std_logic_arith
This is the library that defines some types and basic arithmetic operations for representing integers in standard ways. This is a Synopsys extention. The source code is instd_logic_arith.vhd and is freely redistributable.
The unsigned type
The signed type
The arithmetic functions: +, -, *
The comparison functions: <, <=, >, >=, =, /=
The shift functions: shl, shr
The conv_integer function
The conv_unsigned function
The conv_signed function
The conv_std_logic_vector function
-------------------------------------------------------------------------------------------------------------------------------
The conv_integer function
function conv_integer(arg: integer) return integer;
function conv_integer(arg: unsigned) return integer;
function conv_integer(arg: signed) return integer;
function conv_integer(arg: std_ulogic) return small_int;
These functions convert the arg argument to an integer. If the argument contains any undefined elements, a runtime warning is produced and 0 is returned.
The function provided by the std_logic_arith library can't convert a std_logic_vector to an integer because it is impossible to determine if it represents an unsigned or signed value.Functions that do this are included in the std_logic_unsigned and std_logic_signed libraries.
Examples
signal b : std_logic;
signal u1 : unsigned (3 downto 0);
signal s1 : signed (3 downto 0);
signal i1, i2, i3 : integer;
...
u1 <= "1001";
s1 <= "1001";
b <= 'X';
wait for 10 ns;
i1 <= conv_integer(u1); -- 9
i2 <= conv_integer(s1); -- -7
i3 <= conv_integer(b); -- warning produced in simulator
以上蓝字说明了一切
1.前一个包含在std_logic_arith.all程序包中
2.后一个在std_logic_unsigned 和std_logic_signed 中都有包含
计算机中的运算都是用2进制补码的，本人总结，在做算法时候，包含std_logic_signed 这个包比较好，这样conv_integer 就把需要转换的数据变成了带符号的整数，然而std_logic_unsigned就把需要转换的数据变成了无符号的整数。
3. conv_std_logic_vector 函数的转换结果是将被转换的数据先转换成2进制补码形式，然后取其低“位长”，作为输出。
如：a<=conv_std_logic_vector（-79，6）----（-79）2c=（10110001）
b<=conv_std_logic_vector（-2，6）-----（-2）2c=（11111110）
c<=conv_std_logic_vector（100，6）-----（100）2c=（01100100）
输出结果：a=110001，b=111110，c=100100
4.本人做了这样的转换conv_std_logic_vector （conv_integer （data），16），data是16范围内的有符号数，但是接口是32位的标准矢量形式（std_logic_vector(31 downto 0)刚开始包括的是std_logic_unsigned 包，发现数据变得面目全非了，想来想去，认为conv_integer （data）的原因，就该了下包，结果就正确了。