onnxruntime
Fix attention perf regression
#8682
Merged

Fix attention perf regression #8682

tianleiwu
tianleiwu undo change in attention cpu
730a0a9c
tianleiwu fix perf regression
1eb0370e
tianleiwu fix build error
d300c2f2
tianleiwu remove undir_mask buffer
c2434b29
tianleiwu disable persistent softmax by default
ba98aba9
tianleiwu update comments
e4166581
tianleiwu tianleiwu requested a review from yufenglee yufenglee 4 years ago
tianleiwu tianleiwu requested a review from viboga viboga 4 years ago
tianleiwu tianleiwu requested a review 4 years ago
tianleiwu tianleiwu requested a review from gh-yewang gh-yewang 4 years ago
gh-yewang
gh-yewang approved these changes on 2021-08-11
tianleiwu tianleiwu merged f661c186 into master 4 years ago
tianleiwu tianleiwu deleted the tlwu/fix_attention_cpu_perf_regression branch 4 years ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone