2.6.26-0.25.rc3.git4 or VirtualBox ?

[sean darcy]

I'm using VirtualBox to run rawhide.

kernel-2.6.26-0.25.rc3.git4.fc10.i686 locks up hard with:

unable to handle kernel NULL pointer dereference at 00000246

2.6.25.3-18 works fine.

Does anyone else see this, or is it VirtualBox?

sean

[David]

sean darcy wrote:
> I'm using VirtualBox to run rawhide.
>
> kernel-2.6.26-0.25.rc3.git4.fc10.i686 locks up hard with:
>
> unable to handle kernel NULL pointer dereference at 00000246
>
> 2.6.25.3-18 works fine.
>
> Does anyone else see this, or is it VirtualBox?
>
> sean

I see what is probably the same thing.

[Arjan van de Ven]

On Sun, 25 May 2008 16:47:39 -0400
sean darcy  wrote:
> I'm using VirtualBox to run rawhide.
>
> kernel-2.6.26-0.25.rc3.git4.fc10.i686 locks up hard with:
>
> unable to handle kernel NULL pointer dereference at 00000246
>
> 2.6.25.3-18 works fine.

I wouldn't call it  "fine";
VirtualBox is one of the top causes of kernel oopses that we're seeing
on Linux right now... I'm not surprised it's not working well with
2.6.26 either.

[sean darcy]

Arjan van de Ven wrote:
> On Sun, 25 May 2008 16:47:39 -0400
> sean darcy  wrote:
> > I'm using VirtualBox to run rawhide.
> >
> > kernel-2.6.26-0.25.rc3.git4.fc10.i686 locks up hard with:
> >
> > unable to handle kernel NULL pointer dereference at 00000246
> >
> > 2.6.25.3-18 works fine.
>
> I wouldn't call it  "fine";
> VirtualBox is one of the top causes of kernel oopses that we're seeing
> on Linux right now... I'm not surprised it's not working well with
> 2.6.26 either.

"fine" means it WFM. It does all I ask.

I'm certainly not a vbox fanatic. In fact, I've only been using it for a 
couple of days. That's why I posted: is this a vbox problem, or a kernel 
issue.

If no one else is seeing a hard lockup on bare metal, then it's vbox. 
AFAICS from the response, 1 guy sees the same lockup on bare metal, 2 
don't like vbox.

sean

[zjsun]

Hi, sean

I am not an expert on Virtual box or kernel, but I know
kernel-2.6.26-0.25.rc3.git4.fc10.i686 is for F10, I use virtualbox
1.6.0 on Fedora 9, it works fine realy.  AFAIK virtual box only
supports Fedora9 or older Fedora series now. F10 is not in the scope
of consideration now since its Alpha version is not ready now.

Thanks

Zongjun

2008/5/26 sean darcy :
> Arjan van de Ven wrote:
> > On Sun, 25 May 2008 16:47:39 -0400
> > sean darcy  wrote:
> > > I'm using VirtualBox to run rawhide.
> > >
> > > kernel-2.6.26-0.25.rc3.git4.fc10.i686 locks up hard with:
> > >
> > > unable to handle kernel NULL pointer dereference at 00000246
> > >
> > > 2.6.25.3-18 works fine.
> >
> > I wouldn't call it  "fine";
> > VirtualBox is one of the top causes of kernel oopses that we're seeing
> > on Linux right now... I'm not surprised it's not working well with
> > 2.6.26 either.
>
> "fine" means it WFM. It does all I ask.
>
> I'm certainly not a vbox fanatic. In fact, I've only been using it for a
> couple of days. That's why I posted: is this a vbox problem, or a kernel
> issue.
>
> If no one else is seeing a hard lockup on bare metal, then it's vbox. AFAICS
> from the response, 1 guy sees the same lockup on bare metal, 2 don't like
> vbox.
>
> sean
>
> --
> fedora-devel-list mailing list
> [e-mail]
> https://www.redhat.com/mailman/listinfo/fedora-devel-list

[Arjan van de Ven]

On Sun, 25 May 2008 21:26:09 -0400
sean darcy  wrote:
> Arjan van de Ven wrote:
> > On Sun, 25 May 2008 16:47:39 -0400
> > sean darcy  wrote:
> > > I'm using VirtualBox to run rawhide.
> > >
> > > kernel-2.6.26-0.25.rc3.git4.fc10.i686 locks up hard with:
> > >
> > > unable to handle kernel NULL pointer dereference at 00000246
> > >
> > > 2.6.25.3-18 works fine.
> >
> > I wouldn't call it  "fine";
> > VirtualBox is one of the top causes of kernel oopses that we're
> > seeing on Linux right now... I'm not surprised it's not working
> > well with 2.6.26 either.
>
> "fine" means it WFM. It does all I ask.
>
> I'm certainly not a vbox fanatic. In fact, I've only been using it
> for a couple of days. That's why I posted: is this a vbox problem, or
> a kernel issue.

you're probably seeing:

http://www.kerneloops.org/search.php?search=g_abExecMemory

which a lot of other folks are seeing, so much so that this is in the
top 15 of all kernel oopses/warnings reported in the last week.

Given the backtrace and the high frequency of this thing, I'm pretty
sure it's a VirtualBox bug.

[Pavel Shevchuk]

After Sun bought Innotek i can't wait for xen dom0 kernel availability
in fedora9. For me vbox is dead =(

On 5/25/08, sean darcy  wrote:
> I'm using VirtualBox to run rawhide.
>
> kernel-2.6.26-0.25.rc3.git4.fc10.i686 locks up hard with:
>
> unable to handle kernel NULL pointer dereference at 00000246
>
> 2.6.25.3-18 works fine.
>
> Does anyone else see this, or is it VirtualBox?
>
> sean
>
> --
>  fedora-devel-list mailing list
>  [e-mail]
>  https://www.redhat.com/mailman/listinfo/fedora-devel-list