ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 14:59:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CK-TILE] Default2DEpilogue, example and adding nullptr_t type for D (#2752)

Browse Source 
* Init commit

* Quick fix, CI fails

* Remove CDElementWise

* Add CDEELementWise

---------

Co-authored-by: Thomas Ning <Thomas.Ning@amd.com>
This commit is contained in:
Mateusz Ozga
2025-08-28 21:45:50 +02:00
committed by GitHub
parent 038ea82315
commit 0758883fa4
20 changed files with 636 additions and 441 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
include/ck_tile/ops/rmsnorm2d/pipeline/rmsnorm2d_fwd_pipeline_model_sensitive_pass.hpp
View File

@@ -221,7 +221,7 @@ struct Rmsnorm2dFwdPipelineModelSensitiveT5Pass
}
else
{
Epilogue{}(y_window_, rmsn);
Epilogue{}(y_window_, rmsn, nullptr);
}
}
};

2
include/ck_tile/ops/rmsnorm2d/pipeline/rmsnorm2d_fwd_pipeline_one_pass.hpp
View File

@@ -160,7 +160,7 @@ struct Rmsnorm2dFwdPipelineOnePass
}
else
{
Epilogue{}(y_window_, rmsn);
Epilogue{}(y_window_, rmsn, nullptr);
}
}
};

2
include/ck_tile/ops/rmsnorm2d/pipeline/rmsnorm2d_fwd_pipeline_two_pass.hpp
View File

@@ -195,7 +195,7 @@ struct Rmsnorm2dFwdPipelineTwoPass
});
static_assert(kFusedQuant == Rmsnorm2dFusedQuantEnum::NO_SWEEP);
Epilogue{}(y_window, rmsn);
Epilogue{}(y_window, rmsn, nullptr);
move_tile_window(gamma_window, {-Block_N});
move_tile_window(y_window, {0, -Block_N});