Mojo function

load_matrix_a

load_matrix_a[m: Int, n: Int, k: Int](a_ptr: UnsafePointer[SIMD[float32, 1]], tile_row: Int, tile_col: Int, ldm: Int) -> SIMD[float32, 4]

For shape m16n8k8 type tf32 loads matrix A tile from memory to registers in specific order to be used by tensor cores to perform a warp sync mma op.

load_matrix_a[m: Int, n: Int, k: Int](a_ptr: UnsafePointer[SIMD[float16, 1]], tile_row: Int, tile_col: Int, ldm: Int) -> SIMD[float16, 4]

For shape m16n8k8 & type fp16 loads matrix A tile from memory to registers in specific order to be used by tensor cores to perform a warp sync mma op.

load_matrix_a[m: Int, n: Int, k: Int](a_ptr: UnsafePointer[SIMD[bfloat16, 1]], tile_row: Int, tile_col: Int, ldm: Int) -> SIMD[bfloat16, k.__floordiv__(2)]

For type bfp16 loads matrix A tile from memory to registers in specific order to be used by tensor cores to perform a warp sync mma op.

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