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 function

gemv_gpu_dispatch

def gemv_gpu_dispatch[transpose_b: Bool = False, elementwise_lambda_fn: Optional[def[dtype: DType, width: SIMDLength, *, alignment: Int = Int(1)](IndexList[Int(2)], SIMD[dtype, width]) capturing thin -> None] = None, pdl_level: PDLLevel = PDLLevel.ON, tile_m: Int = Int(1)](kernel_func: GEMVAlgorithm, c: TileTensor[Engine=c.Engine, address_space=c.address_space, linear_idx_type=c.linear_idx_type], a: TileTensor[Engine=a.Engine, address_space=a.address_space, linear_idx_type=a.linear_idx_type], b: TileTensor[Engine=b.Engine, address_space=b.address_space, linear_idx_type=b.linear_idx_type], ctx: DeviceContext)

Launches the GPU GEMV kernel indicated by kernel_func with appropriate grid and block dims.

Translates a GEMVAlgorithm variant into a concrete kernel call with shape-derived launch parameters, handling input/output layout transformation where needed.

Parameters:

• ​transpose_b (Bool): When True, B is treated as transposed (N, K) row-major.
• ​elementwise_lambda_fn (Optional[def[dtype: DType, width: SIMDLength, *, alignment: Int = Int(1)](IndexList[Int(2)], SIMD[dtype, width]) capturing thin -> None]): Optional epilogue applied element-wise to each output.
• ​pdl_level (PDLLevel): Programmatic dependent launch level.
• ​tile_m (Int): Number of output rows processed per CTA (used by GEMV_SPLIT_K).

Args:

• ​kernel_func (GEMVAlgorithm): The GEMV algorithm variant to launch.
• ​c (TileTensor[Engine=c.Engine, address_space=c.address_space, linear_idx_type=c.linear_idx_type]): Rank-2 output TileTensor.
• ​a (TileTensor[Engine=a.Engine, address_space=a.address_space, linear_idx_type=a.linear_idx_type]): Rank-2 input matrix TileTensor.
• ​b (TileTensor[Engine=b.Engine, address_space=b.address_space, linear_idx_type=b.linear_idx_type]): Rank-2 input vector or matrix TileTensor.
• ​ctx (DeviceContext): Device context for kernel launch.

def gemv_gpu_dispatch[EpilogueFnType: ElementwiseEpilogueFn, //, transpose_b: Bool = False, pdl_level: PDLLevel = PDLLevel.ON, tile_m: Int = Int(1)](kernel_func: GEMVAlgorithm, c: TileTensor[Engine=c.Engine, address_space=c.address_space, linear_idx_type=c.linear_idx_type], a: TileTensor[Engine=a.Engine, address_space=a.address_space, linear_idx_type=a.linear_idx_type], b: TileTensor[Engine=b.Engine, address_space=b.address_space, linear_idx_type=b.linear_idx_type], epilogue_fn: EpilogueFnType, ctx: DeviceContext)

Launches the GEMV kernel indicated by kernel_func, storing through epilogue_fn.

Parameters:

• ​EpilogueFnType (ElementwiseEpilogueFn): Type of epilogue_fn (inferred).
• ​transpose_b (Bool): When True, B is treated as transposed (N, K) row-major.
• ​pdl_level (PDLLevel): Programmatic dependent launch level.
• ​tile_m (Int): Number of output rows processed per CTA (used by GEMV_SPLIT_K).

Args:

• ​kernel_func (GEMVAlgorithm): The GEMV algorithm variant to launch.
• ​c (TileTensor[Engine=c.Engine, address_space=c.address_space, linear_idx_type=c.linear_idx_type]): Rank-2 output TileTensor. Only its shape is read; epilogue_fn stores the result.
• ​a (TileTensor[Engine=a.Engine, address_space=a.address_space, linear_idx_type=a.linear_idx_type]): Rank-2 input matrix TileTensor.
• ​b (TileTensor[Engine=b.Engine, address_space=b.address_space, linear_idx_type=b.linear_idx_type]): Rank-2 input vector or matrix TileTensor.
• ​epilogue_fn (EpilogueFnType): Stores each output element.
• ​ctx (DeviceContext): Device context for kernel launch.