[Bug 233863] r345425 on PowerMac G5 may require kern.smp.disabled=1 and must set usefdt=1 which causes net interface reorder

[bugzilla-noreply at freebsd.org]

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=233863

Mark Millard <marklmi26-fbsd at yahoo.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #203889|0                           |1
        is obsolete|                            |

--- Comment #30 from Mark Millard <marklmi26-fbsd at yahoo.com> ---
Created attachment 203981
  --> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=203981&action=edit
Investigatory stand/powerpc/ofw/ofwfdt.c and sys/contrib/libfdt/fdt_rw.c
patches

The MPC750 iMac G4 PowerMac4,1 needs to have its
/memory0/available information used and respected
or the "extra" address range usage will end up
trashed. openfirmware may well report to exclude
some address ranges having nothing to do with
openfirmware's internal memory use.

With this change the MPC750 iMac G3 PowerMac4,1
example boots and runs in *BOTH* non-usefdt mode
and usefdt mode.

It also means that every PowerMac that I have access
to that non-usefdt mode historically worked on now
also works for *BOTH* non-usefdt mode and usefdt mode:
G3, G4, and G5.

Textually the patches would need adjustments to be
acceptable for FreeBSD check-in. Also it is possible
that some things might have other directions things
could instead be taken. But they give solid evidence
about what needs to be addressed.

Note:

There are list messages with notes about other things
I ran into while trying to find what would make things
work again for both non-usefdt mode and usefdt mode.
>From what I observe, none are important for this bugzilla
(beyond what the various attached patches deal with).

-- 
You are receiving this mail because:
You are the assignee for the bug.