ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-24 14:54:47 +00:00
Code Issues Packages Projects Releases Wiki Activity

[CK_TILE] add tensorwise quant in grouped gemm (#3007)

Browse Source 
* add tensorwise quant in grouped gemm

* fix example issue

* update test cases

* format codes

* clang format

* use GTEST_FAIL

* fix a bug in test_grouped_gemm_util

* skip test when use wmma on grouped_quant kernel

* change cmake

* change code based on comments

---------

Co-authored-by: ThomasNing <thomas.ning@amd.com>

[ROCm/composable_kernel commit: 3c12a02827]
This commit is contained in:
kyle-256
2025-10-24 22:41:54 +08:00
committed by GitHub
parent 480f05ffd9
commit b49f5d9de5
11 changed files with 671 additions and 68 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
include/ck_tile/ops/gemm_quant/kernel/grouped_gemm_quant_kernel.hpp
View File

@@ -393,6 +393,13 @@ struct QuantGroupedGemmKernel
aq_block_window,
bq_block_window);
}
else if constexpr(kQuantType == QuantType::TensorQuant)
{
const AccDataType aq_scale = type_convert<AccDataType>(*aq_ptr);
const AccDataType bq_scale = type_convert<AccDataType>(*bq_ptr);
EpiloguePipeline{}(
c_block_window, c_block_tile, c_block_window, smem_ptr_0, aq_scale, bq_scale);
}
}
// For persistent kernels