Hi,

After having a look to the contents of initrd.img-3.0.0-2-mckinley and
of initrd.img-2.6.32-5-mckinley, I have notice some points about the
first

one that don't boot.
It confirms that busybox is missing even if the included
conf/initramfs.conf

says BUSYBOX=y.
Also it appears to have a multiarch content and also libm.so is
missing.

May it be a problem?

Regards,
Patrice.

ps: the diff result is attached.



--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 73d68774970b09434568c644acfe5253@mail.igh.cnrs.fr" >http://lists.debian.org/73d68774970b09434568c644acfe5253@mail.igh.cnrs.fr

I've committed a fix for #638068 to git.

It doesn't address the fact that klibc/sh appears to not be
"good enough" to boot ia64 (which is an entirely different
and, I assume, klibc bug), but it should improve the handling
enough to make it at least do as requested for both BUSYBOX=y
and BUSYBOX=n (which should fix ia64, as long as you use =y)

Can someone test out git head and let me know if this works
for you. It looks "obviously correct" to me, but I can't
readily test it to make sure it DTRT.

... Adam



--
To UNSUBSCRIBE, email to debian-kernel-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: 20120701063705.GT23052@0c3.net">http://lists.debian.org/20120701063705.GT23052@0c3.net