Hi all,
This patch has been in my patch directory for a while.
The PLEB is a SA-1100-based ARM computer developed at CSE at the
University of New South Wales. I have discovered some of the earlier
models would not set register 1 properly, which was required for Linux
to boot. This was inside their (very old) kernel tree but which they
never submitted for inclusion (?) It is a Photon1 with catapult
bootloader combination.
- G.
--
char p[] = "\xeb\x1f\x5e\x89\x76\x08\x31\xc0\x88\x46\x07\x89\x46\x0c\xb0\x0b"
"\x89\xf3\x8d\x4e\x08\x8d\x56\x0c\xcd\x80\x31\xdb\x89\xd8\x40\xcd"
"\x80\xe8\xdc\xff\xff\xff/bin/sh";
On Wed, Mar 26, 2003 at 12:13:58AM +0800, Geoffrey Lee wrote:
> The PLEB is a SA-1100-based ARM computer developed at CSE at the
> University of New South Wales. I have discovered some of the earlier
> models would not set register 1 properly, which was required for Linux
> to boot. This was inside their (very old) kernel tree but which they
> never submitted for inclusion (?) It is a Photon1 with catapult
> bootloader combination.
I've been killing these - people should really be passing the right
value of r1 to the kernel. Think what happens when 200 different
machine types add these 3 lines.
A saner solution would be to define this appropriately if we're only
being built for one platform.
BTW, please send ARM stuff to the linux-arm-kernel list. See
http://lists.arm.linux.org.uk/mailman/listinfo/
--
Russell King ([email protected]) The developer of ARM Linux
http://www.arm.linux.org.uk/personal/aboutme.html
[cc-ing to linux-arm-kernel ]
On Tue, Mar 25, 2003 at 04:38:43PM +0000, Russell King wrote:
> On Wed, Mar 26, 2003 at 12:13:58AM +0800, Geoffrey Lee wrote:
> > The PLEB is a SA-1100-based ARM computer developed at CSE at the
> > University of New South Wales. I have discovered some of the earlier
> > models would not set register 1 properly, which was required for Linux
> > to boot. This was inside their (very old) kernel tree but which they
> > never submitted for inclusion (?) It is a Photon1 with catapult
> > bootloader combination.
>
> I've been killing these - people should really be passing the right
> value of r1 to the kernel. Think what happens when 200 different
> machine types add these 3 lines.
>
I totally agree that it is wrong not to set r1.
Thank goodness that in 2.4.x not many arm boards need this hack :)
Anyway, inside the (very old) pleb kernel tree, it was marked as a "hack".
> A saner solution would be to define this appropriately if we're only
> being built for one platform.
>
Yes. That sounds good.
-- G.
--
char p[] = "\xeb\x1f\x5e\x89\x76\x08\x31\xc0\x88\x46\x07\x89\x46\x0c\xb0\x0b"
"\x89\xf3\x8d\x4e\x08\x8d\x56\x0c\xcd\x80\x31\xdb\x89\xd8\x40\xcd"
"\x80\xe8\xdc\xff\xff\xff/bin/sh";