Re: AC_C_LONG_DOUBLE is obsolescent.

[Bob Friesenhahn]

On Wed, 9 Apr 2008, Eric Blake wrote:
> According to Bob Friesenhahn on 4/9/2008 12:35 PM:
> | I am confused since I don't see how the usefullness of this test can be
> | obsolescent.
>
> The test was merely whether compilers support the type 'long double'; as
> all modern compilers obey this part of the C standard, it is no longer
> worth checking whether long double is syntactically valid, and rather just
> assuming that it is.

I always assumed that the test worked as described and verified that 
'long double' has more range than 'double'.  It seemed to be doing 
more than ensure that the compiler does not choke when requested to 
provide 'long double'.  My application only uses 'long double' for 
cumulative addition so it does not require much support from the math 
library.  I have yet to encounter a problem with a system which passes 
the long double test.

You have now used the term "modern compilers" which I assume means 
C'99 rather than C'89?

> Yes, current systems is intended to mean any system that an autoconf
> configure script will run on.  It excludes platforms like Solaris 4, which
> are no longer supported by the vendor.

Solaris 8 is the oldest version of Solaris supported by Sun, however, 
I see no reason why Solaris 2.1 should not be able to execute current 
autoconf.  Perhaps it might need to be augmented with a newer shell 
and a few updated utilities.  The founding principle of autoconf is 
"feature based testing" and I hope that this principle is not going 
away any time soon.  I have some 10 year old systems here which are 
still running strong and still just as useful with running their 
respective applications as the day they were built.  If I was to 
upgrade their OS they would no longer be useful.

Bob
======================================
Bob Friesenhahn
address@hidden, http://www.simplesystems.org/users/bfriesen/
GraphicsMagick Maintainer,    http://www.GraphicsMagick.org/