Re: gcc bug affects NTLMSSP authentication helper

If you get it's normal output redirected to a file, and leave stderr to
console ?

Or perhaps do one of the display and wait 15 second messages like some
packages do?

Rob

----- Original Message -----
From: "Adrian Chadd" <adrian@creative.net.au>
To: "Chemolli Francesco (USI)" <ChemolliF@GruppoCredit.it>
Cc: <squid-dev@squid-cache.org>
Sent: Friday, February 23, 2001 10:21 PM
Subject: Re: gcc bug affects NTLMSSP authentication helper

> On Fri, Feb 23, 2001, Chemolli Francesco (USI) wrote:
> > > Ah. Right. Well, how about you try the latest gcc to see if
> > > they've fixed it, and if they haven't, forward them the
> > > information (code snippet triggering the bug, resulting
> > > assembly code, gcc flags, etc) so they can try to fix it?
> >
> > Sorry, I forgot to add this info.
> > The GCC shipped in RH7.0 (2.96) works fine.
>
> Right. Well, the best thing here is to bail out of configure
> and tell users about the problem with a nice long explanation.
> If they want to compile without -O, add something on the command
> line to do it.
>
> I didn't like how squid-freebsd just 'disables' -O completely
> with a single line explanation. On some machines that single
> line goes by *WAY* too quick.
>
> Which brings up an interesting point to me. Is there some
> way to get configure to not display those "checking for.."
> lines? That way, we could actually have configure display
> the configuration options rather than being blown away by
> the checks (which we enable by some command line option..)
>
>
> Adrian
>
> --
> Adrian Chadd "Romance novel?"
> <adrian@creative.net.au> "Girl Porn."
> - http://www.sinfest.net/d/20010202.html
>
Received on Fri Feb 23 2001 - 04:29:05 MST