[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: SMP i586?

On Tuesday 25 November 2003 10:47 am, Michael K. Johnson wrote:
> On Mon, Nov 24, 2003 at 09:07:42AM +0100, Klaasjan Brand wrote:
> > Search the list archives; I believe SMP was dropped from RH9 or Fedora
> > because it never was really popular for i586 (and didn't perform that
> > well too). You can build an i586 SMP kernel yourself...

> That's correct.  We don't have any functional i586 SMP hardware that
> I'm aware of either.

> If it's just for one machine, it's probably best for Lamar to just
> rebuild the kernel for his machine (sorry, Lamar!), but if there are
> really a significant number of people who will not only use after
> release, but *test before release* on i586 SMP, then it's at least
> worth discussion of whether or not to re-enable it.

I was just curious as to the why.  The fact that Red Hat has no SMP i586 
hardware is enough for me.  As to performance, it is better than the UP i586 
kernel on this box.  At 133MHz, every little bit helps.

As far as building kernels, that's the easy part.  Edit the configs, edit the 
spec file, issue a rpmbuild --target=i586 -ba kernel-2.4.spec and go away for 
awhile during the build.  Then install the fresh kernel RPM's (which has a 
special release tag, of course), which automatically updates things.  Well, 
at least it's supposed to, but this box doesn't do grub, it does lilo (GRUB 
locks the box during boot, but LILO boots right up -- but that's probably due 
to /boot being on a RAID).  I have found that this is the most reliable, 
least trouble way for me at least to build a custom kernel.  But, then again, 
I have maintained various RPM's for over four years, so this cycle is a 
familiar one.

>  (Part of the
> downside being forcing everyone to download those bits when they
> get the ISOs, and then we get to also argue about what other bits
> are being traded off the ISOs...)

You ain't a kiddin'....
Lamar Owen
Director of Information Technology
Pisgah Astronomical Research Institute
1 PARI Drive
Rosman, NC  28772

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]