NVIDIA / NVIDIA/cutlass

[QST]How to print smem in CuteDSL

Open
#2,834 3 comments 2 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

? - Needs Triage inactive-30d inactive-90d question
Dominant language
C++
Stars
10.5k
Forks
2.1k
Avg merge
3d 11h
Merged PRs (30d)
7

Description

tXsH0 = thr_copy_h0_s2r.partition_S(sH0[:,:,0])
                                   ^^^^^^^^^^
 File "/mnt/shared-storage-user/anaconda3/envs/fla/lib/python3.12/site-packages/nvidia_cutlass_dsl/python_packages/cutlass/cute/tensor.py", line 559, in _check_can_load_store
   raise NotImplementedError(
NotImplementedError: load & store swizzled memory is not supported yet: tensor<ptr<f32, smem, align<1024>, S<3,4,3>> o ((32,2),(8,8),(1,2)):((1,2048),(32,256),(0,4096))>

It seems unable to print smem

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the CuteDSL call to thr_copy_h0_s2r.partition_S(sH0[:,:,0]) and inspect _check_can_load_store in cutlass/cute/tensor.py, where the reported NotImplementedError is raised. Determine whether the requested behavior is printing the shared-memory tensor or supporting its load/store path, then verify the result against this failing example.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
tooling
Issue type
Feature
Difficulty
3/5
Estimated time
1-2 days
Activity status
Quiet
Clarity
Needs clarification
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.