Skip to main content

/

DocsCode

Mojo struct

B200MatmulSmem

struct B200MatmulSmem[a_type: DType, b_type: DType, c_type: DType, transpose_b: Bool, *, config: MatmulConfig[a_type, b_type, c_type, transpose_b]]

Shared memory layout for B200 SM100 matrix multiplication kernel.

This struct manages the shared memory allocation for:

  • Input tiles (A and B matrices) with multi-stage pipelining
  • Output tile (C matrix) for accumulation
  • Synchronization barriers for producer-consumer coordination
  • CLC (Cluster Launch Control) barriers and response storage
  • TMEM (Tensor Memory) address and deallocation barrier

The memory is organized to support asynchronous TMA loads and efficient bank-conflict-free access patterns for tensor core operations.

Type aliases are provided for tile types (ATile, BTile, CTile) to enable cleaner function signatures.

Fields​

  • ​input_tiles (B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].InputTiles):
  • ​output_tiles (B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].OutputTiles):
  • ​pipelines (B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].Pipelines):

Implemented traits​

AnyType, ImplicitlyDestructible

comptime members​

__del__is_trivial​

comptime __del__is_trivial = True

a_smem_layout​

comptime a_smem_layout = B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].Layouts.a_smem_layout

ATileArray​

comptime ATileArray = B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].InputTiles.ATileArray

b_smem_layout​

comptime b_smem_layout = B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].Layouts.b_smem_layout

BK​

comptime BK = config.block_tile_shape.__getitem__[Int](2)

BM​

comptime BM = config.block_tile_shape.__getitem__[Int](0)

BN​

comptime BN = config.block_tile_shape.__getitem__[Int](1)

BTileArray​

comptime BTileArray = B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].InputTiles.BTileArray

c_smem_layout​

comptime c_smem_layout = B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].Layouts.c_smem_layout

CTileArray​

comptime CTileArray = B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].OutputTiles.CTileArray

InputTiles​

comptime InputTiles = StandardTileStorage[a_type, b_type, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BM, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BK, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BN, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BK, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].num_pipeline_stages]

Layouts​

comptime Layouts = SmemLayouts[a_type, b_type, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BM, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BN, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BK, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].OutputM, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].OutputN, config.a_swizzle, config.b_swizzle, transpose_b]

num_accum_pipeline_stages​

comptime num_accum_pipeline_stages = config.num_accum_pipeline_stages

num_clc_pipeline_stages​

comptime num_clc_pipeline_stages = config.num_clc_pipeline_stages

num_group_pipeline_stages​

comptime num_group_pipeline_stages = (B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].num_pipeline_stages // config)

num_output_stages​

comptime num_output_stages = config.num_output_stages

num_pipeline_stages​

comptime num_pipeline_stages = config.num_pipeline_stages

OutputM​

comptime OutputM = config.output_tile_shape.__getitem__[Int](0)

OutputN​

comptime OutputN = config.output_tile_shape.__getitem__[Int](1)

OutputTiles​

comptime OutputTiles = OutputTileStorage[c_type, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].OutputM, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].OutputN, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].num_output_stages]

Pipelines​

comptime Pipelines = SmemPipelineBundle[B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].num_group_pipeline_stages, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].num_accum_pipeline_stages, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].num_clc_pipeline_stages, StandardTilePayload[a_type, b_type, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BM, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BK, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BN, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BK, B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].num_pipeline_stages]]

Methods​

a_tiles​

a_tiles(ref[AddressSpace._value._mlir_value] self) -> B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].ATileArray

Returns:

B200MatmulSmem

b_tiles​

b_tiles(ref[AddressSpace._value._mlir_value] self) -> B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].BTileArray

Returns:

B200MatmulSmem

c_tiles​

c_tiles(ref[AddressSpace._value._mlir_value] self) -> B200MatmulSmem[a_type, b_type, c_type, transpose_b, config=config].CTileArray

Returns:

B200MatmulSmem

ab_pipeline_size​

static ab_pipeline_size() -> Int

Total size of A+B tiles for all pipeline stages (in elements).

Returns:

Int

c_output_size​

static c_output_size() -> Int

Size of C tiles for all output stages (in elements).

Returns:

Int

total_tile_size​

static total_tile_size() -> Int

Total tile storage size (A+B+C) in elements.

Returns:

Int

View source

View source

Was this page helpful?

Models

Cookbook

Puzzles

Blog

Community

Contact