Tom Lane wrote:
> Roman Kononov <[EMAIL PROTECTED]> writes:
> > In float4mul() and float4div(), the computation should be double precision.
> Why?  It's going to have to fit in a float4 eventually anyway.

One issue is in the patch comment:

        !    *  Computations that slightly exceed FLOAT8_MAX are non-Infinity,
        !    *  but those that greatly exceed FLOAT8_MAX become Infinity.  
        !    *  it is difficult to tell if a value is really infinity or the 
        !    *  of an overflow.  The solution is to use a boolean indicating if
        !    *  the input arguments were infiity, meaning an infinite result is
        !    *  probably not the result of an overflow.  This allows various
        !    *  computations like SELECT 'Inf'::float8 + 5.

        +    *  Underflow has similar issues to overflow, i.e. if a computation 
        +    *  slighly smaller than FLOAT8_MIN, the result is non-zero, but if 
it is
        +    *  much smaller than FLOAT8_MIN, the value becomes zero.  However,
        +    *  unlike overflow, zero is not a special value and can be the 
        +    *  of a computation, so there is no easy way to pass a boolean
        +    *  indicating whether a zero result is reasonable or not.  It might
        +    *  be possible for multiplication and division, but because of 
        +    *  such tests would probably not be reliable.

For overflow, it doesn't matter, but by using float8, you have a much
larger range until you underflow to zero.  I will make adjustments to
the patch to use this, and add comments explaining its purpose.

  Bruce Momjian   [EMAIL PROTECTED]

  + If your life is a hard drive, Christ can be your backup. +

---------------------------(end of broadcast)---------------------------
TIP 5: don't forget to increase your free space map settings

Reply via email to