Skip to content

Issues: NVIDIA/Fuser

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

220 Open 601 Closed
220 Open 601 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

Indexing error with circular buffering and unswitch bug Something isn't working
#4189 opened Apr 3, 2025 by naoyam
@naoyam
InsertReshardingsPass failed to decompose a set with both shard additions and deletions. Multi-GPU
#4188 opened Apr 3, 2025 by wujingyue
Verify the IdModel-based indexer matches with the legacy indexer
#4175 opened Apr 2, 2025 by naoyam
@naoyam
1
Outer reduction performance regression rope
#4172 opened Apr 2, 2025 by naoyam
1
@liqiangxl
RoPE benchmark performance fluctuations rope
#4171 opened Apr 2, 2025 by naoyam
@naoyam
Split-K predicate is missing on TMA store Matmuls
#4164 opened Apr 1, 2025 by jacobhinkle
@jacobhinkle
1
Error scheduling Hopper matmul with use_smem_epilogue and splitk Matmuls
#4159 opened Mar 31, 2025 by jacobhinkle
@jacobhinkle
getUnrollFactor incorrectly assumes that all vectorized inputs are using the same vectorized width
#4149 opened Mar 26, 2025 by liqiangxl
3
InsertReshardingsPass decomposes matmul/linear+allreduce. Multi-GPU
#4133 opened Mar 24, 2025 by wujingyue
Cache isResharding during segmentation. Multi-GPU
#4132 opened Mar 24, 2025 by wujingyue
Detect uncacheable tensors through reshape
#4097 opened Mar 18, 2025 by naoyam
@naoyam
max_persistent_buffer_size may be smaller than total_reduction_numel bug Something isn't working
#4075 opened Mar 13, 2025 by naoyam
@liqiangxl
2
Refactor IndexLowering::handle(const LoadStoreOp* ldst) enhancement New feature or request
#4058 opened Mar 11, 2025 by rdspring1
RFE: Take contiguity caching into nvFuser enhancement New feature or request
#4043 opened Mar 7, 2025 by csarofeen
inplace update done via aliased outputs should have more strict checks
#4036 opened Mar 7, 2025 by jjsjann123
1
benchmarking suite should initialize cuda graphs / profiler interaction Python Benchmarks
#4008 opened Mar 4, 2025 by tfogal
take_along_axis validation error (race condition?) bug Something isn't working
#4003 opened Mar 3, 2025 by naoyam
@naoyam
checking for compatible allocation domain on Fusion::replaceOutput
#3994 opened Feb 28, 2025 by jjsjann123
Extend IdModel to map DIDs for certain patterns. Multi-GPU
#3987 opened Feb 27, 2025 by wujingyue
@naoyam
7
Reenable reshape canceling for Mistral backward rope
#3975 opened Feb 26, 2025 by naoyam
cudaErrorMisalignedAddress when sweeping matmul problems with NN, TN, and TT layouts. Matmuls
#3966 opened Feb 25, 2025 by rdspring1
1
@rdspring1
1
Optimize TMA Store logic to handle pipelining and aliasing. Matmuls
#3961 opened Feb 25, 2025 by rdspring1
@rdspring1
Missing block sync after epilogue compute but before stmatrix (Correctness) Matmuls
#3960 opened Feb 25, 2025 by rdspring1
@rdspring1
Swizzle tiles in matmul without introducing larger grid due to nondivisible splits Matmuls
#3942 opened Feb 21, 2025 by jacobhinkle
@jacobhinkle
1
Allow separate sub-DAG for load and compute warp groups with warp-specialized circular buffering. Matmuls TMA
#3941 opened Feb 21, 2025 by rdspring1
@rdspring1
ProTip! Exclude everything labeled bug with -label:bug.