Product, Matrix Multiply

Multiply and divide scalars and nonscalars or multiply and invert matrices

• 

Libraries:
Simulink / Commonly Used Blocks
Simulink / Math Operations
Simulink / Matrix Operations
HDL Coder / Commonly Used Blocks
HDL Coder / HDL Floating Point Operations
HDL Coder / Math Operations

Description

The Product block outputs the result of multiplying two inputs: two scalars, a scalar and a nonscalar, or two nonscalars that have the same dimensions. The default parameter values that specify this behavior are:

• Multiplication: Element-wise(.*)

• Number of inputs: 2

This table shows the output of the Product block for Multiply Inputs of Different Dimensions with the Product Block using default block parameter values.

Inputs and Behavior	Example
Scalar X Scalar Output the product of the two inputs.
Scalar X Nonscalar Output a nonscalar having the same dimensions as the input nonscalar. Each element of the output nonscalar is the product of the input scalar and the corresponding element of the input nonscalar.
Nonscalar X Nonscalar Output a nonscalar having the same dimensions as the inputs. Each element of the output is the product of corresponding elements of the inputs.

The Divide and Product of Elements blocks are variants of the Product block.

• For information on the Divide block, see Divide.

• For information on the Product of Elements block, see Product of Elements.

The Product block (or the Divide block or Product of Elements block, if appropriately configured) can:

• Numerically multiply and divide any number of scalar, vector, or matrix inputs

• Perform matrix multiplication and division on any number of matrix inputs

The Product block performs scalar or matrix multiplication, depending on the value of the Multiplication parameter. The block accepts one or more inputs, depending on the Number of inputs parameter. The Number of inputs parameter also specifies the operation to perform on each input.

The Product block can input any combination of scalars, vectors, and matrices for which the operation to perform has a mathematically defined result. The block performs the specified operations on the inputs, then outputs the result.

The Product block has two modes: Element-wise mode, which processes nonscalar inputs element by element, and Matrix mode, which processes nonscalar inputs as matrices.

Element-Wise Mode

When you set Multiplication to Element-wise(.*), the Product block is in Element-wise mode, in which it operates on the individual numeric elements of any nonscalar inputs. The MATLAB^® equivalent is the .* operator. In element-wise mode, the Product block can perform a variety of multiplication, division, and arithmetic inversion operations.

The value of the Number of inputs parameter controls both how many inputs exist and whether each is multiplied or divided to form the output. When the Product block is in element-wise mode and has only one input, it is functionally equivalent to a Product of Elements block. When the block has multiple inputs, any nonscalar inputs must have identical dimensions, and the block outputs a nonscalar with those dimensions. To calculate the output, the block first expands any scalar input to a nonscalar that has the same dimensions as the nonscalar inputs.

This table shows the output of the Product block for Multiply Inputs of Different Dimensions with the Product Block, using the indicated values for the Number of inputs parameter.

Parameter Values	Examples
Number of inputs: 2
Number of inputs: */
Number of inputs: /**/
Number of inputs:**
Number of inputs: */*

Matrix Mode

When the value of the Multiplication parameter is Matrix(*), the Product block is in Matrix mode, in which it processes nonscalar inputs as matrices. The MATLAB equivalent is the * operator. In Matrix mode, the Product block can invert a single square matrix, or multiply and divide any number of matrices that have dimensions for which the result is mathematically defined.

The value of the Number of inputs parameter controls both how many inputs exist and whether each input matrix is multiplied or divided to form the output. The syntax of Number of inputs is the same as in element-wise mode. The difference between the modes is in the type of multiplication and division that occur.

Interactions Between Block Inputs and Modes

The interactions between the Product block inputs and its Multiplication modes are:

• 1 or * or /

  The block has one input port. In element-wise mode, the block processes the input as described for the Product of Elements block. In matrix mode, if the parameter value is 1 or *, the block outputs the input value. If the value is /, the input must be a square matrix (including a scalar as a degenerate case) and the block outputs the matrix inverse. See Element-Wise Mode and Matrix Mode for more information.

