[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: Bug#382129: Beta3 won't boot on OldWorld PowerPC Mac


What we need most is for someone who has familiarity with the configuration options and build procedures for the powerpc kernel and initrd to take a look at the situation and provide us testers with a kernel/initrd to test.

Failing that, I'm willing to try building kernels and initrds with various configuration options and/or patches if someone would be willing to teach me how. I can RTFM, if I know which FM to R.

Willingness aside, I fear that I would be casting about at random, since I know very little of the options that changed during the ppc/ ppc64 -> powerpc transition. Again, we need someone who has some experience in the area to take notice and help out.

Please! Help!


On Aug 22, 2006, at 12:27 PM, Harold Johnson wrote:

Thanks for explaining this, Sven. I was unaware of the nature of changes that occurred prior to the 2.6.15 kernel, and this gives me a general idea. This also seems to answer the original question posed by Rick Thomas -- the question which began this thread.

Rick, Let me know if there's anything I can test/do to help you get what you need. I know you're wanting 2.6.16, and I haven't had success there yet -- but perhaps I can help in some other way. I currently have a partition available to test on my PowerBook G3. (You'll recall, it's a "Wallstreet" PowerBook.)


On 8/21/06, Sven Luther <sven.luther@wanadoo.fr> wrote:
Previous to 2.6.15/2.6.14, there where two kernel architectures, ppc and ppc64, with lot of duplication. In 2.6.15 the migration started for ppc64 to have a single architecture for both, namely powerpc, and this was completed for powerpc/chrp in 2.6.16, while prep and apus still stayed in ARCH=ppc, and
nubus is not ported to 2.6 yet.

If you get a kernel source, the thing is that the code changed between what
was in arch/ppc and now is in arch/powerpc.

This is a pretty major migration, and code changed, so for oldworld machines,
problems with 2.6.16 are pretty much linked to this migration.

Reply to: