BUG] Order-dependent wrong rmem values on Windows SM120 with CuTe DSL 4.8 CUDA12 stack; CUDA13 control passes Auto Label Issues #548: Issue #3673 opened by hisashi2026
7s
7s
[FEA][CuTeDSL] Let make_tiled_tma_atom set the TMA L2 promotion Auto Label Issues #547: Issue #3672 opened by zkyue
8s
8s
[BUG][CuTeDSL] cute.copy(..., tma_desc_ptr=p) with a gmem-space p passes the Python checks, then fails in libNVVM Auto Label Issues #546: Issue #3671 opened by zkyue
6s
6s
[BUG][CuTeDSL] cute.copy(..., tma_desc_ptr=TensorMap.get_ptr()) is silently ignored on tiled TMA loads Auto Label Issues #545: Issue #3670 opened by zkyue
8s
8s