Mojo module

matmul

SM100 matmul - COMPATIBILITY LAYER for grouped_matmul.

NOTE: This module is maintained for backward compatibility with grouped_matmul implementations that depend on internal functions (WarpRole, consumer_main_loop, stsm_helper, shared_memory_epilogue, register_epilogue, accum_arrive).

For new code, use sm100_structured directly:

• Import configs from: linalg.matmul.gpu.sm100_structured.config
• Import matmul from: linalg.matmul.gpu.sm100_structured.matmul

comptime values

RLayout32Bits

comptime RLayout32Bits[layout: Layout] = RuntimeLayout[layout, element_type=DType.uint32, linear_idx_type=DType.uint32]

Parameters

• ​layout (Layout):

Structs

• ​WarpRole:

Functions

• ​accum_arrive:
• ​consumer_main_loop:
• ​f32_frag_to_smem:
• ​register_epilogue:
• ​shared_memory_epilogue:
• ​shared_memory_epilogue_transpose:
• ​stsm_helper:

Was this page helpful?

Thank you! We'll create more content like this.

Thank you for helping us improve!

😔 What went wrong?

Some code doesn’t work

It includes inaccurate information

It's missing information I need

It was difficult to understand

Other

Submit