🚀 go-pugleaf

This problem is in printf, not in the compiler.  Sun's printf does
not know about long doubles.  You can change
    printf("%.15Lg\n", myvar);
to
    printf("%.15g\n", (double) myvar);
or you can use the function e64toasc in real.c for printing -- link real.o to
your program.


>As near as I can tell from the doc, the 'long double' extended floating point
>format is supported by gcc-2.4 on 68000-based machines that have a 68881 fpa.  
>On a sun3 which meets the above criteria, I compiled and ran the following
>program with gcc-2.4.2:

>main()
>{
>long double myvar;
>myvar = 1.0 + 1.0e-14;
>printf("%.15Lg\n", myvar);
>}

>And I got the following output:
>1.00000762939453

>Whereas if I replace the long double with a double and the L with an l,
>it gives the expected output:
>1.00000000000001 

>Is this a known bug?  Is there a fix?

>--
>Mark J. Reed			<mark@cad.gatech.edu>		
>Georgia Institute of Technology