• Integer value > 1

  The block has the number of inputs given by the integer value. The inputs are multiplied together in element-wise mode or matrix mode, as specified by the Multiplication parameter. See Element-Wise Mode and Matrix Mode for more information.

• Unquoted string of two or more * and / characters

  The block has the number of inputs given by the length of the character vector. Each input that corresponds to a * character is multiplied into the output. Each input that corresponds to a / character is divided into the output. The operations occur in element-wise mode or matrix mode, as specified by the Multiplication parameter. See Element-Wise Mode and Matrix Mode for more information.

Expected Differences Between Simulation and Code Generation

For element-wise operations on complex floating-point inputs, simulation and code generation results might differ in near-overflow cases. Although complex numbers is selected and non-finite numbers is not selected on the Code Generation > Interface pane of the Configuration Parameters dialog box, the code generator does not emit special case code for intermediate overflows. This method improves the efficiency of embedded operations for the general case that does not include extreme values. If the inputs could include extreme values, you must manage these cases explicitly.

The generated code might not produce the same pattern of NaN and inf values as simulation when these values are mathematically meaningless. For example, if the simulation output contains a NaN, output from the generated code also contains a NaN, but not necessarily in the same place.

Examples

Multiply and Divide Inputs Using the Product Block

This example shows how to multiply and divide several input signals using the Product block.

Multiply Inputs of Different Dimensions with the Product Block

This example shows how to perform element-wise (.*) multiplication of inputs using the Product block. In this example, the Product block multiplies two scalars, a scalar and a vector, and two 2x2 matrices.

Ports

Input

Port_1 — First input to multiply or divide

scalar | vector | matrix | N-D array

First input to multiply or divide, provided as a scalar, vector, matrix, or N-D array.

Data Types: half | single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | Boolean | fixed point

Port_N — Nth input to multiply or divide

scalar | vector | matrix | N-D array

Nth input to multiply or divide, provided as a scalar, vector, matrix, or N-D array.

Data Types: half | single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | Boolean | fixed point

X — Input signal to multiply

scalar | vector | matrix | N-D array

Input signal to be multiplied with other inputs.

Dependencies

To enable one or more X ports, specify one or more * characters for the Number of inputs parameter and set the Multiplication parameter to Element-wise(.*).

Data Types: half | single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | Boolean | fixed point

÷ — Input signal to divide or invert

scalar | vector | matrix | N-D array

Input signal for division or inversion operations.

Dependencies

To enable one or more ÷ ports, specify one or more / characters for the Number of inputs parameter and set the Multiplication parameter to Element-wise(.*).

Data Types: half | single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | Boolean | fixed point

* — Input signal to multiply

scalar | vector | matrix | N-D array

Input signal to be multiplied with other inputs.

Dependencies

To enable one or more * ports, specify one or more * characters for the Number of inputs parameter and set the Multiplication parameter to Matrix(*).

Data Types: half | single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | Boolean | fixed point

Inv — Input signal to divide or invert

scalar | vector | matrix | N-D array

Input signal for division or inversion operations.

Dependencies

To enable one or more Inv ports, specify one or more / characters for the Number of inputs parameter and set the Multiplication parameter to Matrix(*).

Data Types: half | single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | Boolean | fixed point

Output

Port_1 — Output computed by multiplying, dividing, or inverting inputs

scalar | vector | matrix | N-D array

Output computed by multiplying, dividing, or inverting inputs.

Data Types: half | single | double | int8 | int16 | int32 | int64 | uint8 | uint16 | uint32 | uint64 | Boolean | fixed point

Parameters

Main

Number of inputs — Control number of inputs and type of operation

2 (default) | scalar | * or / for each input port

Control two properties of the block:

• The number of input ports on the block

• Whether each input is multiplied or divided into the output

When you specify:

• 1 or * or /

  The block has one input port. In element-wise mode, the block processes the input as described for the Product of Elements block. In matrix mode, if the parameter value is 1 or *, the block outputs the input value. If the value is /, the input must be a square matrix (including a scalar as a degenerate case) and the block outputs the matrix inverse. See Element-Wise Mode and Matrix Mode for more information.

