Re: bug report

["Tim Prince" <tprince at computer dot org>]

According to my understanding, gcc-2.96 is not recommended as a project
base, but that does leave us at a loss to know which version is recommended.
I just installed the gcc-2.95.2.1, as I don't really care about the
modernizations in 2.97, and got a -march=pentiupro failure from it.  The
ICE's which I reported early in 2.96 continue in 2.97. If -march=pentiumpro
is not supported, it should be so documented, or did I miss something along
that line?

----- Original Message -----
From: "Alan Rojer" <rojer@on2.com>
To: <gcc-bugs@gcc.gnu.org>
Sent: Monday, January 01, 2001 11:32 AM
Subject: bug report


> Hi gcc'ers, here's an internal compiler error in gcc 2.96.
>
> I've attached the .ii file, the version information, and the compiler
> command line, etc.  Please let me know if you require any additional
> information.  I found the gnats-web interface awkward since I think
> the attachments are the most useful thing here, and it doesn't seem to
> support them.
>
> This is proprietary code for us, so I would be grateful if you'd treat
> it as such.
>
> This is from a RedHat 7.0 system, freshly installed.  This code has been
> building successfully on many prior g++ compilers (lastest was gcc 2.95.3,
> on a RH 6.2 system, but also on gcc 2.7.2.3 on a RH 5.2 system).  I'm very
> distressed about the busted build, as this is a crucial development effort
> for us.  I am about to try upgrade my gcc rpm to 2.96-69 (this is 2.96-54,
> but I doubt that'll help; I expect I'll be rolling back to 2.95-3 shortly.
>
> Thanks for any help, and thanks also for all the great work you gcc
> workers perform!
>
> - Alan
>
> --
> Alan S. Rojer, Ph.D.
> Director of R&D
> rojer@on2.com
> ON2.COM, Inc.
>