biondizzle/DeepGEMM
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add more GPU architectures support (#112)

Browse Source 
* Add more GPU architectures support

* Update layout.py

* Optimize performance, Add SM90 support, Add 1D2D SM100 support

* Add fmtlib submodule at commit 553ec11

---------

Co-authored-by: fzyzcjy <5236035+fzyzcjy@users.noreply.github.com>
This commit is contained in:
Ray Wang
2025-07-18 11:32:22 +08:00
committed by GitHub
parent 03d0be3d2d
commit 9da4a23561
67 changed files with 5586 additions and 2965 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
csrc/indexing/main.cu Normal file
View File

@@ -0,0 +1,13 @@
#include <deep_gemm/impls/sm90_bf16_gemm.cuh>
#include <deep_gemm/impls/sm100_bf16_gemm.cuh>
#include <deep_gemm/impls/sm90_fp8_gemm_1d1d.cuh>
#include <deep_gemm/impls/sm90_fp8_gemm_1d2d.cuh>
#include <deep_gemm/impls/sm100_fp8_gemm_1d1d.cuh>
#include <deep_gemm/impls/sm100_fp8_gemm_1d2d.cuh>
#include <deep_gemm/impls/smxx_layout.cuh>
using namespace deep_gemm;
int main() {
return 0;
}