This series changes how the fdt for VOF is generated in pegasos2 by
moving the static parts to a dtb and only generate the changing parts
such as memory size and PCI devices programmatically. This simplifies
the code and allows simply adding emulation of Pegasos I which has a
different north bridge and slightly different memory map but otherwise
very similar and can be emulated by reusing parts from the amigaone
machine. The machine was tested with a Pegasos I ROM image and MorphOS.

The first VOF patch (submitted separetely before, the reviewed v3 is
included here) fixes handling the name property in VOF that cannot be
represented in a dts as that always takes the path as the name and
cannot accept an explicit name property but we need the name property
to appear when guest queries properties which previously was worked
around by adding it to every node.

Regards,
BALATON Zoltan

Link to previous version:
https://patchew.org/QEMU/cover.1760798392.git.balaton@eik.bme.hu/

v4:
- rebase on master
- added some R-b tags from and changes suggested by Philippe
- reworded commit message of patch 8 trying to clarify it

v3:
- rebase on master

v2:
- rebase on master
- added some R-b tags from Philippe
- move first patch later (was first to allow merging separately)
- clarify blurb above

BALATON Zoltan (12):
  ppc/vof: Make nextprop behave more like Open Firmware
  hw/ppc/pegasos2: Remove explicit name properties from device tree
  hw/ppc/pegasos2: Change device tree generation
  hw/ppc/pegasos2: Remove fdt pointer from machine state
  hw/ppc/pegasos2: Rename mv field in machine state
  hw/ppc/pegasos2: Add south bridge pointer in the machine state
  hw/ppc/pegasos2: Move PCI IRQ routing setup to a function
  hw/ppc/pegasos2: Move hardware specific parts out of machine reset
  hw/ppc/pegasos2: Introduce abstract superclass
  hw/ppc/pegasos2: Add bus frequency to machine state
  hw/ppc/pegasos2: Add Pegasos I emulation
  hw/ppc/pegasos2: Add VOF support for pegasos1

 MAINTAINERS              |   1 +
 hw/ppc/pegasos2.c        | 769 +++++++++++++++++++--------------------
 hw/ppc/vof.c             |  50 ++-
 pc-bios/dtb/meson.build  |   2 +
 pc-bios/dtb/pegasos1.dtb | Bin 0 -> 857 bytes
 pc-bios/dtb/pegasos1.dts | 125 +++++++
 pc-bios/dtb/pegasos2.dtb | Bin 0 -> 1701 bytes
 pc-bios/dtb/pegasos2.dts | 167 +++++++++
 8 files changed, 703 insertions(+), 411 deletions(-)
 create mode 100644 pc-bios/dtb/pegasos1.dtb
 create mode 100644 pc-bios/dtb/pegasos1.dts
 create mode 100644 pc-bios/dtb/pegasos2.dtb
 create mode 100644 pc-bios/dtb/pegasos2.dts

-- 
2.41.3

On 10/23/25 05:36, BALATON Zoltan wrote:
> This series changes how the fdt for VOF is generated in pegasos2 by
> moving the static parts to a dtb and only generate the changing parts
> such as memory size and PCI devices programmatically. This simplifies
> the code and allows simply adding emulation of Pegasos I which has a
> different north bridge and slightly different memory map but otherwise
> very similar and can be emulated by reusing parts from the amigaone
> machine. The machine was tested with a Pegasos I ROM image and MorphOS.
> 
> The first VOF patch (submitted separetely before, the reviewed v3 is
> included here) fixes handling the name property in VOF that cannot be
> represented in a dts as that always takes the path as the name and
> cannot accept an explicit name property but we need the name property
> to appear when guest queries properties which previously was worked
> around by adding it to every node.
> 
> Regards,
> BALATON Zoltan

Queued.

On Thu, 23 Oct 2025, Harsh Prateek Bora wrote:
> On 10/23/25 05:36, BALATON Zoltan wrote:
>> This series changes how the fdt for VOF is generated in pegasos2 by
>> moving the static parts to a dtb and only generate the changing parts
>> such as memory size and PCI devices programmatically. This simplifies
>> the code and allows simply adding emulation of Pegasos I which has a
>> different north bridge and slightly different memory map but otherwise
>> very similar and can be emulated by reusing parts from the amigaone
>> machine. The machine was tested with a Pegasos I ROM image and MorphOS.
>> 
>> The first VOF patch (submitted separetely before, the reviewed v3 is
>> included here) fixes handling the name property in VOF that cannot be
>> represented in a dts as that always takes the path as the name and
>> cannot accept an explicit name property but we need the name property
>> to appear when guest queries properties which previously was worked
>> around by adding it to every node.
>> 
>> Regards,
>> BALATON Zoltan
>
> Queued.

OK. In that case I can do follow up if needed but I still plan to send 
updated raven series and if time allows updated firmware for sam460ex but 
the latter is unsure if I can finish before the freeze.

Regards,
BALATON Zoltan

On Thu, 23 Oct 2025, BALATON Zoltan wrote:
> On Thu, 23 Oct 2025, Harsh Prateek Bora wrote:
>> On 10/23/25 05:36, BALATON Zoltan wrote:
>>> This series changes how the fdt for VOF is generated in pegasos2 by
>>> moving the static parts to a dtb and only generate the changing parts
>>> such as memory size and PCI devices programmatically. This simplifies
>>> the code and allows simply adding emulation of Pegasos I which has a
>>> different north bridge and slightly different memory map but otherwise
>>> very similar and can be emulated by reusing parts from the amigaone
>>> machine. The machine was tested with a Pegasos I ROM image and MorphOS.
>>> 
>>> The first VOF patch (submitted separetely before, the reviewed v3 is
>>> included here) fixes handling the name property in VOF that cannot be
>>> represented in a dts as that always takes the path as the name and
>>> cannot accept an explicit name property but we need the name property
>>> to appear when guest queries properties which previously was worked
>>> around by adding it to every node.
>>> 
>>> Regards,
>>> BALATON Zoltan
>> 
>> Queued.
>
> OK. In that case I can do follow up if needed but I still plan to send 
> updated raven series and if time allows updated firmware for sam460ex but the 
> latter is unsure if I can finish before the freeze.

