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

GEMVAlgorithm

struct GEMVAlgorithm

Enumerates the GEMV kernel algorithm variants used by the GPU dispatcher.

Each variant targets a distinct operand shape, memory access pattern, or hardware capability. The dispatcher in gemv_gpu selects among these based on runtime shape information and target architecture.

Implemented traits

AnyType, Copyable, Deinitable, EnumLike, Equatable, Hashable, ImplicitlyCopyable, Movable, RegisterPassable, TrivialRegisterPassable, Writable

comptime members

GemmMmaCpasync

comptime GemmMmaCpasync = GEMVAlgorithm(Int(6))

GemvKernel

comptime GemvKernel = GEMVAlgorithm(Int(0))

GemvKernelVector

comptime GemvKernelVector = GEMVAlgorithm(Int(1))

GemvSplitK

comptime GemvSplitK = GEMVAlgorithm(Int(2))

GevmKernel

comptime GevmKernel = GEMVAlgorithm(Int(4))

GevmKernelVector

comptime GevmKernelVector = GEMVAlgorithm(Int(3))

MatmulNaive

comptime MatmulNaive = GEMVAlgorithm(Int(5))

Methods

__int__

def __int__(self) -> Int

Returns:

Int

write_to

def write_to(self, mut writer: T)