Re: Bizarre difference in uptimes for same kernel

Michael L. Galbraith (
Fri, 12 Sep 1997 20:40:01 +0200 (MET DST)

On Thu, 11 Sep 1997, Peter Benie wrote:

> linux kernel account writes ("Re: Bizarre difference in uptimes for same kernel"):
> > On Tue, 9 Sep 1997, Rob Partington wrote:
> >
> > > Is it possible that compiling the kernel with pgcc instead of gcc
> > > actually makes for a more stable kernel?
> >
> > I've compiled everything on my system with pgcc and I have did not see the
> > lockups you described
> My experience with pgcc is not so good. I compiled the kernel (2.0.30,
> SMP Pentium Pro) with gcc and pgcc. The gcc the kernel works fine (apart
> from lockups with ptrace). The pgcc kernel is fine 99% of the time, but
> suffers from random weirdness... I found I had lots of lpd processes
> running (they're supposed to read their bit of the printcap and die
> immediately) and ypbind stopped trying to write into the binding directory
> (perhaps it couldn't get its domain from uname?). Needless to say, I
> switched back to the gcc kernel fairly quickly.
> Peter Benie

Hi Peter,

The latest/greatest is pretty nice. I'm using egcs-970910 with the pgcc
patches hand integrated. It's working well enough to compile XF3.3.1
and run with _no_ detectable errors. No earlier pgcc version could compile
X and not have runtime glitches happen. I ran 2.1.54 compiled O9 also with
no problems.. no version of gcc or pgcc could do that with my setup. Take
another look.. it's coming along very nicely indeed.

I'm still having problems with the c++ libs, but then I've _always_ had
problems there :) The compiler itself is working very nicely IMHO.