Skip to content

Batched GEMM#120

Merged
asroy merged 1 commit into
developfrom
jfy_batched_gemm_rebased
Mar 21, 2022
Merged

Batched GEMM#120
asroy merged 1 commit into
developfrom
jfy_batched_gemm_rebased

Conversation

@j4yan
Copy link
Copy Markdown
Contributor

@j4yan j4yan commented Mar 10, 2022 •
edited by rocking5566
Loading

  • refactored deviceBatchedGemm;
  • removed GridwiseBatchedGemm;
  • added fp32 and int8 to the profiler.

@asroy
Copy link
Copy Markdown
Contributor

asroy commented Mar 10, 2022

Also need to add ctest

@j4yan j4yan changed the title Batched GEMM WIP: Batched GEMM Mar 10, 2022
@asroy
Copy link
Copy Markdown
Contributor

asroy commented Mar 14, 2022

Is this PR ready for review?

@j4yan
Copy link
Copy Markdown
Contributor Author

j4yan commented Mar 14, 2022

Is this PR ready for review?

Except the performance issue, yes.

@j4yan j4yan changed the title WIP: Batched GEMM Batched GEMM Mar 18, 2022
@j4yan j4yan force-pushed the jfy_batched_gemm_rebased branch from 58d6f5f to 001df06 Compare March 19, 2022 01:07
@j4yan
refactored deviceBatchedGemm; removed GridwiseBatchedGemm; added fp32…
6bafabf 
… and int8 to profiler

changed long_index_t to index_t when computing memory offset

uncomment other ops in profiler

added test for batched_gemm
@j4yan j4yan force-pushed the jfy_batched_gemm_rebased branch from 001df06 to 6bafabf Compare March 19, 2022 01:10
asroy
asroy reviewed Mar 21, 2022
View reviewed changes
Comment thread library/src/tensor_operation_instance/gpu/batched_gemm/CMakeLists.txt
device_batched_gemm_xdl_f16_f16_f16_gmk_gnk_gmn_instance.cpp;
device_batched_gemm_xdl_f16_f16_f16_gkm_gkn_gmn_instance.cpp;
device_batched_gemm_xdl_f16_f16_f16_gkm_gnk_gmn_instance.cpp;
device_batched_gemm_xdl_f32_f32_f32_gmk_gkn_gmn_instance.cpp;
Copy link
Copy Markdown
Contributor

@asroy asroy Mar 21, 2022

Choose a reason for hiding this comment

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

There is not BF16 GEMM?

asroy
asroy approved these changes Mar 21, 2022
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.

In future PR, pleaes

  1. add bfp16 batched GEMM
  2. Reimplement test using ckProfiler

@asroy asroy merged commit cb87b04 into develop Mar 21, 2022
@illsilin illsilin deleted the jfy_batched_gemm_rebased branch December 8, 2023 16:03
linqun added a commit that referenced this pull request Nov 26, 2025
@linqun
[CK_TILE] Port hw independent changes from internal repo to develop b…
a62f2a2 
…ranch

It includes PR#96, #114, #120, #121.
@linqun linqun mentioned this pull request Nov 26, 2025
Merged
7 tasks
linqun added a commit that referenced this pull request Dec 10, 2025
@linqun
[CK_TILE] Port hw independent changes from internal repo to develop b…
2d81c14 
…ranch

It includes PR#96, #114, #120, #121.
illsilin pushed a commit that referenced this pull request Dec 12, 2025
@linqun
[CK_TILE] Port hw independent changes from internal repo to develop b…
fc7bf0a 
…ranch (#3301)

* [CK_TILE] Port hw independent changes from internal repo to develop branch

It includes PR#96, #114, #120, #121.

* correct rebase error
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

@j4yan @asroy