onnxruntime
Rope imbedding kernel to use avx2
#23694
Merged

Rope imbedding kernel to use avx2 #23694

liqunfu merged 22 commits into main from liqun/Intel-ROPE-kernel-to-use-AVX2
liqunfu
liqunfu profile init code
4523b3d0
liqunfu from patch
59e27609
liqunfu Merge branch 'liqun/GQA' into Intel-ROPE-kernel-to-use-AVX2
1611fccc
liqunfu Merge branch 'Intel-ROPE-kernel-to-use-AVX2' into liqun/Intel-ROPE-ke…
6e2f414e
liqunfu liqunfu requested a review 1 year ago
github-actions
github-actions commented on 2025-02-14
liqunfu node_name and remove profiler wrapper
35bf5176
liqunfu m:erge branch 'main' into liqun/GQA
e1232fc8
liqunfu Merge branch 'liqun/GQA' into liqun/Intel-ROPE-kernel-to-use-AVX2
b36fa2c8
github-actions
github-actions commented on 2025-02-14
snnn
snnn commented on 2025-02-14
liqunfu remove profiling code
3964acc6
github-actions
github-actions commented on 2025-02-15
liqunfu undo test_gqa_cpu.py
40a68542
liqunfu lint
43bdb444
liqunfu some edit
46353d84
github-actions
github-actions commented on 2025-02-15
liqunfu fix data correctness in interleaved cases
e13ac56d
jywu-msft jywu-msft requested a review from fajin-corp fajin-corp 1 year ago
jywu-msft jywu-msft requested a review from yihonglyu yihonglyu 1 year ago
yihonglyu
liqunfu one more data correctness fix, add MLAS RoPE test to covert all scena…
9867ee78
github-actions
github-actions commented on 2025-02-19
github-advanced-security
github-advanced-security commented on 2025-02-19
liqunfu
liqunfu
liqunfu lint
754be929
liqunfu fix build - declaration error
4dc471b4
liqunfu unused RoPERegisterAllShortExecuteTests ci failure
5ecee3c6
liqunfu Merge branch 'liqun/Intel-ROPE-kernel-to-use-AVX2' of https://github.…
4e128da7
liqunfu missing implementation
62d5eefb
yihonglyu
yihonglyu commented on 2025-02-19
fajin-corp
fajin-corp commented on 2025-02-19
fajin-corp
fajin-corp commented on 2025-02-19
liqunfu add benchmark, etc.
d6de70f9
github-advanced-security
github-advanced-security commented on 2025-02-19
github-actions
github-actions commented on 2025-02-19
liqunfu Update onnxruntime/test/mlas/bench/bench_rope.cpp
89e1411f
liqunfu
liqunfu unaligned store
b46927ef
yihonglyu
yihonglyu commented on 2025-02-20
yihonglyu
yihonglyu commented on 2025-02-20
liqunfu sin ->sin_data, constexpr
484d1285
fajin-corp
fajin-corp approved these changes on 2025-02-20
yihonglyu
yihonglyu approved these changes on 2025-02-20
liqunfu liqunfu merged af04b202 into main 1 year ago
liqunfu liqunfu deleted the liqun/Intel-ROPE-kernel-to-use-AVX2 branch 1 year ago

Login to write a write a comment.

Login via GitHub

Assignees
No one assigned
Labels
Milestone