• Integer value > 1

  The block has the number of inputs given by the integer value. The inputs are multiplied together in element-wise mode or matrix mode, as specified by the Multiplication parameter. See Element-Wise Mode and Matrix Mode for more information.

• Unquoted string of two or more * and / characters

  The block has the number of inputs given by the length of the character vector. Each input that corresponds to a * character is multiplied into the output. Each input that corresponds to a / character is divided into the output. The operations occur in element-wise mode or matrix mode, as specified by the Multiplication parameter. See Element-Wise Mode and Matrix Mode for more information.

Programmatic Use

Block Parameter: Inputs

Type: character vector

Values: '2' | '**' | '*/' | '*/*' | ...

Default: '2'

Multiplication — Element-wise (.*) or Matrix (*) multiplication

Element-wise(.*) (default) | Matrix(*)

Specify whether the block performs Element-wise(.*) or Matrix(*) multiplication.

Programmatic Use

Block Parameter: Multiplication

Type: character vector

Values: 'Element-wise(.*)' | 'Matrix(*)'

Default: 'Element-wise(.*)'

Apply over — How to apply function along specified dimensions

All dimensions (default) | Specified dimension

Specify how to apply the function along specified dimensions.

• All dimensions — Apply function for all input values for all dimensions.

• Specified dimension — Apply function for all input values for specified dimension.

For example, in this model, Multiplication is set to Element-wise(.*), and Apply over is set to All dimensions. The block returns the product of all values from all dimensions.

Dependencies

To enable this parameter, set Number of inputs to * and Multiplication to Element-wise (.*).

Programmatic Use

Block Parameter: CollapseMode

Type: character vector

Values: 'All dimensions' | 'Specified dimension'

Default: 'All dimensions'

Dimension — Dimension along which to multiply

1 (default) | positive integer

Specify the dimension along which to multiply, as a positive integer. For example, for a 2-D matrix, 1 applies the function to each column, and 2 applies the function to each row.

For example, in this model, Multiplication is set to Element-wise(.*), Apply over is set to Specified dimension, and Dimension is set to 2. The block returns the product of all values from each row.

Dependencies

To enable this parameter:

• Set Number of inputs to *

• Set Multiplication to Element-wise (.*)

• Set Apply over to Specified dimension

Programmatic Use

Block Parameter: CollapseDim

Type: character vector

Values: '1' | '2' | ...

Default: '1'

Sample time (-1 for inherited) — Interval between samples

-1 (default) | scalar | vector

Specify the time interval between samples. To inherit the sample time, set this parameter to -1. For more information, see Specify Sample Time.

Dependencies

This parameter is visible only if you set it to a value other than -1. To learn more, see Blocks for Which Sample Time Is Not Recommended.

Programmatic Use

To set the block parameter value programmatically, use the set_param function.

Parameter:	SampleTime
Values:	"-1" (default) | scalar or vector in quotes

Signal Attributes

Require all inputs to have the same data type — Require that all inputs have the same data type

off (default) | on

Specify if input signals must all have the same data type. If you enable this parameter, then an error occurs during simulation if the input signal types are different.

Programmatic Use

Block Parameter: InputSameDT

Type: character vector

Values: 'off' | 'on'

Default: 'off'

Output minimum — Minimum output value for range checking

[] (default) | scalar

Lower value of the output range that the software checks.

The software uses the minimum to perform:

• Parameter range checking (see Specify Minimum and Maximum Values for Block Parameters) for some blocks.

• Simulation range checking (see Specify Signal Ranges and Enable Simulation Range Checking).

• Automatic scaling of fixed-point data types.

• Optimization of the code that you generate from the model. This optimization can remove algorithmic code and affect the results of some simulation modes such as SIL or external mode. For more information, see Optimize using the specified minimum and maximum values (Embedded Coder).

Tips

Output minimum does not saturate or clip the actual output signal. Use the Saturation block instead.

Programmatic Use

To set the block parameter value programmatically, use the set_param function.

Parameter:	OutMin
Values:	'[]' (default) | scalar in quotes

Output maximum — Maximum output value for range checking

[] (default) | scalar

Upper value of the output range that the software checks.

The software uses the maximum value to perform:

• Parameter range checking (see Specify Minimum and Maximum Values for Block Parameters) for some blocks.

• Simulation range checking (see Specify Signal Ranges and Enable Simulation Range Checking).

• Automatic scaling of fixed-point data types.

• Optimization of the code that you generate from the model. This optimization can remove algorithmic code and affect the results of some simulation modes such as SIL or external mode. For more information, see Optimize using the specified minimum and maximum values (Embedded Coder).

Tips

Output maximum does not saturate or clip the actual output signal. Use the Saturation block instead.

Programmatic Use

To set the block parameter value programmatically, use the set_param function.

Parameter:	OutMax
Values:	'[]' (default) | scalar in quotes

Output data type — Specify the output data type

Inherit: Inherit via internal rule (default) | Inherit: Keep MSB | Inherit: Match scaling | Inherit: Inherit via back propagation | Inherit: Same as first input | double | single | half | int8 | uint8 | int16 | uint16 | int32 | uint32 | int64 | uint64 | fixdt(1,16) | fixdt(1,16,0) | fixdt(1,16,2^0,0) | <data type expression>

Choose the data type for the output. The type can be inherited, specified directly, or expressed as a data type object such as Simulink.NumericType. For more information, see Control Data Types of Signals.

When you select an inherited option, the block behaves as follows:

• Inherit: Inherit via internal rule — Simulink^® chooses a data type to balance numerical accuracy, performance, and generated code size, while taking into account the properties of the embedded target hardware. If you change the embedded target settings, the data type selected by the internal rule might change. For example, if the block multiplies an input of type int8 by a gain of int16 and ASIC/FPGA is specified as the targeted hardware type, the output data type is sfix24. If Unspecified (assume 32-bit Generic), in other words, a generic 32-bit microprocessor, is specified as the target hardware, the output data type is int32. If none of the word lengths provided by the target microprocessor can accommodate the output range, Simulink software displays an error in the Diagnostic Viewer.

• Inherit: Keep MSB– Simulink chooses a data type that maintains the full range of the operation, then reduces the precision of the output to a size appropriate for the embedded target hardware.

  Tip

  For more efficient generated code, clear the Saturate on integer overflow parameter.

  This rule never produces overflows.

• Inherit: Match scaling– Simulink chooses a data type whose scaling matches the scaling of the input types. If the full range of the type does not fit on the embedded target hardware, the range is reduced yielding a type appropriate for the embedded target hardware. This rule can produce overflows. This rule does not support multiplication between complex signals

  The Inherit: Keep MSB and Inherit: Match scaling rules do not support multiplication between complex signals or signals with non-zero bias. The rules support only multiplication and division ('**', '*/', '/*') between two inputs, matrix multiplication of two inputs, and collapsing product of two elements of a vector.

  It is not always possible for the software to optimize code efficiency and numerical accuracy at the same time. If the internal rule doesn’t meet your specific needs for numerical accuracy or performance, use one of the following options:

  • Specify the output data type explicitly.

  • Use the simple choice of Inherit: Same as input.

  • Explicitly specify a default data type such as fixdt(1,32,16) and then use the Fixed-Point Tool to propose data types for your model. For more information, see fxptdlg (Fixed-Point Designer).

  • To specify your own inheritance rule, use Inherit: Inherit via back propagation and then use a Data Type Propagation block. Examples of how to use this block are available in the Signal Attributes library Data Type Propagation Examples block.

• Inherit: Inherit via back propagation — Use data type of the driving block.

• Inherit: Same as first input — Use data type of first input signal.

Dependencies

When input is a floating-point data type smaller than single precision, the Inherit: Inherit via internal rule output data type depends on the setting of the Inherit floating-point output type smaller than single precision configuration parameter. Data types are smaller than single precision when the number of bits needed to encode the data type is less than the 32 bits needed to encode the single-precision data type. For example, half and int16 are smaller than single precision.

Programmatic Use

Block Parameter: OutDataTypeStr

Type: character vector

Values: 'Inherit: Inherit via internal rule | 'Inherit: Keep MSB'| 'Inherit: Match scaling'| 'Inherit: Same as first input' | 'Inherit: Inherit via back propagation' | 'double' | 'single' | 'half' | 'int8' | 'uint8' | 'int16' | 'uint16' | 'int32' | 'uint32' | 'uint64'|'int64'|'fixdt(1,16)' | 'fixdt(1,16,0)' | 'fixdt(1,16,2^0,0)' | '<data type expression>'

Default: 'Inherit: Inherit via internal rule'

Lock output data type setting against changes by the fixed-point tools — Option to prevent fixed-point tools from overriding Output data type

off (default) | on

Select this parameter to prevent the fixed-point tools from overriding the Output data type you specify on the block. For more information, see Use Lock Output Data Type Setting (Fixed-Point Designer).

Programmatic Use

To set the block parameter value programmatically, use the set_param function.

Parameter:	LockScale
Values:	'off' (default) | 'on'

Integer rounding mode — Rounding mode for fixed-point operations

Floor (default) | Ceiling | Convergent | Nearest | Round | Simplest | Zero

Select the rounding mode for fixed-point operations. You can select:

Ceiling

Rounds positive and negative numbers toward positive infinity. Equivalent to the MATLAB ceil function.

Convergent

Rounds number to the nearest representable value. If a tie occurs, rounds to the nearest even integer. Equivalent to the Fixed-Point Designer™ convergent function.

Floor

Rounds positive and negative numbers toward negative infinity. Equivalent to the MATLAB floor function.

Nearest

Rounds number to the nearest representable value. If a tie occurs, rounds toward positive infinity. Equivalent to the Fixed-Point Designer nearest function.

Round

Rounds number to the nearest representable value. If a tie occurs, rounds positive numbers toward positive infinity and rounds negative numbers toward negative infinity. Equivalent to the Fixed-Point Designer round function.

Simplest

Chooses between rounding toward floor and rounding toward zero to generate rounding code that is as efficient as possible.

Zero

Rounds number toward zero. Equivalent to the MATLAB fix function.

For more information, see Rounding Modes (Fixed-Point Designer).

Block parameters always round to the nearest representable value. To control the rounding of a block parameter, enter an expression using a MATLAB rounding function into the mask field.

Programmatic Use

Block Parameter: RndMeth

Type: character vector

Values: 'Ceiling' | 'Convergent' | 'Floor' | 'Nearest' | 'Round' | 'Simplest' | 'Zero'

Default: 'Floor'

Saturate on integer overflow — Method of overflow action

off (default) | on

Specify whether overflows saturate or wrap.

• on — Overflows saturate to either the minimum or maximum value that the data type can represent.

• off — Overflows wrap to the appropriate value that the data type can represent.

For example, the maximum value that the signed 8-bit integer int8 can represent is 127. Any block operation result greater than this maximum value causes overflow of the 8-bit integer.

• With this parameter selected, the block output saturates at 127. Similarly, the block output saturates at a minimum output value of -128.

• With this parameter cleared, the software interprets the overflow-causing value as int8, which can produce an unintended result. For example, a block result of 130 (binary 1000 0010) expressed as int8 is -126.

Tips

• Consider selecting this parameter when your model has a possible overflow and you want explicit saturation protection in the generated code.

• Consider clearing this parameter when you want to optimize efficiency of your generated code. Clearing this parameter also helps you to avoid overspecifying how a block handles out-of-range signals. For more information, see Troubleshoot Signal Range Errors.

• When you select this parameter, saturation applies to every internal operation on the block, not just the output or result.

• In general, the code generation process can detect when overflow is not possible. In this case, the code generator does not produce saturation code.

Programmatic Use

To set the block parameter value programmatically, use the set_param function.

Parameter:	SaturateOnIntegerOverflow
Values:	'off' (default) | 'on'

Mode — Select data type mode

Inherit (default) | Built in | Fixed Point

Select the category of data to specify.

• Inherit — Inheritance rules for data types. Selecting Inherit enables a second menu/text box to the right where you can select the inheritance mode.

• Built in — Built-in data types. Selecting Built in enables a second menu/text box to the right where you can select a built-in data type.

• Fixed point — Fixed-point data types. Selecting Fixed point enables additional parameters that you can use to specify a fixed-point data type.

