Haocong WANG 020148d0f7 [BlockScale GEMM] FP8 Blockscale GEMM optimization and ckProfiler (#1913) ...

* Added two kernel for M=32 problem

* Comment the first one

* Enable multiply_multiply for Scale_Block_M = 1 for deepseek

* Modify the a_thread offset since the A data load is different from B.

* edit fp8 ab scale for Scale_Block_M=1

* edit GemmSpec to MNKPadding

* enable blockwise pipelie v1 and v2. v1 is work for small K.

* add instance for gemm_ab_scale

* fix cmakelist of ckProfiler

* optimize blockscale gemm. todo: reduce vgpr usage

* fix a correctness bug

* sanity checked

* revert ckprofiler cmake changes

* clang format

* revert unnecessary changes.

* remove commented codes.

---------

Co-authored-by: mtgu0705 <mtgu@amd.com>
Co-authored-by: chenjun <junchen2@amd.com>

2025-02-25 15:42:20 +08:00

..

ck/library

[BlockScale GEMM] FP8 Blockscale GEMM optimization and ckProfiler (#1913)

2025-02-25 15:42:20 +08:00