For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /max/get-started.md).
Mojo struct
MmaOpSM100_BlockScaled_SS
struct MmaOpSM100_BlockScaled_SS[c_type: DType, a_type: DType, b_type: DType, sfa_dtype: DType, sfb_dtype: DType, scaling_kind: UMMAKind, block_tile_shape: IndexList[3], mma_shape: IndexList[3], /, *, accum_type: DType = DType.float32, cta_group: Int = 1, cluster_shape: IndexList[3] = Index[Int, Int, Int](1, 1, 1), a_swizzle: TensorMapSwizzle = TensorMapSwizzle.SWIZZLE_128B, b_swizzle: TensorMapSwizzle = TensorMapSwizzle.SWIZZLE_128B, transpose_b: Bool = False, enable_small_sfb: Bool = False]
Fields
- idesc (
UMMAInsDescriptor[scaling_kind]): - mask (
UInt16):
Implemented traits
AnyType,
Copyable,
Defaultable,
ImplicitlyCopyable,
ImplicitlyDeletable,
Movable,
RegisterPassable,
TrivialRegisterPassable
Methods
__init__
def __init__() -> Self
mma
def mma(self, a: TileTensor[address_space=AddressSpace.SHARED, linear_idx_type=a.linear_idx_type, element_size=a.element_size], b: TileTensor[address_space=AddressSpace.SHARED, linear_idx_type=b.linear_idx_type, element_size=b.element_size], sfa_smem: TileTensor[address_space=AddressSpace.SHARED, linear_idx_type=sfa_smem.linear_idx_type, element_size=sfa_smem.element_size], sfb_smem: TileTensor[address_space=AddressSpace.SHARED, linear_idx_type=sfb_smem.linear_idx_type, element_size=sfb_smem.element_size], c_tmem: UInt32, sfa_tmem: UInt32, sfb_tmem: UInt32, init_c: Bool, sfb_tmem_adj: UInt32 = UInt32(0))
TileTensor overload for block-scaled MMA input tiles.
Creates MMA descriptors directly from swizzle parameters.
commit
def commit(self, ptr_mbar: UnsafePointer[address_space=AddressSpace.SHARED])
wait
def wait(self)
Was this page helpful?
Thank you! We'll create more content like this.
Thank you for helping us improve!