onnxruntime
Add composable kernel GEMM baseline for kernel explorer
#12364
Merged

Add composable kernel GEMM baseline for kernel explorer #12364

zhangyaobit merged 15 commits into master from guangyunhan/ke-ck-blas
cloudhan
cloudhan
cloudhan
cloudhan Split GemmBase RocBlasGemm
81db5b99
cloudhan Add composable kernel GEMM baseline
cdb9de75
cloudhan Make linter happy
f5efe4e5
cloudhan cloudhan force pushed from 3051f5b6 to f5efe4e5 3 years ago
cloudhan cloudhan marked this pull request as ready for review 3 years ago
cloudhan cloudhan requested a review from zhangyaobit zhangyaobit 3 years ago
zhangyaobit
zhangyaobit commented on 2022-07-30
zhangyaobit
zhangyaobit commented on 2022-07-30
zhangyaobit
zhangyaobit commented on 2022-07-30
cloudhan Address review comment
11d82cc3
cloudhan Update bert cases with batchsize
6fb96d9b
cloudhan Adjust includes to fix IWYU lint
4e65b5bb
cloudhan Only builds and links used ck kernels to improve building time
d9dbb685
cloudhan cloudhan requested a review from zhangyaobit zhangyaobit 3 years ago
zhangyaobit
zhangyaobit commented on 2022-08-02
cloudhan Remove warmup run on SelectImpl
efb327bd
cloudhan Add comment to utility function
ff8395f6
cloudhan Mute cpplint
c7b1207b
cloudhan Make RocBlasGemm<T>::SelectImpl semantically correct
7f0efab7
cloudhan Add reduced basic test cases for ck gemm
55a1e6da
zhangyaobit
zhangyaobit commented on 2022-08-02
cloudhan More robust gemm testing
f61461c3
cloudhan Fix warnings
beb70be3
cloudhan Fix grammar
a32c0062
zhangyaobit
zhangyaobit approved these changes on 2022-08-05
zhangyaobit zhangyaobit merged f39354d7 into master 3 years ago
zhangyaobit zhangyaobit deleted the guangyunhan/ke-ck-blas branch 3 years ago

Login to write a write a comment.

Login via GitHub

Reviewers
Assignees
No one assigned
Labels
Milestone