Patches to allow usefdt mode that works on a 2 socket PowerMac3, 6 example too --and makes more work on 2-socket/1-core-each PowerMac11, 2

[Mark Millard]

On 2019-Apr-12, at 13:13, Dennis Clarke <dclarke at blastwave.org> wrote:

> On 4/12/19 3:19 PM, Mark Millard via freebsd-ppc wrote:
>> With the following 2 patches for converting
>> openfirmware to fdt content. . .
>> PowerMac11,2 example for usefdt mode:
>> A) bge0 and bge1 are back in their historical places.
>> B) powerd on the PowerMac11,2 works again.
>> C) sysctl -a | grep cpufreq lists items for all the cpus
>> D) probably more.
>> PowerMac3,6 example for usefdt mode:
>> E) gem0 is present again and even works.
>> F) Both CPUs are used again.
>> F) probably more.
>> (powerpd and cpufreq's are not operable/present even in
>> non-usefdt mode.)
>> This message does not deal with other investigatory
>> patches for other issues than converting openfirmware
>> to fdt. But my test environment has all my investigatory
>> patches in order to avoid other things getting in the
>> way of my investigations.
>> The code comments are fairly explicit about
>> what and why for the changes.
>> I treat the patches as investigatory, not ready
>> in form for being official FreeBSD material. There
>> are likely questions of if the change go in the
>> right long-term direction --or even if old PowerMacs
>> will continue to be viewed as worth supporting
>> (because they compete with time spent on modern
>> support).
>> I'll note that I've never had the 2-socket/1-core-each
>> PowerMac7,2 get any visible behavior after the Kernel
>> entry point message with any variation of usefdt mode.
>> Failing so early, I've not figured out any way to
>> investigate it hanging up. It does boot in non-usefdt
>> mode with my other investigatory patches in place.
>> The openfirmware to fdt conversion patches are (white
>> space details might not have been preserved in the
>> message):
> .
> .
> .<awesome deep magic snippage>
> .
> .
>> ===
>> Mark Millard
> 
> Would you be so kind as to paste all this into :
> 
> https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=233863
> 
> Really I would like to run some tests and follow up in the bug reports.

Okay I'll paste them in as attachments. But be warned:

The 2 files do not deal with threads being stuck sleeping
(and, so, the fans going) or other such. The stuck-sleeping
problem happens for both multi-socket G5's and multi-socket
G4's. (I do not have access to single-socket multi-core
powerpc64 or powerpc machines to test.)

So do not expect too much from these patches: They address
some necessary issues but are not sufficient for everything.

These patches for the openfirmware->fdt translation are
closer to being reasonable for FreeBSD official use
than my highly context-specific stuck-sleeping patches for
usefdt mode.


===
Mark Millard
marklmi at yahoo.com
( dsl-only.net went
away in early 2018-Mar)