Skip to content

refactor dynamic xdlops iGemm#13

Merged
asroy merged 12 commits into
developfrom
xdlops_refactor
Aug 19, 2021
Merged

refactor dynamic xdlops iGemm#13
asroy merged 12 commits into
developfrom
xdlops_refactor

Conversation

@zjing14
Copy link
Copy Markdown
Contributor

@zjing14 zjing14 commented Aug 13, 2021

  • Refactor gridwise-/blockwise-/xdlops-gemm

@zjing14 zjing14 requested a review from asroy August 16, 2021 13:42
asroy
asroy reviewed Aug 16, 2021
View reviewed changes
Comment thread host/driver_offline/include/driver_dynamic_gemm_xdlops_v2r3.hpp Outdated
asroy
asroy reviewed Aug 16, 2021
View reviewed changes
Comment thread composable_kernel/include/tensor_operation/gridwise_dynamic_gemm_xdlops_v2r3.hpp Outdated
asroy
asroy reviewed Aug 16, 2021
View reviewed changes
Comment thread composable_kernel/include/tensor_operation/blockwise_gemm_xdlops.hpp Outdated
asroy
asroy reviewed Aug 16, 2021
View reviewed changes
Comment thread composable_kernel/include/tensor_operation/blockwise_gemm_xdlops.hpp Outdated
asroy
asroy reviewed Aug 16, 2021
View reviewed changes
Comment thread composable_kernel/include/tensor_operation/blockwise_gemm_xdlops.hpp Outdated
asroy
asroy reviewed Aug 16, 2021
View reviewed changes
Comment thread composable_kernel/include/tensor_operation/blockwise_gemm_xdlops.hpp Outdated
asroy
asroy suggested changes Aug 16, 2021
View reviewed changes
Copy link
Copy Markdown
Contributor

@asroy asroy left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Besides some change request, please also merge with this branch

#8

@asroy
Copy link
Copy Markdown
Contributor

asroy commented Aug 16, 2021

I've merged #8 into develop

asroy
asroy suggested changes Aug 18, 2021
View reviewed changes
Comment thread composable_kernel/include/tensor_operation/blockwise_gemm_xdlops.hpp Outdated
Comment thread composable_kernel/include/tensor_operation/blockwise_gemm_xdlops.hpp Outdated
Comment thread composable_kernel/include/tensor_operation/gridwise_gemm_xdlops_v2r3.hpp Outdated
Comment thread composable_kernel/include/tensor_operation/gridwise_gemm_xdlops_v2r3.hpp Outdated
@asroy
Copy link
Copy Markdown
Contributor

asroy commented Aug 19, 2021

I tried this PR, but I cannot compile V4R4R4XDLNHWC for fp16

/root/workspace/composable_kernel/composable_kernel/include/tensor_operation/xdlops_gemm.hpp:619:53: error: function 'GetMfma<_Float16, 32, 32>' with deduced return type cannot be used before it is defined

Maybe something wrong with the tuning parameter?
@zjing14

@asroy asroy self-requested a review August 19, 2021 14:51
@asroy asroy merged commit a2ad6d3 into develop Aug 19, 2021
@aserio aserio mentioned this pull request Aug 20, 2021
Closed
asroy added a commit that referenced this pull request Dec 1, 2023
@asroy
Shuffle in thread (#13)
1cf54e8 
* adding in-thread shuffle

* update softmax example

* refactor grid gemm

* refactor gemm: layouts

* bug fix

* clean

* clean
@illsilin illsilin deleted the xdlops_refactor branch December 7, 2023 18:39
carlushuang pushed a commit that referenced this pull request Mar 26, 2024
@illsilin
Merge pull request #13 from ROCmSoftwarePlatform/merge_from-_public_repo
984889f 
Merge from  public repo
@illsilin illsilin mentioned this pull request Oct 15, 2025
Merged
7 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

1 more reviewer

@asroy asroy asroy approved these changes

Reviewers whose approvals may not affect merge requirements

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@zjing14 @asroy