Mojo struct

DecodeSM100PVSS

@register_passable(trivial) struct DecodeSM100PVSS[operand_type: DType, accum_type: DType, *, config: MLA_SM100_Decode_Config]

Implemented traits

AnyType, Copyable, ImplicitlyCopyable, ImplicitlyDestructible, Movable, TrivialRegisterType

`comptime` members

`copyinitis_trivial`

comptime __copyinit__is_trivial = True

`delis_trivial`

comptime __del__is_trivial = True

`moveinitis_trivial`

comptime __moveinit__is_trivial = True

`ALayout`

comptime ALayout = tile_layout_k_major[operand_type, DecodeSM100PVSS[operand_type, accum_type, config=config].BM, DecodeSM100PVSS[operand_type, accum_type, config=config].BK, config.swizzle_mode]()

`BK`

comptime BK = config.BK1

`BLayout`

comptime BLayout = tile_layout_mn_major[operand_type, DecodeSM100PVSS[operand_type, accum_type, config=config].BN, DecodeSM100PVSS[operand_type, accum_type, config=config].BK, config.kv_swizzle_mode]()

`BM`

comptime BM = config.BM

`BN`

comptime BN = DecodeSM100PVSS[operand_type, accum_type, config=config].MMA_N

`MMA_K`

comptime MMA_K = MLA_SM100_Decode_Config.MMA_K

`MMA_M`

comptime MMA_M = config.MMA_M

`MMA_N`

comptime MMA_N = config.MMA_PV_N

`num_k_mmas`

comptime num_k_mmas = (DecodeSM100PVSS[operand_type, accum_type, config=config].BK // 16)

`operand_size`

comptime operand_size = size_of[operand_type]()

`UMMAPVSS`

comptime UMMAPVSS = UMMAInsDescriptor.create[UMMAKind.KIND_F16, accum_type, operand_type, operand_type, Index[dtype=DType.uint32](DecodeSM100PVSS[operand_type, accum_type, config=config].MMA_M, DecodeSM100PVSS[operand_type, accum_type, config=config].MMA_N), transpose_b=False]()

Methods

`descriptor_v_block`

static descriptor_v_block(kv_smem: UnsafePointer[Scalar[operand_type], MutAnyOrigin, address_space=AddressSpace.SHARED]) -> MMASmemDescriptorPair

Returns:

MMASmemDescriptorPair

`descriptor_p_block`

static descriptor_p_block(p_smem: UnsafePointer[Scalar[operand_type], MutAnyOrigin, address_space=AddressSpace.SHARED]) -> MMASmemDescriptorPair

Returns:

MMASmemDescriptorPair

`mma`

static mma[*, stage_idx: Int = 0](a: MMASmemDescriptorPair, b: MMASmemDescriptorPair, c: UInt32, *, c_scale: UInt32, elect: Int32)

Implemented traits​

comptime members​

__copyinit__is_trivial​

__del__is_trivial​

__moveinit__is_trivial​

ALayout​

BK​

BLayout​

BM​

BN​

MMA_K​

MMA_M​

MMA_N​

num_k_mmas​

operand_size​

UMMAPVSS​

Methods​

descriptor_v_block​

descriptor_p_block​

mma​