Re: [Cooker] Re: kernel-secure - back to UP/non-enterprise?

2002-04-28 Thread Charles A. Shirley


On Sun, 28 Apr 2002, Leon Brooks wrote:

> On Sunday 28 April 2002 11:10, Levi Ramsey wrote:
> > On Sun Apr 28  9:51 +0800, Leon Brooks wrote:
> >> Running Mandrake 5.2 in front of me is a dual PPro-200 system on a TD6NF
> >> mobo. One example does not a statistic make, but...
> 
> > Am I just being stupid, or aren't PPro's i686's?
> 
> It could be me being stupid, but IIRC they didn't make the multi-CPU stuff 
> entirely happy until very late in the series.
> 
> > IIRC, the chip architecture in the PPro's is the P6, which is the basis
> > of everything up to the P4.
> 
> I'm guessing that `basis' != complete feature set.
> 
> Cheers; Leon

As I recollect, P-II was electrically the same as the P-pro, but had MMX.
In fact, www.Powerleap.com will sell you an accellerator for your UP P-Pro
machine using a coppermine celeron processor past 700 MHz, and a PPGA
celeron up to 533 MHz for Dual processor systems.  So, as long as i686
optimization doesn't include MMX, etc, I think making the SMP kernels i686
optimized is a good move.  I think i486 had better SMP support than
classic Pentium, and none of the Pentium clones had any.

Best Regards,
Chuck Shirley





Re: [Cooker] Re: kernel-secure - back to UP/non-enterprise?

2002-04-27 Thread Leon Brooks

On Sunday 28 April 2002 11:10, Levi Ramsey wrote:
> On Sun Apr 28  9:51 +0800, Leon Brooks wrote:
>> Running Mandrake 5.2 in front of me is a dual PPro-200 system on a TD6NF
>> mobo. One example does not a statistic make, but...

> Am I just being stupid, or aren't PPro's i686's?

It could be me being stupid, but IIRC they didn't make the multi-CPU stuff 
entirely happy until very late in the series.

> IIRC, the chip architecture in the PPro's is the P6, which is the basis
> of everything up to the P4.

I'm guessing that `basis' != complete feature set.

Cheers; Leon





Re: [Cooker] Re: kernel-secure - back to UP/non-enterprise?

2002-04-27 Thread Levi Ramsey

On Sun Apr 28  9:51 +0800, Leon Brooks wrote:
> Running Mandrake 5.2 in front of me is a dual PPro-200 system on a TD6NF mobo. 
> One example does not a statistic make, but...

Am I just being stupid, or aren't PPro's i686's?

IIRC, the chip architecture in the PPro's is the P6, which is the basis
of everything up to the P4.

-- 
Levi Ramsey
[EMAIL PROTECTED]
[EMAIL PROTECTED]

When it comes down to desperation,
You make the best of your situation.
Linux 2.4.18-11mdk
 11:01pm  up 6 days, 57 min,  7 users,  load average: 0.01, 0.13, 0.15




Re: [Cooker] Re: kernel-secure - back to UP/non-enterprise?

2002-04-27 Thread Leon Brooks

On Saturday 27 April 2002 18:51, Juan Quintela wrote:
> Making SMP versions to work only in i686 & upper is a good move
> because Pentium support for multiprocessing is quite bad, and anyways,
> there is almost no i586 SMP boards (comparing with i686/athlon boards).

Running Mandrake 5.2 in front of me is a dual PPro-200 system on a TD6NF mobo. 
One example does not a statistic make, but...

Cheers; Leon





Re: [Cooker] Re: kernel-secure - back to UP/non-enterprise?

2002-04-27 Thread Peter Ruskin

On Saturday 27 Apr 2002 16:20, Alexander Skwar wrote:
> »Juan Quintela« sagte am 2002-04-27 um 12:51:01 +0200 :
> > - i586
> > - i686
> > - i686SMP
> > - i686-4GB
>
> Hm, are there any disadvantages in the -4GB part compared to the plain
> kernels?  If not, then why not build every kernel with high mem support
> and drop the non -4GB ones?
>
> Alexander Skwar

Some things don't work with high memory support -- for example Win4Lin
-- 
Peter Ruskin, Wrexham, Wales.  AMD Athlon XP 1600+, 512MB RAM.
Registered Linux User 219434.  Mandrake Linux release 8.3 (Cooker) 
Kernel 2.4.18-12mdk,  XFree86 4.2.0, patch level 9mdk.
KDE: 3.0.1 (CVS >= 20020327).  Qt: 3.0.3.  Up 22 hours 45 minutes.





Re: [Cooker] Re: kernel-secure - back to UP/non-enterprise?

2002-04-27 Thread Alexander Skwar

»Juan Quintela« sagte am 2002-04-27 um 12:51:01 +0200 :
> - i586
> - i686
> - i686SMP
> - i686-4GB

Hm, are there any disadvantages in the -4GB part compared to the plain
kernels?  If not, then why not build every kernel with high mem support
and drop the non -4GB ones?

Alexander Skwar
-- 
How to quote:   http://learn.to/quote (german) http://quote.6x.to (english)
Homepage:   http://www.iso-top.de  |Jabber: [EMAIL PROTECTED]
   iso-top.de - Die günstige Art an Linux Distributionen zu kommen
   Uptime: 3 days 10 hours 19 minutes