Skip to content

Actions: NVIDIA/cutlass

Actions

Auto Label Issues

Actions

Loading...
Loading

Show workflow options

Create status badge

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

Filter by Event

Filter by Status

Filter by Branch

Filter by Actor

[BUG] Cute print_tensor does not agree with CUTLASS for e2m1 Auto Label Issues #162: Issue #2764 opened by tapicl
8s
8s
[QST] CuTeDSL should drop cuda.stream when creating mangle name. Auto Label Issues #161: Issue #2763 opened by PannenetsF
8s
8s
[BUG] ModuleNotFoundError: No module named 'cutlass' Auto Label Issues #160: Issue #2762 opened by whatdhack
8s
8s
[BUG] Race condition causing correctness issue of SM90 array/grouped gemms when PDL enabled Auto Label Issues #159: Issue #2760 opened by Algy
7s
7s
[QST] Build cutlass-dsl Python package from source Auto Label Issues #158: Issue #2759 opened by Flamefire
9s
9s
[QST] Will Cute DSL be compatible with NVSHMEM in the future? Auto Label Issues #157: Issue #2757 opened by Amicoyuan
8s
8s
[QST] About Example 23: Ampere GEMM Operand Reduction Fusion with kGemmSplitKParallel Auto Label Issues #156: Issue #2756 opened by kitecats
5s
5s
[QST] how to use cutlass in pytorch backward Auto Label Issues #155: Issue #2755 opened by xkwnzxxjw
9s
9s
[BUG] Why are shuffle_sync_up, shuffle_sync_down, warp_reduction_sum, etc. not exposed in the API, though present in examples? Auto Label Issues #154: Issue #2754 opened by narain1
9s
9s
[BUG] nvidia-cutlass-dsl needs __version__ attribute Auto Label Issues #153: Issue #2753 opened by jduprat
8s
8s
[BUG] cutlass profiler gemm fails on certain shapes, on Blackwell (B200, GB200, GB300) Auto Label Issues #152: Issue #2751 opened by michael604work
10s
10s
[BUG] [CuTeDSL] PDL example has missing attribute, AttributeError: module 'cutlass.cute.arch' has no attribute 'griddepcontrol_launch_dependents' Auto Label Issues #151: Issue #2748 opened by lessw2020
8s
8s
[QST] [CuTeDSL] Kernel Launch Failure due to requesting too much SMEM Auto Label Issues #150: Issue #2747 opened by HanGuo97
8s
8s
[QST] Is nvfp4 gemv supported on RTX 5090? Auto Label Issues #149: Issue #2743 opened by ZhuYe0729
9s
9s
[BUG] [CuTe DSL] Cold L2 leads to IMA in grouped_blockscaled_gemm Auto Label Issues #148: Issue #2737 opened by reubenconducts
16s
16s
[QST] Is it possible to run Hopper-specific kernels on Blackwell? Auto Label Issues #147: Issue #2736 opened by ankutalev
8s
8s
[QST]Will Blockscaled GEMM support Linear Layout Scale Factor? Auto Label Issues #146: Issue #2735 opened by HydraQYH
8s
8s
[QST] Does CUTLASS provide APIs like binary op min/max and dtype's INF/NAN Auto Label Issues #145: Issue #2733 opened by Rachmanino
11s
11s
[QST][CuteDSL] Why cute.make_layout throw an error about dynamic/static shape Auto Label Issues #144: Issue #2730 opened by zfan2356
8s
8s
[BUG] [Cutlass C++] Illegal instruction on Sm120a with certain shape Auto Label Issues #143: Issue #2728 opened by AI4SClab
8s
8s
[BUG] Build cutlass with arch=100a failed Auto Label Issues #142: Issue #2727 opened by Peakulorain
13s
13s
[QST] Why do you suck Nvidia? Why give 99kb SRAM to 600W workstation cards WHAATTTTTTTT!!!!!!!!!!!!!!!!!!!!!!!!! Auto Label Issues #141: Issue #2726 opened by IISuperluminaLII
23s
23s
[QST]How to configure blockM and blockN in GemmUniversal? Auto Label Issues #140: Issue #2725 opened by 4grass
14s
14s
[QST]Legal threadblock, warp, mma shape Auto Label Issues #139: Issue #2724 opened by mbernaschi
13s
13s
[BUG] for loop and while loop have different behavior Auto Label Issues #138: Issue #2723 opened by HarryWu99
14s
14s