DeepSpeed
Fix multiple issues in Op Builder and refactor implementation
#9085
Merged

Fix multiple issues in Op Builder and refactor implementation #9085

hwchen2017 merged 13 commits into master from hongwei/op_builder
hwchen2017
hwchen2017 remove build for cpu
9b2a0601
hwchen2017 move cuda/rocm related code to cudaopbuilder
c7f4a24e
hwchen2017 Fix cuda mismatch problem
651fe0a5
hwchen2017 Fix problem in prebuilt ops
eea2048d
hwchen2017 remove hardcoded cpp standard
4a67eaff
hwchen2017 Cache cpu info
4f53ff0f
hwchen2017 Unify compatibility check
3c55eb0d
hwchen2017 Support a suffix
c59029ea
hwchen2017 Minor fix
6aaaf8c1
hwchen2017 restore jit mode
f96ed521
hwchen2017 trim unit test
4046c9de
hwchen2017 support bf16 on cuda 12.2 or higher; remove used code
d8eda136
hwchen2017 hwchen2017 requested a review from pengdurice pengdurice 4 days ago
hwchen2017 hwchen2017 requested a review from delock delock 4 days ago
hwchen2017 hwchen2017 requested a review from sfc-gh-truwase sfc-gh-truwase 4 days ago
hwchen2017 hwchen2017 requested a review from tjruwase tjruwase 4 days ago
hwchen2017 hwchen2017 requested a review from loadams loadams 4 days ago
hwchen2017 hwchen2017 requested a review from tohtana tohtana 4 days ago
chatgpt-codex-connector
gpu-test-labeler gpu-test-labeler added skip-gpu-test
hwchen2017 Don't include debug info by default
1502e83c
gpu-test-labeler gpu-test-labeler removed skip-gpu-test
gpu-test-labeler gpu-test-labeler added skip-gpu-test
sfc-gh-truwase
sfc-gh-truwase commented on 2026-11-15
sfc-gh-truwase
sfc-gh-truwase approved these changes on 2026-11-15
hwchen2017 hwchen2017 merged e35282bd into master 2 days ago
hwchen2017 hwchen2017 deleted the hongwei/op_builder branch 2 days ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone