Guidelines for Supported Blocks and Data Types - By Numbered List - MATLAB & Simulink

Guidelines for Supported Blocks and Data Types - By Numbered List

The HDL modeling guidelines are a set of recommended guidelines that you can follow when creating Simulink^® model for code generation with HDL Coder™. The guidelines for supported blocks and data types consist of guidelines for using various blocks in the HDL Coder block library, and about the supported data types. Each modeling guideline for HDL code generation has a different level of severity that indicates the levels of compliance requirements. To learn more about these severity levels, see HDL Modeling Guidelines Severity Levels.

These tables list the guidelines for supported data types in HDL Coder and for various blocks in the HDL Coder block library. The guidelines start from 2.1 and are divided into subsections. In the table, you see that certain guidelines have an associated model check. You can follow the modeling pattern recommended for these guidelines by running that check in the HDL Code Advisor. To learn more about the HDL Code Advisor, see Check HDL Compatibility of Simulink Model Using HDL Code Advisor.

Guidelines 2.1: HDL RAMs and HDL Operations Library

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.1.1	RAM Block Access Considerations	Recommended	None
2.1.2	Serial to Parallel Conversion	Recommended	None

Guidelines 2.2: Logic and Bit Operations Library

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.2.1	Logical and Arithmetic Bit Shift Operations	Informative	None
2.2.2	Usage of Logical Operator, Bitwise Operator, and Bit Reduce Blocks	Informative	None
2.2.3	Use Boolean Output for Compare to Constant and Relational Operator Blocks	Strongly Recommended	None

Guidelines 2.3: Lookup Table and Signal Routing Blocks

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.3.1	Generate FPGA Block RAM from Lookup Tables	Strongly Recommended	None
2.3.2	Recommended Block Parameter Settings of Multiport Switch Block for Numeric and Enumerated Types	Recommended	None
2.3.3	Guidelines for Using Selector Blocks to Extract Input Elements from Vector or Matrix Signals	Recommended	None
2.3.4	Guidelines for Using Assignment Blocks to Write Elements in Vectors, Matrices, and 3-D Arrays	Recommended	None

Guidelines 2.4: Ports and Subsystems

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.4.1	Virtual Subsystem: Use as DUT	Mandatory	Model Check: Check for invalid top level subsystem
2.4.2	Atomic and Virtual Subsystems: Generate Reusable HDL Files	Recommended	None
2.4.3	Variant Subsystem: Using Variant Subsystems for HDL Code Generation	Mandatory	None
2.4.4	Model References: Build Model Design Using Smaller Partitions	Recommended	None
2.4.5	Block Settings of Enabled and Triggered Subsystems	Mandatory	None

Guideline 2.5: Rate Change and Constant Blocks

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.5.1	Usage of Rate Conversion Blocks	Recommended	None
2.5.2	Use Discrete and Finite Sample Time for Constant Block	Recommended	Model Check: Check for infinite and continuous sample time sources
2.5.3	Dynamically Change Sample Offset for Downsample Block	Informative	None

Guideline 2.6: Delay Blocks

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.6.1	Appropriate Usage of Delay Blocks as Registers	Recommended	Check for obsolete Unit Delay Enabled/Resettable Blocks
2.6.2	Absorb Delays to Avoid Timing Difference	Recommended	None
2.6.3	Map Large Delays to Block RAM	Recommended	None
2.6.4	Required HDL Settings for Goto and From Blocks	Mandatory	None

Guideline 2.7: Multiplication and Accumulation Operations

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.7.1	Designing Multipliers and Adders for Efficient Mapping to DSP Blocks on FPGA	Strongly Recommended	None
2.7.2	Set ConstMultiplierOptimization HDL Block Property to auto for Gain Block	Recommended	None
2.7.3	Use ShiftAdd Architecture of Divide Block for Fixed-Point Types	Recommended	None
2.7.4	Use Gain Block for Fixed-Point Constant Operations	Strongly Recommended	None

Guideline 2.8: MATLAB Function Blocks

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.8.1	Update Persistent Variables at End of MATLAB Function	Strongly Recommended	None
2.8.2	Avoid Algebraic Loop Errors from Persistent Variables inside MATLAB Function Blocks	Mandatory	None
2.8.3	Use hdlfimath Setting and Specify fi Objects inside MATLAB Function Block	Strongly Recommended	Check for MATLAB Function block settings

Guideline 2.9: Stateflow Charts

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.9.1	Choose State Machine Type based on HDL Implementation Requirements	Strongly Recommended	None
2.9.2	Specify Block Configuration Settings of Stateflow Chart	Strongly Recommended	Check for Stateflow chart settings
2.9.3	Insert Unconditional Transition State for Else Statement in HDL Code	Recommended	None
2.9.4	Data Type Settings and Casting in Stateflow Chart for HDL Code Generation	Informative	None
2.9.5	Using Absolute Time Temporal Logic in Stateflow Charts	Mandatory	None
2.9.6	Modeling Error (default) State in Stateflow Charts	Informative	None
2.9.7	Enable Clock-Driven Outputs of Stateflow Charts (Moore Charts Only)	Informative	None
2.9.8	Enumeration type for active state monitoring in a Stateflow chart with no default value	Informative	None

Guidelines 2.10: Data Types

Guideline ID	Title	Severity	Associated Model Check/Coding Standard Rule
2.10.1	Use Boolean for Logical Data and Ufix1 for Numerical Data	Mandatory	None
2.10.2	Specify Data Type of Gain Blocks	Recommended	None
2.10.3	Enumerated Data Type Restrictions	Mandatory	None
2.10.4	Choose Optimal Simulink Block to Compute Sine and Cosine Functions with Fixed-Point Data Types	Recommended	None
2.10.5	Choose Optimal Simulink Block to Compute Sine and Cosine Functions with Floating-Point Data Types	Recommended	None
2.10.6	Guidelines for Using Rounding and Saturation Settings for Fixed-Point Data Types	Recommended	None