httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jim Jagielski <...@jaguNET.com>
Subject Re: Configure and non-gcc platforms
Date Mon, 04 Aug 1997 21:36:29 GMT
The general fix should be that Configure then not assume any value
for OPTIM then.

I don't think Configure should do any prompting at all... Q&A
"sessions" are a major pain, IMO.

I propose:
  o Configure no longer set OPTIM if undefined
  o TestLib be able to simply block stdout and not stderr, thus errors
    are seen (providing a hint of what the problem is). Obviously,
    doing this for the "lib" function doesn't make sense, since
    we want that to be clean, but for the general ANSI "sanity"
    check, it does.

+1s anyone?

Marc Slemko wrote:
> 
> On Sun, 3 Aug 1997, Dean Gaudet wrote:
> 
> > Well since we have the OPTIM variable, I don't see any reason to prompt.
> > It could just be set to "OPTIM=-O" and if the user has trouble it should
> > be obvious to them what to whack.
> 
> But the problem is that _how_ do they whack it?
> 
> They can't do so by just setting OPTIM= in Configuration to an empty
> string, since it is currently impossible to set no OPTIM flags; if there
> are none, Configure will use default ones.
> 
> > 
> > Dean
> > 
> > On Sun, 3 Aug 1997, Randy Terbush wrote:
> > 
> > > Regarding this discussion on 'Configure', perhaps it is time that 
> > > we model from the Perl Configure and offer a defaults (-d) flag, or 
> > > prompt the user to confirm our decision.
> > > 
> > > ??
> > > 
> > > 
> > > 
> > > 
> > 
> 
> 


-- 
====================================================================
      Jim Jagielski            |       jaguNET Access Services
     jim@jaguNET.com           |       http://www.jaguNET.com/
            "Look at me! I'm wearing a cardboard belt!"

Mime
View raw message