Re: [Qemu-devel] [PATCH 0/3] dp8393x update

[Laurent Vivier]

Le 02/01/2015 02:34, Laurent Vivier a écrit :
> Hi Hervé,
> 
> Le 01/01/2015 22:01, Hervé Poussineau a écrit :
>> Hi Laurent,
>>
>> Le 29/12/2014 01:39, Laurent Vivier a écrit :
>>> This is a series of patches I wrote to use dp8393x (SONIC) with
>>> Quadra 800 emulation. I think it is interesting to share them with the
>>> mainline.
>>>
>>> Qdev'ifying allows to remove the annoying warning:
>>> "requested NIC (anonymous, model dp83932) was not created
>>>   (not supported by this machine?)"
>>>
>>> [PATCH 1/3] dp8393x: add registers offset
>>> [PATCH 2/3] dp8393x: add PROM to store MAC address
>>> [PATCH 3/3] qdev'ify dp8393x
>>>
>>
>> I also had some patches to QOM'ify dp8393x.
>> Those are available at
>> http://repo.or.cz/w/qemu/hpoussin.git/shortlog/refs/heads/sonic
>>
>> Main differences are:
>> - dp8393x uses an AddressSpace, instead of an offset in a MemoryRegion
>> in yours
>> - no PROM support, but should be easy to add
>> - rc4030 (MIPS Jazz chipset) also converted to QOM (but that was not the
>> goal of your patch series)
>>
>> Minor points are:
>> - have load/save support
>> - all functions have the same dp8393x_ prefix
>> - old_mmio-style functions are not used anymore
>>
>> What do you think of them?
> 
> I don't know if it's a good idea to use AddressSpace into device. For
> me, AddressSpace must stay in the machine definition. SysBus is there
> for that. But it seems to be a good way to do DMA. I have to think about
> that...

Using AddressSpace is really a very good idea, in fact, but I don't like
the way you pass it to the device (a qdev_set_prop()).

I think we should do as it is done in PCI. This must be managed at
sysbus level, not at the device level.

Find attached a (not tested) patch to show what I'm thinking about.

Regards,
Laurent

sysbus_dma_rw.patch
Description: Text Data

signature.asc
Description: OpenPGP digital signature