This code is included when there are fixed-point data types in the model and fixed-point word size checks are enabled. The limit for each type is the max and min values that can be represented by the type. The header file is included to perform some checks on the target (e.g., because fixpoint slope/bias can cause overflow).
The recommended workflow is to set the correct values of the bits in the "Configuration Parameters" -> "Hardware Implementation" -> Device Details section based on the values defined by the compiler in the limits.h file. You can avoid these checks completely in the following way if the customer does not want these checks:
From MATLAB R2024a, this option is available directly as checkbox in 'Configuration Parameters' -> 'Code Generation' -> 'Interface' -> 'Advanced parameters' -> 'Suppress generation of fixed-point word size checks'
Suppress generation of fixed-point word size checks: https://www.mathworks.com/help/releases/R2024a/rtw/ref/suppressgenerationoffixedpointwordsizechecks.html
From MATLAB R2017a onwards you can add a TLC option which will ensure that the Fixed Point Word Size checking code will not be generated. To use this option, navigate to 'Configuration Parameters' -> 'All Parameters' tab. Then search for "TLC Options" and add the following text to the TLC options check box "-aRemoveFixptWordSizeChecks=1".
Configure Generated Code with TLC: https://www.mathworks.com/help/releases/R2024a/ecoder/ug/configuring-generated-code-with-tlc.html
For older versions of MATLAB (before R2017a), you can disable these #define checks by setting the 'TargetPreprocMaxBitsSint' and 'TargetPreprocMaxBitsUint' to a value of '0'. To do this, navigate to 'Configuration Parameters' and click on the 'List' tab on the top of the dialog box. Then search for the above two parameters and set them to '0'. If we generate code then the #define checks will be disabled by #if 0 statements encapsulating the above statements.