• Expression — Expressions that evaluate to data types. Selecting Expression enables a second menu/text box to the right, where you can enter the expression.

For more information, see Specify Data Types Using Data Type Assistant.

Dependencies

To enable this parameter, click the Show data type assistant button.

Data type override — Specify data type override mode for this signal

Inherit | Off

Select the data type override mode for this signal.

• When you select Inherit, Simulink inherits the data type override setting from its context, that is, from the block, Simulink.Signal object or Stateflow^® chart in Simulink that is using the signal.

• When you select Off, Simulink ignores the data type override setting of its context and uses the fixed-point data type specified for the signal.

For more information, see Specify Data Types Using Data Type Assistant in the Simulink documentation.

Dependencies

To enable this parameter, set Mode to Built in or Fixed point.

Tips

The ability to turn off data type override for an individual data type provides greater control over the data types in your model when you apply data type override. For example, you can use this option to ensure that data types meet the requirements of downstream blocks regardless of the data type override setting.

Signedness — Specify signed or unsigned

Signed (default) | Unsigned

Specify whether the fixed-point data is signed or unsigned. Signed data can represent positive and negative values, but unsigned data represents positive values only.

• Signed, specifies the fixed-point data as signed.

• Unsigned, specifies the fixed-point data as unsigned.

For more information, see Specify Data Types Using Data Type Assistant.

Dependencies

To enable this parameter, set the Mode to Fixed point.

Word length — Bit size of the word that holds the quantized integer

16 (default) | integer from 0 to 32

Specify the bit size of the word that holds the quantized integer. For more information, see Specifying a Fixed-Point Data Type.

Dependencies

To enable this parameter, set Mode to Fixed point.

Fraction length — Specify fraction length for fixed-point data type

0 (default) | scalar integer

Specify fraction length for fixed-point data type as a positive or negative integer. For more information, see Specifying a Fixed-Point Data Type.

Dependencies

To enable this parameter, set Scaling to Binary point.

Scaling — Method for scaling fixed-point data

Best precision (default) | Binary point | Slope and bias

Specify the method for scaling your fixed-point data to avoid overflow conditions and minimize quantization errors. For more information, see Specifying a Fixed-Point Data Type.

Dependencies

To enable this parameter, set Mode to Fixed point.

Slope — Specify slope for the fixed-point data type

2^0 (default) | positive, real-valued scalar

Specify slope for the fixed-point data type. For more information, see Specifying a Fixed-Point Data Type.

Dependencies

To enable this parameter, set Scaling to Slope and bias.

Bias — Specify bias for the fixed-point data type

0 (default) | real-valued scalar

Specify bias for the fixed-point data type as any real number. For more information, see Specifying a Fixed-Point Data Type.

Dependencies

To enable this parameter, set Scaling to Slope and bias.

Block Characteristics

Data Types	Boolean | double | fixed point | half | integer | single
Direct Feedthrough	yes
Multidimensional Signals	yes
Variable-Size Signals	yes
Zero-Crossing Detection	no

Extended Capabilities

C/C++ Code Generation

Generate C and C++ code using Simulink® Coder™.

You can generate Halide code if you have Embedded Coder^® license. For more information, see Speed Up Generated Code Execution with Halide Code (Embedded Coder).

HDL Code Generation

Generate VHDL, Verilog and SystemVerilog code for FPGA and ASIC designs using HDL Coder™.

HDL Coder™ provides additional configuration options that affect HDL implementation and synthesized logic.

HDL Architecture

The default Linear implementation generates a chain of N operations (multipliers) for N inputs.

Architecture	Parameters	Description
Linear(default)	None	Generate a product operator * in the generated code.
ShiftAdd	LatencyStrategy	Perform product operations on fixed-point types by using multiple shift and add operations. The block has pipelined implementation that introduces additional latency in the generated code. ShiftAdd architecture maps to efficient circuits in ASIC applications. This architecture is useful in applications on FPGA target boards that do not have DSP units. The MAX latency for different word lengths is calculated by using this formula:ceil(log2(min(in1WL,in2WL))). in1WL is the first input word length and in2WL is the second input word length. For example, if both inputs are of word length four the MAX latency is two. When you use fixed-point data types, the word length of the inputs must be less than 63.

