------- Comment #3 from rob1weld at aol dot com  2009-01-23 12:25 -------
> From: Ben Elliston <b...@air.net.au>
> To: rob1w...@aol.com
> Sent: Thu, 22 Jan 2009 8:12 pm
> Subject: Re: Patch for config.guess
>
> Hi Rob,
> 
> Thanks for your report.
> 
> > The script config.guess lumps together many of Sun's Operating
> > Systems into one identifier. To save attempting to parse all the
> > names of the various versions it is easier to rely on the Kernel
> > version number.
> 
> > This is discussed here:
> > http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38783
> 
> I read the bug report.  Solaris is no different to other operating
> systems that we have detected in config.guess for many years.  We have
> traditionally tried to strike a balance between providing enough
> information for GNU packages that care (ie. GCC) and providing too
> much information (such as the specific build number of the kernel).
> 
> If there are specific kernel problems that you are trying to work
> around, it is usually more reliable to have a configure test that
> switches on the config.guess triplet and then actually detects the
> presence of the problem.
> 
> Cheers, Ben

Thanks for considering my request and reading my Bug Report over 
at gcc's Bugzilla.

> ... tried to strike a balance between providing enough information for  
> GNU packages that care (ie. GCC) and providing too much information ...

Agreed. It would not have been great to make the Identifier larger and 
it would certainly not have been pretty, but there is a lot of overlap 
between different OSes.

Since "My Operating System is correctly reported" (or you could file a 
Bug Report) I have nothing to worry about. The people who program gcc 
(and other software) will simply have to figure out from the given 
Identifier which OS they are dealing with and know what to test for.

Thanks,
Rob


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=38783

Reply via email to