onnxruntime
Enabling S8S8 and S8U8 handling in QGemm for AVX2 and AVX-VNNI
#21123
Closed
Overview
Commits
4
Changes
View On GitHub
  • onnxruntime
    • core/mlas/lib
      • amd64
        • File
          QgemmU8S8KernelAvx2.asm
      • File
        mlasi.h
      • File
        platform.cpp
      • File
        qgemm.h
      • File
        qgemm_kernel_avx2.cpp
      • x86_64
        • File
          QgemmU8S8KernelAvx2.S
    • test/mlas/unittest
      • File
        test_qgemm.cpp

Loading comments...

FAQ Terms Privacy Refunds Impressum