Skip to content

Issues: NVIDIA/cutlass

Beta
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Clear current search query, filters, and sorts
267 Open 1,102 Closed
267 Open 1,102 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

[FEA] cute hopper conv example CuTe CuTe Functionality feature request New feature or request help wanted Interested in support from contributors inactive-30d inactive-90d
#1064 opened Aug 25, 2023 by zhang662817 updated Mar 5, 2024 CUTLASS 3.5
2
[BUG] Implicitly generate unexpected LDGSTS instructions for A100 bug Something isn't working inactive-30d inactive-90d
#1231 opened Dec 4, 2023 by cctry updated Apr 3, 2024
3
[BUG] Wrong traits for 64-bit integer bug Something isn't working inactive-30d inactive-90d
#1219 opened Nov 29, 2023 by JieRen98 updated Apr 5, 2024 CUTLASS 3.4
4
[QST] How does local_tile work? Could you provide an detailed explanation? CuTe CuTe Functionality inactive-30d inactive-90d question Question
#1243 opened Dec 6, 2023 by ziyuhuang123 updated May 22, 2024
4
[QST] thread num assert in sm70_epilogue_vectorized ? - Needs Triage inactive-30d inactive-90d question Question
#1334 opened Feb 5, 2024 by mammoth831 updated Jun 6, 2024
3
[BUG] Illegal CUDA shared memory access in SM90 GEMM TMA Warpspecialized at ClusterBarrier::init ? - Needs Triage bug Something isn't working inactive-30d inactive-90d
#1247 opened Dec 6, 2023 by kadeng updated Jun 21, 2024
7
[QST] Epilogue Swizzle inactive-30d inactive-90d question Question
#1383 opened Mar 6, 2024 by jeromeku updated Jul 6, 2024
5
[QST] How to Use Gemv in a Cuda Kernel inactive-30d inactive-90d question Question
#1401 opened Mar 14, 2024 by StarrickLiu updated Jul 23, 2024
5
[QST] Is there any fp16xfp16 GEMM sample using CUTE with a performance comparable to cublas? feature request New feature or request good first issue Good first issue for contributors help wanted Interested in support from contributors question Question
#1686 opened Aug 6, 2024 by xiaonans updated Aug 6, 2024
[BUG] Disordered header files: detail::is_prefetch used before declaration. bug Something isn't working inactive-30d inactive-90d
#1484 opened Apr 15, 2024 by rchardx updated Aug 14, 2024
5
[QST] How to use swizzle to avoid bank conflict? ? - Needs Triage inactive-30d inactive-90d question Question
#1234 opened Dec 5, 2023 by ziyuhuang123 updated Aug 16, 2024
5
[QST] How local_partition works? ? - Needs Triage CuTe CuTe Functionality inactive-30d inactive-90d question Question
#1244 opened Dec 6, 2023 by ziyuhuang123 updated Aug 16, 2024
5
[QST] 2D Convolution for NCHW Row-Major images, kernels and output ? - Needs Triage inactive-30d inactive-90d question Question
#1496 opened Apr 19, 2024 by chart21 updated Aug 17, 2024
2
[QST]iterator store interface[store(Fragment &frag,TensorCoord const &tile_offset)] not offset in units of whole tiles inactive-30d inactive-90d question Question
#1314 opened Jan 20, 2024 by zhiyu-deep updated Aug 20, 2024
13
[QST] make error caused by glibc 2.32 in ubuntu 18.04 "//usr/local/lib/libpthread.so.0: undefined reference to" ? - Needs Triage inactive-30d inactive-90d question Question
#1490 opened Apr 17, 2024 by Zmy6 updated Aug 22, 2024
4
[QST] How to do concurrent GEMMs ? inactive-30d inactive-90d question Question
#1418 opened Mar 23, 2024 by jomivaan updated Aug 23, 2024
8
[BUG] Broken copy.hpp bug Something isn't working inactive-30d inactive-90d
#1508 opened Apr 28, 2024 by kroburg updated Aug 27, 2024
5
[QST] StreamK ReductionStrategy: "Atomic" or "Mixed" inactive-30d inactive-90d question Question
#1488 opened Apr 16, 2024 by HanGuo97 updated Aug 27, 2024
4
[BUG] Python EVT Pytorch Emitter Broken bug Something isn't working inactive-30d inactive-90d
#1462 opened Apr 8, 2024 by jeromeku updated Aug 27, 2024
4
[QST] use FastLinearCombinationClamp to convert half accumulator to int8_t output? ? - Needs Triage inactive-30d inactive-90d question Question
#1516 opened Apr 30, 2024 by hychiang-git updated Aug 28, 2024
2
[QST] Epilogue Reduction ? - Needs Triage inactive-30d inactive-90d question Question
#1518 opened Apr 30, 2024 by jeromeku updated Aug 28, 2024
2
[QST] cutlass::Array and cute::Tensor --- using CUTLASS utility structs/classes with CUTE (such as NumericArrayConverter) ? - Needs Triage inactive-30d inactive-90d question Question
#1532 opened May 10, 2024 by HanGuo97 updated Sep 7, 2024
2
[FEA] CUTLASS should ensure all its symbols are hidden from shared object libraries feature request New feature or request inactive-30d
#1027 opened Aug 4, 2023 by jrhemstad updated Sep 16, 2024 CUTLASS 3.4
3
[QST] Epilogue Broadcast: Adapter vs GemmUniversal inactive-30d question Question
#1459 opened Apr 7, 2024 by jeromeku updated Sep 20, 2024
7
[QST] Hopper mixed precision gemm always worse than FP8 ? - Needs Triage inactive-30d inactive-90d question Question
#1549 opened May 24, 2024 by divchenko updated Sep 23, 2024
8
ProTip! Updated in the last three days: updated:>2025-05-07.