HDL Block Properties

If you use the block in matrix multiplication mode, you can specify the DotProductStrategy. This setting determines whether you want to implement the matrix multiplication by using a tree of adders and multipliers, or use the Multiply-Accumulate block implementation. The default is Fully Parallel.

Note

The DotProductStrategy must be set to Fully Parallel when you use the Native Floating Point mode.

For more information, see DotProductStrategy (HDL Coder).

See also Design Considerations for Matrices and Vectors (HDL Coder).

General
ConstrainedOutputPipeline	Number of registers to place at the outputs by moving existing delays within your design. Distributed pipelining does not redistribute these registers. The default is 0. For more details, see ConstrainedOutputPipeline (HDL Coder).
DSPStyle	Synthesis attributes for multiplier mapping. To use this property, set HDL architecture to Linear. The default is none. See also DSPStyle (HDL Coder).
InputPipeline	Number of input pipeline stages to insert in the generated code. Distributed pipelining and constrained output pipelining can move these registers. The default is 0. For more details, see InputPipeline (HDL Coder).
OutputPipeline	Number of output pipeline stages to insert in the generated code. Distributed pipelining and constrained output pipelining can move these registers. The default is 0. For more details, see OutputPipeline (HDL Coder).
LatencyStrategy	To enable this property, set HDL architecture to ShiftAdd. Specify whether to map the blocks in your design to MAX, Min, CUSTOM, or ZERO latency for fixed-point and floating-point types. The default is MAX. See also LatencyStrategy (HDL Coder).
CustomLatency	To enable this property, set HDL architecture to ShiftAdd. When LatencyStrategy is set to CUSTOM, use this property to specify a custom latency value between ZERO and MAX for fixed-point types. See also LatencyStrategy (HDL Coder).

Native Floating Point
HandleDenormals	Specify whether you want HDL Coder to insert additional logic to handle denormal numbers in your design. Denormal numbers are numbers that have magnitudes less than the smallest floating-point number that can be represented without leading zeros in the mantissa. The default is inherit. See also HandleDenormals (HDL Coder).
LatencyStrategy	Specify whether to map the blocks in your design to inherit, Max, Min, Zero, or Custom for the floating-point operator. The default is inherit. See also LatencyStrategy (HDL Coder).
NFPCustomLatency	To specify a value, set LatencyStrategy to Custom. HDL Coder adds latency equal to the value that you specify for the NFPCustomLatency setting. See also NFPCustomLatency (HDL Coder).
MantissaMultiplyStrategy	Specify how to implement the mantissa multiplication operation during code generation. By using different settings, you can control the DSP usage on the target FPGA device. The default is inherit. See also MantissaMultiplyStrategy (HDL Coder).

Complex Data Support

The default (linear) implementation supports complex data.

Complex division is not supported. For block implementations of the Product block in divide mode or reciprocal mode, see HDL Code Generation on the Divide block reference page.

Optimization Support

You can apply the sharing or streaming optimizations for Product, Divide and Reciprocal blocks using ShiftAdd architecture.

Block (Input)	Resource Sharing	Streaming
Product(*)	Yes	Yes
Divide(/* or */)	Yes	Yes
Reciprocal (/)	Yes	No

When you use sharing optimization on these blocks, make sure to enclose the block in the atomic subsystem and set the Default parameter behavior configuration parameter to Inlined.

For more information, see Resource Sharing (HDL Coder) and Streaming (HDL Coder).

Restrictions

• HDL code generation does not support more than two inputs for the block.

• Product block with /* for Number of inputs block parameter performs a division where the second input is divided by the first input. This mode has the same restrictions that apply to the Divide block. See HDL Code Generation.

PLC Code Generation
Generate Structured Text code using Simulink® PLC Coder™.

Fixed-Point Conversion
Design and simulate fixed-point systems using Fixed-Point Designer™.

Version History

Introduced before R2006a

R2024a: Product and Matrix Multiply Blocks Have New Parameter Name

The Multiply over parameter has been renamed to Apply over for the Product and Matrix Multiply blocks.