Matrix Vector Multiply Configuration Parameters - 2025.1 English

For the Matrix Vector Multiply (GeMV) library element, use the following list of configurable parameters and default values.

Table 103 Matrix Vector Multiply Configuration Parameters

Name	Type	Default	Description
DATA_A	typename	cint16	Input Matrix A Data Type.
DATA_B	typename	cint16	Input Vector B Data Type.
DIM_A	unsigned	16	Input Matrix A Dimension (number of matrix rows).
DIM_B	unsigned	16	Input Vector B Dimension (number of matrix columns).
SHIFT	unsigned	16	See Common Configuration Parameters
ROUND_MODE	unsigned	0	See Common Configuration Parameters
UUT_SSR	unsigned	1	Super Sample Rate.
CASC_LEN	unsigned	1	Cascade length.
NUM_FRAMES	unsigned	1	The number of batches of input data that will be processed per iteration.
DIM_A_LEADING	unsigned	1	COL_MAJOR = 1 ROW MAJOR = 0
USE_MATRIX_RELOAD	unsigned	0	0 - Matrix A via iobuffer 1 - reloadable Matrix A via RTP ports
API_IO	unsigned	0	0 - Vector B via iobuffer 1 - Vector B via stream, only when reloadable Matrix A via RTP port
DUAL_IP	unsigned	0	0 - single vector input per kernel 1 - dual vector inputs per kernel
NUM_OUTPUTS	unsigned	1	1 - single output per SSR rank 2 - dual outputs per SSR rank, stream only
NITER	unsigned	16	See Common Configuration Parameters
DIFF_TOLERANCE	unsigned	0	See Common Configuration Parameters
STIM_TYPE_A	unsigned	0	See STIM_TYPE in Common Configuration Parameters
STIM_TYPE_B	unsigned	0	See STIM_TYPE in Common Configuration Parameters
SAT_MODE	unsigned	1	See Common Configuration Parameters