Often times we find certain pieces of code to be repetitive and called multiple times within the RTL. They mostly do not consume simulation time and might involve complex calculations that need to be done with different data values. In such cases, we can declare a
function and place the repetitive code inside the function and allow it to return the result. This will reduce the amount of lines in the RTL drastically since all you need to do now is to do a function call and pass data on which the computation needs to be performed. In fact, this is very similar to the functions in C.
The purpose of a function is to return a value that is to be used in an expression. A function definition always start with the keyword
function followed by the return type, name and a port list enclosed in parantheses. Verilog knows that a function definition is over when it finds the
endfunction keyword. Note that a function shall have atleast one input declared and the return type will be
void if the function does not return anything.
function [automatic] [return_type] name ([port_list]); [statements] endfunction
automatic will make the function reentrant and items declared within the task are dynamically allocated rather than shared between different invocations of the task. This will be useful for recursive functions and when the same function is executed concurrently by N processes when forked.
There are two ways to declare inputs to a function:
function [7:0] sum; input [7:0] a, b; begin sum = a + b; end endfunction function [7:0] sum (input [7:0] a, b); begin sum = a + b; end endfunction
Returning a value from a function
The function definition will implicitly create an internal variable of the same name as that of the function. Hence it is illegal to declare another variable of the same name inside the scope of the function. The return value is initialized by assigning the function result to the internal variable.
sum = a + b;
Calling a function
A function call is an operand with an expression and has a syntax as shown below.
reg [7:0] result; reg [7:0] a, b; initial begin a = 4; b = 5; #10 result = sum (a, b); end
- A function cannot contain any time-controlled statements like #, @, wait, posedge, negedge
- A function cannot start a task because it may consume simulation time, but can call other functions
- A function should have atleast one input
- A function cannot have non-blocking assignments or
- A function cannot have any triggers
- A function cannot have an output or inout
Functions that call itself are called recursive functions. In the example shown below, a recursive function is written to compute the factorial of a given number.
module tb; initial begin integer result = factorial(4); $display("factorial(4) = %0d", result); end function automatic integer factorial(integer i); integer result = i; // This function is called within the body of this // function with a different argument if (i) begin result = i * factorial(i-1); $display("i=%0d result=%0d", i, result); end else result = 1; return result; endfunction endmodule
xcelium> run i=1 result=1 i=2 result=2 i=3 result=6 i=4 result=24 factorial(4) = 24 xmsim: *W,RNQUIE: Simulation is complete