[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] Re: [PATCH 1/2] Factor out common SharpSL PDA code
From: |
Dmitry Baryshkov |
Subject: |
[Qemu-devel] Re: [PATCH 1/2] Factor out common SharpSL PDA code |
Date: |
Sat, 31 May 2008 16:40:00 +0000 (UTC) |
User-agent: |
Pan/0.132 (Waxed in Black) |
andrzej zaborowski wrote:
> On 30/05/2008, Dmitry Baryshkov <address@hidden> wrote:
>> andrzej zaborowski wrote:
>> > I think of the Tosa as part of spitz family, but we can split the
>> > file if you think it's advantageous. hw/spitz.c file would be named
>> > hw/zaurus.c but I had to be paranoid about trademarked names at that
>> > time, and afaik the SCOOP chip is only used on the Zaurus.
>>
>> SCOOP is a custom ASiC used only on zaurus. However not all zaurii are
>> spitz'. I moved all common functionality to separate file (should be
>
> For that matter akita isn't spitz either...
It's like a variation of spitz. The difference between spitz and tosa is
rather big.
>
>> named hw/zaurus.c :)) to make things more clear. And separated
>
> Ok, I can rename when committing and optimally I'd use device.h instead
> of sharpsl.h
Fine with me. Please commit it so.
>
>
>>
>>
>> > BTW, are you planning to add the other Tosa peripherals?
>>
>> Yes. I'm looking at it now. The main problem probably will be TC6393XB
>> which is a SiC containing NAND controller, FB/LCD, OHCI and (unused in
>> tosa) MMC controller. The current code ("return 3") is just a
>> placeholder to make machine type detector work. wm9712 (audio codec +
>> ts interface) and kbd seem to be simpler to implement.
>
> Great stuff. The WM9712 can be also used in emulating some of the newer
> Palms.
Yup. However doing it properly will probably require refactoring of hw/
ac97.c. I'll probably start from tc6393xb.
>> BTW: I wasn't able to get PXA MMC emulation working neither with my
>> code, nor with spitz emulation. Does it work for you?
>
> It worked at one time but now I see the card intialisation stops after
> CMD1, so no, it doesn't.
>
Yes, I saw that. Linux gets timeouts and nothing more.
--
With best wishes
Dmitry