[haiku-development] Re: PPC sync / isync cleanup PATCH and adding missing kernel calls.
- From: Alexander von Gluck <kallisti5@xxxxxxxxxxx>
- To: <haiku-development@xxxxxxxxxxxxx>
- Date: Thu, 08 Oct 2009 08:01:29 -0500
On Thu, 08 Oct 2009 10:13:20 +0200 CEST, "Axel Dörfler"
<axeld@xxxxxxxxxxxxxxxx> wrote:
> Alexander von Gluck <kallisti5@xxxxxxxxxxx> wrote:
>> Really? Wow, I was not aware Haiku ever booted properly on PPC.
>
> Just to be clear, with "booted fine" I meant: it panicked in the kernel
> stating there were no boot devices (natural, since there is no PCI
> support yet).
>
Ha, ok that's a lot less exciting and a little more expected.
It would be nice to get back to the point of getting the kernel
started. My biggest fear is that the PPC port will miss some critical
code updates and x86 will leave it in the dust. PPC will get so outdated
and disconnected that no-one will want to put the time in to fix the
basics
(this kind of already happened. PPC wouldn't even compile before the first
wave of my patches)
>> That
>> rules out problems in the basics of implementation of the
>> openfirmware
>> bootloader.
>
> Maybe. One thing that was previously not there was the relocation of
> the kernel, and other changes might have broken fundamental things as
> well (there are also a bunch of TODOs that might need to be resolved).
I would love to work on the TODOs and think it's a great next step,
unfortunately
my C++ is just weak enough that I am really not a whole lot of help. I
know a
little PPC assembly (yeah I learned x68 assembly before C... what of it?
:P) but
not enough to be mapping page stacks and stuff.
If anyone out there closely knows PPC and has some free time, please take
a look at
the TODOs... maybe this would be a good candidate for the next gSOC? PPC
programming
is still useful for embedded devices, there still has to be a class out
there somewhere
for it.
--Alex
http://haikufire.com
Other related posts:
