-
Notifications
You must be signed in to change notification settings - Fork 1.9k
Issues
is:issue state:open
is:issue state:open
Issue creation is restricted in this repository
Search results
[BUG] ICE: Load/Store operand must be a pointer when loading/storing aligned FP4 tensor
bugSomething isn't workingSomething isn't workingStatus: Open.#3349 In NVIDIA/cutlass;[BUG] Allow boolean types for sm100_utils.make_smem_layout_epi
bugSomething isn't workingSomething isn't workingStatus: Open.#3348 In NVIDIA/cutlass;- Status: Open.#3343 In NVIDIA/cutlass;
[BUG] CuTeDSL segfaults when using vector<fp8>
bugSomething isn't workingSomething isn't workingStatus: Open.#3342 In NVIDIA/cutlass;- Status: Open.#3340 In NVIDIA/cutlass;
[BUG] SFA/SFB s2t (UTCCP) should be issued only by the leader CTA
bugSomething isn't workingSomething isn't workingStatus: Open.#3331 In NVIDIA/cutlass;[QST] Why is the accumulation dtype in the CUTLASS FP4 GEMV example in FP16, instead of FP32?
questionQuestionQuestionStatus: Open.#3330 In NVIDIA/cutlass;[BUG] Unhelpful error when cutedsl runtime shared libs aren't found
bugSomething isn't workingSomething isn't workingStatus: Open.#3329 In NVIDIA/cutlass;[BUG]
bugSomething isn't workingSomething isn't workingStatus: Open.#3327 In NVIDIA/cutlass;- Status: Open.#3325 In NVIDIA/cutlass;
- Status: Open.#3319 In NVIDIA/cutlass;
[BUG] The MHA-FP8 results from running CuteDSL on Thor are incorrect.
bugSomething isn't workingSomething isn't workingStatus: Open.#3318 In NVIDIA/cutlass;