Skip to content

INT8 GEMM support #93

@albiol2004

Description

@albiol2004

The C++ matmul kernels (aie2/mm.cc, aie2p/mm.cc) already have vectorized INT8
matmul templates (i8→i8, i8→i16, i8→i32, MAC shape 8x8x8) and compile flags
(-Di8_i8_ONLY, etc.), but the Python GEMM operator only accepts bf16 input.

This would wire up INT8 through the Python layer:

  • design.py: add "i8" to dtype_in, "i8"/"i16"/"i32" to dtype_out,
    add i8 MAC dims (8,8,8) to microkernel_mac_dim_map
  • op.py: add i8 kernel flags, min tile sizes, skip bf16 emulation for int8
  • reference.py: int8 golden reference
  • test.py: int8 test cases

The NPU does ~50 TOPs INT8 vs ~3-5 TOPs bf16, so this would be a big
throughput gain for quantized inference.

If I'm not missing anything it should be trivial, I'm already on it.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions