Re: new clang report

On ons, 2011-02-09 at 20:30 +0200, Peter Eisentraut wrote:
> Regression tests (world):
> 
> --- src/test/regress/expected/float8.out
> +++ src/test/regress/results/float8.out
> @@ -384,7 +384,15 @@
>  SELECT '' AS bad, f.f1 * '1e200' from FLOAT8_TBL f;
>  ERROR:  value out of range: overflow
>  SELECT '' AS bad, f.f1 ^ '1e200' from FLOAT8_TBL f;
> -ERROR:  value out of range: overflow
> + bad | ?column? 
> +-----+----------
> +     |        0
> +     |      NaN
> +     |      NaN
> +     |      NaN
> +     |      NaN
> +(5 rows)
> +
>  SELECT 0 ^ 0 + 0 ^ 1 + 0 ^ 0.0 + 0 ^ 0.5;
>   ?column? 
>  ----------

So issue here is actually that clang has an option
      -fmath-errno          Indicate that math functions should be treated as updating errno.

If you pass this option, then the regression tests pass.  If not, you
get the above difference.  So the question is, do we

a) legislate that -fmath-errno is required, or

b) fix dpow() to handle this case somehow (how?), or

c) provide an alternative expected file?