[meta-xilinx] [PATCH v1] qemuzynmp: Add the qemuzynqmp machine

[Alistair Francis]

On Mon, May 16, 2016 at 9:39 AM, Alistair Francis
<alistair.francis at xilinx.com> wrote:
> On Sun, May 15, 2016 at 10:10 AM, Peter Crosthwaite
> <crosthwaitepeter at gmail.com> wrote:
>> On Sun, May 15, 2016 at 1:17 AM, Nathan Rossi <nathan at nathanrossi.com> wrote:
>>> On Sat, May 14, 2016 at 2:54 AM, Alistair Francis
>>> <alistair.francis at xilinx.com> wrote:
>>>> On Fri, May 13, 2016 at 2:35 AM, Nathan Rossi <nathan at nathanrossi.com> wrote:
>>>>> On Fri, May 13, 2016 at 6:11 AM, Alistair Francis
>>>>> <alistair.francis at xilinx.com> wrote:
>>>>>> Add the qemuzynqmp machine based on the physical ZCU02 revB board.
>>>>>
>>>>> Hi Alistair,
>>>>>
>>>>> So I am a little confused, is this machine supposed to emulate the
>>>>> ep108 (the args to QEMU appear to be for the ep108) or the zcu102? or
>>>>> are both able to be emulated?
>>>>
>>>> Hey Nathan,
>>>>
>>>> This machine is supposed to be emulating the ZCU102.
>>>>
>>>> At the moment we only have the EP108 in upstream QEMU. It isn't worth
>>>> adding a ZCU102 model upstream yet as at the moment we don't have
>>>> enough modeled upstream to notice any differences between the boards.
>>>
>>> Ah ok, so in the future QEMU will have models for both the EP108 and
>>> the ZCU102 boards. In that case I don't actually see a need for a
>>> 'qemuzynqmp' machine, since the ep108-zynqmp already provides the
>>> setup for QEMU. The same can be done for the zcu102-zynqmp board, this
>>> would also make it easier to maintain consistency (aka changes only
>>> need to be made to the zcu102-zynqmp machine).
>>>
>>
>> +1, qemuzynq as well as the "zynq" machine model in upstream QEMU are
>> legacy convention that we should replace with concrete boards going
>> forward.
>
> Ok, I'll look into adding a ZCU102 machine similar to what we do for the EP108.

How can I run the EP108 on QEMU? I can't find how I can start QEMU with Yocto.

Thanks,

Alistair

>
> Thanks,
>
> Alistair