Mojo struct

STMatrixOffsets

@register_passable(trivial) struct STMatrixOffsets[BM: Int, BN: Int, *, num_threads: Int, accum_type_size: Int, curr_repeat: Int, cumulative_repeat: Int, m_mma: Int]

Implemented traits

AnyType, Copyable, ImplicitlyCopyable, Movable, UnknownDestructibility

Aliases

`copyinitis_trivial`

comptime __copyinit__is_trivial = True

`delis_trivial`

comptime __del__is_trivial = True

`moveinitis_trivial`

comptime __moveinit__is_trivial = True

`b32_per_repeat`

comptime b32_per_repeat = ((STMatrixLayout[BM, BN, num_threads=num_threads, accum_type_size=accum_type_size].elements_per_repeat * accum_type_size) // 4)

`local_frag_size_b32`

comptime local_frag_size_b32 = (curr_repeat * STMatrixOffsets[BM, BN, num_threads=num_threads, accum_type_size=accum_type_size, curr_repeat=curr_repeat, cumulative_repeat=cumulative_repeat, m_mma=m_mma].b32_per_repeat)

`ptr_offset`

comptime ptr_offset = (STMatrixOffsets[BM, BN, num_threads=num_threads, accum_type_size=accum_type_size, curr_repeat=curr_repeat, cumulative_repeat=cumulative_repeat, m_mma=m_mma].b32_per_repeat * ((STMatrixLayout[BM, BN, num_threads=num_threads, accum_type_size=accum_type_size].repeat * m_mma) + cumulative_repeat))

`STLayout`

comptime STLayout = STMatrixLayout[BM, BN, num_threads=num_threads, accum_type_size=accum_type_size]

`tmem_col_offset`

comptime tmem_col_offset = ((cumulative_repeat * STMatrixLayout[BM, BN, num_threads=num_threads, accum_type_size=accum_type_size].frag_simdwidth) * 4)

`tmem_offset`

comptime tmem_offset = ((STMatrixOffsets[BM, BN, num_threads=num_threads, accum_type_size=accum_type_size, curr_repeat=curr_repeat, cumulative_repeat=cumulative_repeat, m_mma=m_mma].tmem_row_offset << 16) + STMatrixOffsets[BM, BN, num_threads=num_threads, accum_type_size=accum_type_size, curr_repeat=curr_repeat, cumulative_repeat=cumulative_repeat, m_mma=m_mma].tmem_col_offset)

`tmem_row_offset`

comptime tmem_row_offset = (16 * m_mma)

Methods

`init`

__init__() -> Self

Implemented traits​

Aliases​

__copyinit__is_trivial​

__del__is_trivial​

__moveinit__is_trivial​

b32_per_repeat​

local_frag_size_b32​

ptr_offset​

STLayout​

tmem_col_offset​

tmem_offset​

tmem_row_offset​

Methods​

__init__​