ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-21 13:29:20 +00:00
Code Issues Packages Projects Releases Wiki Activity
676 Commits 751 Branches 38 Tags
54af9c3aa069cbb328d752ca7ec9084fa8f47de5
Commit Graph

3 Commits

Author SHA1 Message Date
rocking5566
686e38c5f0 [What] Fix bug of verification fail on E Matrix (#371) 
[Why] We need to sync lds even in first loop because Gemm also use the same LDS.

[ROCm/composable_kernel commit: c366de553e]
 2022-08-22 07:50:28 -05:00
Chao Liu
215b32cd3f fix build issue (#357) 
* fix build

* excludeexample_gemm_max_xdl_fp16 from testing due to random failure on gfx908

[ROCm/composable_kernel commit: 5ee304595c]
 2022-08-13 15:58:31 -05:00
rocking5566
ada72bead2 Gemm multiple d multiple r (#335) 
* Imitate XXX_gemm_multiple_d, add XXX_gemm_multiple_d_multiple_r for gemm + reduction

* Implement run of kernel

* Add example

* Fix parameter of typo

* Rewrite the reduceMax example

* Rewrite the reduceMean + reduceMeanSquare example

* Refine naming

* Refine folder name

* refine naming

* Rewrite the gemm + bias + relu + add + layernorm example

* Rewrite the gemm + layernorm example

* clang-format

* Fix bug if sync lds

* Fix compile error

[ROCm/composable_kernel commit: 6c3c06bf1f]
 2022-08-13 01:07:12 -05:00