Re: [PATCH 2/5] hw/arm/aspeed: Select console UART from machine

[Philippe Mathieu-Daudé]

On 8/31/21 3:34 PM, Cédric Le Goater wrote:
> On 8/31/21 1:23 PM, Andrew Jeffery wrote:
>> Hi Cédric, Peter,
>>
>> On Tue, 31 Aug 2021, at 20:09, Cédric Le Goater wrote:
>>> On 8/28/21 5:58 PM, Peter Delevoryas wrote:
>>>> I think I’m a little confused on this part. What I meant by “most machines 
>>>> just use UART5” was that most DTS’s use “stdout-path=&uart5”, but fuji 
>>>> uses “stdout-path=&uart1”. I /do/ see that SCU510 includes a bit related 
>>>> to UART, but it’s for disabling booting from UART1 and UART5. I just care 
>>>> about the console aspect, not booting.
>>>
>>> The UART can be switched with SCU70[29] on the AST2500, btw.
>>
>> If it helps, neither the AST2600's "Boot from UART" feature nor the 
>> AST2[456]00's "Debug UART" feature are related to which UART is used as 
>> the BMC console by u-boot and/or the kernel - the latter is entirely a 
>> software thing.
> 
> ok. 
> 
> I don't think we should initialize all 5 UARTs of SoC and let the user define 
> all the expected devices on the command. Unless we want to do something like
> 'macs_mask' ? but at the SoC level. It might be overkill for the need.

I'm not sure I'm following what you are suggesting. If we are talking
about QEMU device initialization, QEMU must initialize all devices
on the board, regardless the guest code uses them or not.