On Sat, Apr 17, 2004 at 12:28:15AM +0100, Colin Watson wrote: > On Fri, Apr 16, 2004 at 07:58:48PM +0200, Pip Oomen wrote: > > Until I found this: > > > > http://lists.penguinppc.org/yaboot-users/2003/yaboot-users-200307/msg00017.html > > > > Then it hit me that we do not have to wait for yaboot 2 to be released, > > but that it might very well be possible to patch yaboot to allow for > > bigger kernels: > > > > --- yaboot-1.3.11/second/fs_of.c 2002-09-15 05:12:33.000000000 +0200 > > +++ yaboot-1.3.11.new/second/fs_of.c 2004-04-15 11:41:11.000000000 +0200 > > @@ -44,7 +44,7 @@ > > #include "debug.h" > > > > #define LOAD_BUFFER_POS 0x600000 > > -#define LOAD_BUFFER_SIZE 0x400000 > > +#define LOAD_BUFFER_SIZE 0x600000 > > > > static int of_open(struct boot_file_t* file, const char* dev_name, > > struct partition_t* part, const char* file_name); > > > > Lo, and behold, this makes the system boot the debian-installer kernel :) > > Yeah, that's what I thought, what I suggested on debian-powerpc, and > what was suggested in bug #242348. The upstream maintainer has > inarticulate objections to it and wants it tested everywhere in the > universe before he'll buy it. > > I think it may be easier just to reduce the extremely large size of the > kernels we're using. Apparently the 2.6 packages in preparation have > smaller kernels, or there's powerpc-small.
Well, the current 2.6.5 powerpc kernel has miboot friendly size, so it should be ok. We also tried booting the pmac-elf kernel with builtin initrd on a prismo from the netbooting OF directly, but there was some error because the compressed kernel + initrd was bigger than 4Mo. That said, this 2.6.5 kernel use a initrd for module booting. Did you make progress on 2.6 d-i on ppc ? Friendly, Sven Luther -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]