coreboot + Linux = ?

[Peter Stuge]

I came across this commit included in 2.6.24:
http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=...

Does anyone know what the current attitude towards making Linux play
nice with coreboot is?

//Peter

Quoting Peter Stuge :
> I came across this commit included in 2.6.24:
> http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=...
>
> Does anyone know what the current attitude towards making Linux play
> nice with coreboot is?

That looks like it is just for the "EFI frame buffer" correct? Why  
would it have any effect on coreboot?

Thanks - Joe

[Peter Stuge]

On Sat, Mar 15, 2008 at 09:59:15PM -0400, [e-mail] wrote:
> > I came across this commit included in 2.6.24:
> > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=...
> >
> > Does anyone know what the current attitude towards making Linux play
> > nice with coreboot is?
>
> That looks like it is just for the "EFI frame buffer" correct?

Yes.
> Why would it have any effect on coreboot?

It wouldn't. The code along with the list of people involved in
that commit tells me that many are interested in EFI + Linux.

coreboot is an alternative. I was asking if anyone knew about
interest for a stronger bond between coreboot and Linux.

//Peter

Quoting Peter Stuge :
> On Sat, Mar 15, 2008 at 09:59:15PM -0400, [e-mail] wrote:
> > > I came across this commit included in 2.6.24:
> >
> > http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commit;h=...
> > > Does anyone know what the current attitude towards making Linux play
> > > nice with coreboot is?
> >
> > That looks like it is just for the "EFI frame buffer" correct?
>
> Yes.
> > Why would it have any effect on coreboot?
>
> It wouldn't. The code along with the list of people involved in
> that commit tells me that many are interested in EFI + Linux.
>
> coreboot is an alternative. I was asking if anyone knew about
> interest for a stronger bond between coreboot and Linux.

Ah, gotcha...

Thanks - Joe

[ron minnich]

On Sat, Mar 15, 2008 at 6:11 PM,   wrote:
> > coreboot is an alternative. I was asking if anyone knew about
> > interest for a stronger bond between coreboot and Linux.
>
> Ah, gotcha...

Actually, it requires only that somebody start dropping patches in.
But what, really, do we need? EFI makes all kinds of demands on Linux
... coreboot doesn't.

Thanks

ron

> EFI makes all kinds of demands on Linux ... coreboot doesn't.

That was my point exactly, I was just being a little more suttle about it....

Thanks - Joe

[Peter Stuge]

On Sat, Mar 15, 2008 at 06:22:41PM -0800, ron minnich wrote:
> On Sat, Mar 15, 2008 at 6:11 PM,   wrote:
> > > coreboot is an alternative. I was asking if anyone knew about
> > > interest for a stronger bond between coreboot and Linux.
> >
> > Ah, gotcha...
>
> Actually, it requires only that somebody start dropping patches in.
> But what, really, do we need?

As you know my grand plan is to get rid of all things legacy. I do
think there is, or will be, a need.

On the way to that 2.6.24 I took a detour into Ulrich Drepper's
excellent What Every Programmer Should Know About Memory.

On the topic of HT hypercube information in sysfs:

--8<-- http://people.redhat.com/drepper/cpumemory.pdf page 45
Each processor constitutes its own node as can be seen
by the bits set in the value in cpumap file in the node*
directories. The distance files in those directories con-
tains a set of values, one for each node, which represent
a cost of memory accesses at the respective nodes. In
this example all local memory accesses have the cost 10,
all remote access to any other node has the cost 20. [26]
..
[26] This is, by the way, incorrect. The ACPI information is appar-
ently wrong since, although the processors used have three coherent
HyperTransport links, at least one processor must be connected to a
Southbridge. At least one pair of nodes must therefore have a larger
distance.
-->8--

For people to do a really good job with libNUMA it seems benefitial
to replace the wrong information with something that is right.

Any advantage coreboot can offer is good I think.

//Peter

[Peter Stuge]

On Sun, Mar 16, 2008 at 04:51:30PM +0100, Peter Stuge wrote:
> Any advantage coreboot can offer is good I think.

Make that: Every advantage coreboot can offer is worthwhile.

//Peter

[Stefan Reinauer]

ron minnich wrote:
> On Sat, Mar 15, 2008 at 6:11 PM,   wrote:
> > > coreboot is an alternative. I was asking if anyone knew about
> > > interest for a stronger bond between coreboot and Linux.
> >
> > Ah, gotcha...
>
> Actually, it requires only that somebody start dropping patches in.
> But what, really, do we need? EFI makes all kinds of demands on Linux
> ... coreboot doesn't.

This is our big marketing weakness leading to answers such as "fix your
bios". Believe it or not, I just implemented an SMBIOS implementation
for coreboot in order to be able to use ACPI without acpi=force on the
command line. That's the way to go? We do have requirements to the OS,
but we do not pronounce them enough. One requirement would be: Don't
look for SMBIOS if you already found a coreboot lbtable and do use ACPI
in that case, because if thats buggy, we can fix it.
Or: we initialize video with 1024x786x16 instead of 80x25 and have a
corebootfb read that resolution from lbtable, behaving like vesafb
otherwise.
There are things that come to mind when actually thinking beyound
booting a cluster or server.

Stefan

[Rudolf Marek]

Hi,
> Believe it or not, I just implemented an SMBIOS implementation
> for coreboot in order to be able to use ACPI without acpi=force on the
> command line.

My DSDT in Asus A8V-E SE does not need any acpi=force option. All you need is to 
switch the PMIO block from legacy to ACPI mode in Coreboot. Please can you tell 
me more why you needed acpi=force?

Rudolf

[Torsten Duwe]

On Sunday 16 March 2008, Peter Stuge wrote:
> > Why would it have any effect on coreboot?
>
> It wouldn't. The code along with the list of people involved in
> that commit tells me that many are interested in EFI + Linux.

Just because someone is on the Cc list does not necessarily mean they're 
interested.
> coreboot is an alternative. I was asking if anyone knew about
> interest for a stronger bond between coreboot and Linux.

It is not. Coreboot is an alternative to intel's "platform initialisation", 
which they propose _along_ with EFI. Coreboot + EFI + Linux? Why not, once 
it's all open source. OTOH, Coreboot + filo + Linux will always outperform it 
8-)

[Peter Stuge]

On Tue, Mar 18, 2008 at 11:30:31PM +0100, Torsten Duwe wrote:
> Just because someone is on the Cc list does not necessarily mean
> they're interested.

Fair enough. But at least they are receiving and supposedly reading,
and probably also feeling. :)
> > coreboot is an alternative. I was asking if anyone knew about
> > interest for a stronger bond between coreboot and Linux.
>
> It is not.

Yes and no. I prefer to talk about the way that it is.
> OTOH, Coreboot + filo + Linux will always outperform it

And $Subject even more so. ;)

//Peter