For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Mojo struct

NullMask

struct NullMask

Mask that's effectively a noop.

Implemented traits

AnyType, Copyable, Deinitable, DevicePassable, ImplicitlyCopyable, MHAMask, Movable, RegisterPassable, TrivialRegisterPassable

comptime members

apply_log2e_after_mask

comptime apply_log2e_after_mask = False

check_mask_during_decoding

comptime check_mask_during_decoding = False

device_type

comptime device_type = NullMask

mask_out_of_bound

comptime mask_out_of_bound = True

mask_safe_out_of_bounds

comptime mask_safe_out_of_bounds = True

Methods

get_type_name

static def get_type_name() -> String

Returns:

String

name

static def name() -> String

Returns:

String

mask

def mask[dtype: DType, width: SIMDLength, //, *, element_type: DType = DType.uint32](self, coord: IndexList[Int(4), element_type=element_type], score_vec: SIMD[dtype, width]) -> SIMD[dtype, width]

Returns:

SIMD[dtype, width]

status

def status[*, element_type: DType = DType.uint32](self, seq_id: UInt32, tile_offset: IndexList[Int(2), element_type=element_type], tile_size: IndexList[Int(2), element_type=element_type]) -> TileMaskStatus

Returns:

TileMaskStatus

start_column

def start_column[BM: Int, BN: Int, page_size: Int](self, seq_id: UInt32, row: UInt32) -> UInt32

Returns:

UInt32

start_column_alignment

static def start_column_alignment[BM: Int, BN: Int, page_size: Int]() -> Int

Returns:

Int

total_iters

def total_iters[BM: Int, BN: Int, page_size: Int](self, seq_id: UInt32, row: UInt32, num_cols: UInt32) -> UInt32

The total number of column iterations for which this mask returns either `TileMaskStatus.NO_MASK' or 'TileMaskStatus.PARTIAL_MASK'.

Parameters:

• ​BM (Int): Query tile height (number of query rows per block).
• ​BN (Int): Key tile width (number of key columns per block).
• ​page_size (Int): The KV cache page size in key columns (0 or 1 if unpaged).

Args:

• ​seq_id (UInt32): The sequence/batch index.
• ​row (UInt32): The starting query-row index of the current query tile.
• ​num_cols (UInt32): The exclusive upper bound on key column indices.

Returns:

UInt32

last_masked_set_end

def last_masked_set_end[BM: Int, BN: Int, page_size: Int](self, seq_id: UInt32, row: UInt32, num_cols: UInt32) -> UInt32

Returns:

UInt32

count_nonfull_sets

static def count_nonfull_sets(BM: Int, BN: Int) -> Int

Returns:

Int

masked_set_ends

def masked_set_ends[BM: Int, BN: Int, page_size: Int](self, seq_id: UInt32, row: UInt32, num_cols: UInt32) -> StaticTuple[UInt32, NullMask.count_nonfull_sets(BM, BN)]

Returns:

StaticTuple[UInt32, NullMask.count_nonfull_sets(BM, BN)]

nonfull_sets

static def nonfull_sets[BM: Int, BN: Int]() -> StaticTuple[TileMaskStatus, NullMask.count_nonfull_sets(BM, BN)]

Returns:

StaticTuple[TileMaskStatus, NullMask.count_nonfull_sets(BM, BN)]

mask_strategies

static def mask_strategies[BM: Int, BN: Int]() -> StaticTuple[MaskStrategy, NullMask.count_nonfull_sets(BM, BN)]

Returns:

StaticTuple[MaskStrategy, NullMask.count_nonfull_sets(BM, BN)]

mask_bits

def mask_bits(self, seq_id: UInt32, score_row: Int32, col_start: Int32, num_keys: Int32) -> UInt32

Returns:

UInt32

sliding_window_size

static def sliding_window_size() -> Int

Returns:

Int