Date: Thu, 26 Aug 2010 19:08:49 +0200
From: Jon Clausen <alpine@ymmv.dk>
To: alpine-devel@lists.alpinelinux.org
Subject: Re: [alpine-devel] Soekris 4801, alpine 2.0, kernel trouble
Message-ID: <20100826170848.GA5518@nx8220.ymmv.dk>
References: <20100826120329.GA7723@nx8220.ymmv.dk>
 <AANLkTim_gytd4jh_B5DHPNdd=_6wy8Ddxeg=eQD_AcLr@mail.gmail.com>
Precedence: list
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <AANLkTim_gytd4jh_B5DHPNdd=_6wy8Ddxeg=eQD_AcLr@mail.gmail.com>
User-Agent: Mutt/1.5.20 (2009-06-14)

On Thu, 26 Aug, 2010 at 17:08:08 +0200, Natanael Copa wrote:
> On Thu, Aug 26, 2010 at 2:03 PM, Jon Clausen <alpine@ymmv.dk> wrote:

> > It's my first hike into "The Alps", so there's a lot to get used to, but
> > overall I do like it :)
> 
> nice!

:)

> > The long story short:
> >
> > Booting the 2.6.32.18 kernel with the default parameters on the soekris
> > results in a kernel panic.
> 
> not good.

Agreed

> > After trying out the kernels of alpine 2 beta-* and rc-* I found that:
> >
> > 2.6.32.15 boots ok
> > 2.6.32.16 upwards 'panic'
> 
> This is very useful information.

Good :)

> form the git log, the only thing that changed really was support for EFI.
> 
> http://git.alpinelinux.org/cgit/aports/log/main/linux-grsec?h=2.0-stable
> 
> > After searching the web I found a hint that 'noreplace-paravirt' had worked
> > for someone else. So I tried adding that to the append line in syslinux.cfg
> > - and lo and behold: Now the kernel boots :)
> 
> Thats interesting. I wonder how that could be?

Frankly, I have no idea - except I did notice a kernel message:

"Booting paravirtualized kernel on bare hardware"

But my first thought was that noreplace-paravirt would make matters worse.

> Could you try the vserver iso image too? It has a different kernel,
> without the grsecurity patches. The syslinux.cfg is broken so you'd
> need enter 'vserver' at the boot prompt where it complains about
> missing grsec.

Will do. Probably not tonight, but yeah it makes sense to test that.

> > So far so good, except now I get:
> >
> > pci 0000:00:07.0: BAR 6: address space collision on of device
> > [0x3f000000-0x3f00ffff]
> > pci 0000:00:08.0: BAR 6: address space collision on of device
> > [0x3f000000-0x3f00ffff]
> >
> > - which I'm somewhat unhappy about, since those are two of the three onboard
> > NICs :(
> > (Incidentally the 2.6.32.15 kernel also has these collisions)
> 
> I dont think those causes any problems.

Well... I guess the only way to be sure is to set it up and see what
happens... it's just that *some* of the posts I've seen were talking about:

'seems to run ok for a bit, then bombs out <mumble> <mumble> memory leak?'

or somesuch

> > I'm not too sure how to proceed from here?
> 
> I just pushed 2.6.32.19 kernel. i doubt it makes any difference

I'll have to check that as well...

> > I have a serial console on the box, so it's pretty esay for me to 'grab' the
> > output. In fact I've saved the output from booting the various kernels (with
> > various kernel parameters) but I've nowhere public to put them ATM.
> >
> > They're about 7K-9K each. If anyone's interested, let me know?
> 
> The ones that panic would be interesting to see. Not sure how much i
> can help there though.

OK, I'll tar 'em up and send them in a PM.

- speaking of which;

What's the 'listiquette'?

Attachment size limits, etc?

I see in the archives that people generally respond to both the poster and
the list. Is that the norm? (This message is going to the list only)

/jon
-- 
YMMV


---
Unsubscribe:  alpine-devel+unsubscribe@lists.alpinelinux.org
Help:         alpine-devel+help@lists.alpinelinux.org
---