RE: New kernel for testing

I've run into the same problem as [email protected] trying the kernel 2.4.19-13 with SMP support on a SS1 (2xRT625 bootprom 2.25R).

free_bootmem: base [0] size [c000000]
reserve_bootmem: base [0] size [25e000]
reserve_bootmem: base [25e000] size [1800]
Level 15 Interrupt

In the past, I was successful in compiling a 2.4.19 kernel but this was without SMP support. I've been trying with SMP support ever since but no joy !

Any hopes that a 2.4.x kernel with SMP support will exist for a sparc32?

Eivan

-----Original Message-----
From: Thorsten Kukuk [mailto:[email protected]]
Sent: 29 September 2002 09:32
To: [email protected]
Subject: Re: [suse-sparc] New kernel for testing


On Sat, Sep 28, [email protected] wrote:

> Hi all,
> this sounds like (Hardware-)initialisation Problems on Hypersparcs
> in SMP-Mode or Bugs in arch depended SMP-Kernel-Code on Kernel 2.4.x.
> 
> Thorsten, do you use your 125MHz HyperSparc's as SMP-Engines?

No, currently only as UP system.

> If not, i will try to check this Kernel on my ss10 with only 1 CPU.
> 
> Thank you
> Rolf
> 
> 
> > -----Original Message-----
> > From: Thorsten Kukuk [mailto:[email protected]]
> > Sent: Saturday, September 28, 2002 9:37 AM
> > To: [email protected]
> > Subject: Re: [suse-sparc] New kernel for testing
> > 
> > 
> > On Fri, Sep 27, [email protected] wrote:
> > 
> > > Hi,
> > > at my ss10 (sun4m, 2 x RT626/142MHz/1MCache, 128MB) the new precompiled
> > > Kernel do _not_ boot.
> > > It stops after the first 5 or 6 Lines (init Mem) with Interrupt 15 and
> > > return to siloprompt.
> > > started twice, he hang and a cold boot is nesessary.
> > > 
> > > IMHO it looks like the old Problem on booting by using a newer 
> > gcc instead
> > > of egcs
> > 
> > If your machine needs a egcs compiled kernel, you have to stuck with
> > kernel 2.2.18 (I don't know why, but this is the last 2.2 kernel which
> > works stable for me on SPARC, later one don't do so).
> > egcs is not able to compile a working 2.4.19+ kernel for sparc32.
> > Currently, the only compiler which works for me is gcc 2.95.3.
> > 
> > But I have running the kernel now on 7 32bit SPARCs (all I currently
> > have) without any problems. All machines are building RPMs the whole
> > day, so they have really something to do.
> > 
> > > but a bug / incompatibility in HyperSparc-CPU's or anything else are
> > > possible too.
> > 
> > I only have 125MHz HyperSparc CPUs, but this stable since one week with
> > this kernel.
> > 
> > > I think, Sparcs using spezial Startupcode to "align" 
> > Codesegemts in propper
> > > Memory,
> > > is this Code checked and proof?
> > 
> > I don't think that anybody is working in this area.
> > 
> >   Thorsten
> > 
> > -- 
> > Thorsten Kukuk       http://www.suse.de/~kukuk/        [email protected]
> > SuSE Linux AG        Deutschherrnstr. 15-19        D-90429 Nuernberg
> > --------------------------------------------------------------------    
> > Key fingerprint = A368 676B 5E1B 3E46 CFCE  2D97 F8FD 4E23 56C6 FB4B
> > 
> > -- 
> > To unsubscribe, e-mail: [email protected]
> > For additional commands, e-mail: [email protected]
> > 
> 
> -- 
> To unsubscribe, e-mail: [email protected]
> For additional commands, e-mail: [email protected]

-- 
Thorsten Kukuk       http://www.suse.de/~kukuk/        [email protected]
SuSE Linux AG        Deutschherrnstr. 15-19        D-90429 Nuernberg
--------------------------------------------------------------------    
Key fingerprint = A368 676B 5E1B 3E46 CFCE  2D97 F8FD 4E23 56C6 FB4B

-- 
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]