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.
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
Author
Label
Projects
Milestones
Assignee
Sort
Issues list
allocation order inference creating issue in multidevice_matmul tests
#2276
opened May 21, 2024 by
jjsjann123
Cannot define an New feature or request
matmul
MmaOp
where batch dimension is Broadcast
enhancement
#2273
opened May 20, 2024 by
jacobhinkle
Invalid loop promotion due to propagation of incomplete promotion
#2261
opened May 16, 2024 by
naoyam
Implement User Scheduling in Python Frontend
enhancement
New feature or request
Python API
Issues related to the Python API
#2259
opened May 16, 2024 by
rdspring1
4 of 20 tasks
NoOp scheduler claims zero-dimensional non-trivial fusions
Segmentation
Issues related to nvFuser Segmentation
#2241
opened May 14, 2024 by
jacobhinkle
allocation domain propagation to support reshape operations
allocation domain
issues related to allocation domain support
#2235
opened May 13, 2024 by
jjsjann123
Initial top-down presegmentation phase
enhancement
New feature or request
Segmentation
Issues related to nvFuser Segmentation
#2224
opened May 9, 2024 by
jacobhinkle
Allow reductions in matmul epilogues
enhancement
New feature or request
matmul
#2213
opened May 7, 2024 by
jacobhinkle
Support different cp.async and load stages for operands A and B
enhancement
New feature or request
matmul
#2211
opened May 7, 2024 by
jacobhinkle
Reduction scheduler does not handle allocation domain properly and trigger assert when reduction output has specified allocation domain
allocation domain
issues related to allocation domain support
#2202
opened May 5, 2024 by
jjsjann123
Write a sharded transformer block in nvFuser API.
Top-Down Matmul Dev
#2199
opened May 3, 2024 by
wujingyue
Add Double-Buffer TMA Support for Pointwise and Normalization Kernels
TMA
#2197
opened May 3, 2024 by
rdspring1
Restrict matmul scheduler to accepting only segments that are supported
#2167
opened May 1, 2024 by
jacobhinkle
LayerNorm fusion in Diffusion Transformer has worse performance than expected
Segmentation
Issues related to nvFuser Segmentation
Thunder
#2146
opened Apr 24, 2024 by
parthmannan
Previous Next
ProTip!
Mix and match filters to narrow down what you’re looking for.