forked from pytorch/FBGEMM
-
Notifications
You must be signed in to change notification settings - Fork 11
Pull requests: ROCm/FBGEMM
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[ROCm][CI] Restore ROCm GPU CI: switch to the MI350 runner label
#168
opened Aug 19, 2026 by
aryaman-gupta
Loading…
[ROCm] Make the Python NFP8 dtype selection architecture-aware
#167
opened Aug 19, 2026 by
aryaman-gupta
Loading…
[ROCm] Fix packed-bag pooling truncation in nbit inference forward
#166
opened Aug 19, 2026 by
aryaman-gupta
Loading…
[ROCm] Fix flaky forward-tolerance in backward adagrad ROCm test
#165
opened Aug 18, 2026 by
avbokovoy
Loading…
[ROCm] Fix GPU memory fault in batched_unary_embeddings backward
#164
opened Aug 18, 2026 by
avbokovoy
Loading…
Generalizing the HIP-optimized
warp_per_row kernel
#146
opened Mar 19, 2026 by
aryaman-gupta
•
Draft
Split Kernel Optimization for New feature or request
group_index_select_or_add_2d_kernel
enhancement
#142
opened Feb 2, 2026 by
aryaman-gupta
Loading…
Optimize
group_index_select_or_add_2d_kernel by adding a separate codepath for small embedding dimensions
#135
opened Dec 16, 2025 by
aryaman-gupta
Loading…
Fixes bug in one specialized HIP instantiation of the
warp-per-row kernel
#134
opened Dec 5, 2025 by
aryaman-gupta
Loading…
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.