Or if you haven't sent the pull request yet (I was confused by getting 
cc-d on a pull that did not go to the list, maybe was only for testing) 
then I can still update this series too. I see there was some problem with 
another patch from Thomas so there will be another version of that too. So 
the plan is to update raven series, after that update this one to address 
Philippe's comments then if there's still time I see if I can do the 
sam460ex U-Boot update but you don't have to wait for that last one.

Regards,
BALATON Zoltan

On 10/23/25 17:45, BALATON Zoltan wrote:
> On Thu, 23 Oct 2025, BALATON Zoltan wrote:
>> On Thu, 23 Oct 2025, Harsh Prateek Bora wrote:
>>> On 10/23/25 05:36, BALATON Zoltan wrote:
>>>> This series changes how the fdt for VOF is generated in pegasos2 by
>>>> moving the static parts to a dtb and only generate the changing parts
>>>> such as memory size and PCI devices programmatically. This simplifies
>>>> the code and allows simply adding emulation of Pegasos I which has a
>>>> different north bridge and slightly different memory map but otherwise
>>>> very similar and can be emulated by reusing parts from the amigaone
>>>> machine. The machine was tested with a Pegasos I ROM image and MorphOS.
>>>>
>>>> The first VOF patch (submitted separetely before, the reviewed v3 is
>>>> included here) fixes handling the name property in VOF that cannot be
>>>> represented in a dts as that always takes the path as the name and
>>>> cannot accept an explicit name property but we need the name property
>>>> to appear when guest queries properties which previously was worked
>>>> around by adding it to every node.
>>>>
>>>> Regards,
>>>> BALATON Zoltan
>>>
>>> Queued.
>>
>> OK. In that case I can do follow up if needed but I still plan to send 
>> updated raven series and if time allows updated firmware for sam460ex 
>> but the latter is unsure if I can finish before the freeze.
> 
> Or if you haven't sent the pull request yet (I was confused by getting 
> cc-d on a pull that did not go to the list, maybe was only for testing) 
> then I can still update this series too. I see there was some problem 
> with another patch from Thomas so there will be another version of that 
> too. So the plan is to update raven series, after that update this one 
> to address Philippe's comments then if there's still time I see if I can 
> do the sam460ex U-Boot update but you don't have to wait for that last one.

Sure. no issues.
I have included these patches in the pull request I sent today and am
planning another pull request before soft/hard freeze, so we can include
remaining patches that gets reviewed in that. Thanks.

regards,
Harsh

> 
> Regards,
> BALATON Zoltan

On Thu, 23 Oct 2025, Harsh Prateek Bora wrote:
> On 10/23/25 17:45, BALATON Zoltan wrote:
>> On Thu, 23 Oct 2025, BALATON Zoltan wrote:
>>> On Thu, 23 Oct 2025, Harsh Prateek Bora wrote:
>>>> On 10/23/25 05:36, BALATON Zoltan wrote:
>>>>> This series changes how the fdt for VOF is generated in pegasos2 by
>>>>> moving the static parts to a dtb and only generate the changing parts
>>>>> such as memory size and PCI devices programmatically. This simplifies
>>>>> the code and allows simply adding emulation of Pegasos I which has a
>>>>> different north bridge and slightly different memory map but otherwise
>>>>> very similar and can be emulated by reusing parts from the amigaone
>>>>> machine. The machine was tested with a Pegasos I ROM image and MorphOS.
>>>>> 
>>>>> The first VOF patch (submitted separetely before, the reviewed v3 is
>>>>> included here) fixes handling the name property in VOF that cannot be
>>>>> represented in a dts as that always takes the path as the name and
>>>>> cannot accept an explicit name property but we need the name property
>>>>> to appear when guest queries properties which previously was worked
>>>>> around by adding it to every node.
>>>>> 
>>>>> Regards,
>>>>> BALATON Zoltan
>>>> 
>>>> Queued.
>>> 
>>> OK. In that case I can do follow up if needed but I still plan to send 
>>> updated raven series and if time allows updated firmware for sam460ex but 
>>> the latter is unsure if I can finish before the freeze.
>> 
>> Or if you haven't sent the pull request yet (I was confused by getting cc-d 
>> on a pull that did not go to the list, maybe was only for testing) then I 
>> can still update this series too. I see there was some problem with another 
>> patch from Thomas so there will be another version of that too. So the plan 
>> is to update raven series, after that update this one to address Philippe's 
>> comments then if there's still time I see if I can do the sam460ex U-Boot 
>> update but you don't have to wait for that last one.
>
> Sure. no issues.
> I have included these patches in the pull request I sent today and am
> planning another pull request before soft/hard freeze, so we can include
> remaining patches that gets reviewed in that. Thanks.

OK, so the pull did go to the list. I'm OK with this series as it is, the 
remaining changes were minor and could be done as follow up (or later). 
I'll then concentrate on the raven series and see if I'll have time for 
sam460ex. Thank you for stepping in and taking care of it.

Regards,
BALATON Zoltan