Re: AC_C_LONG_DOUBLE is wrong on IRIX 5.3

[Oliver Kiddle]

Paul Eggert wrote:
> 
> But isn't this the correct answer on Irix?  long double is supported
> and is equivalent to 'double'.  The C standard allows 'long double'
> to be equivalent to 'double'.

That depends on your interpretation of the warning. The warning states
that long double is "unsupported". They have just substituted double
instead of generating an error. I suspect that problems may occur in
programs if there are any other things they assume will then work like
any library/system calls with long double parameters or '%Lf' formats
to printf.

> The compiler warnings are admittedly annoying, but you can ignore them.

> > -  exit (sizeof (long double) < sizeof (double));
> > +  exit (sizeof (long double) <= sizeof (double));
> 
> This patch isn't correct, as the two sizes can be equal on conforming
> implementations.

If the sizes are equal, then it isn't particularly useful for a program
to be using long double instead of double. It is already ruling out
smaller long doubles for Ultrix 4.3. I think it would be more useful if
the check was redefined as `checking for a useful long double type'.
Any program using that macro is going to be looking to see if there is
a more precise double variable it can be using so I think my patch is
right, it should just be combined with rewording the description of the
macro's purpose.

Oliver Kiddle

_____________________________________________________________________
This message has been checked for all known viruses by the 
MessageLabs Virus Scanning Service. For further information visit
http://www.messagelabs.com/stats.asp