qemu-arm
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-arm] [PATCH 0/5] ast2400: some cleanups and a simple memory co


From: Cédric Le Goater
Subject: Re: [Qemu-arm] [PATCH 0/5] ast2400: some cleanups and a simple memory controller model
Date: Tue, 12 Jul 2016 18:20:00 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:45.0) Gecko/20100101 Icedove/45.1.0

Hello Peter,

On 07/12/2016 04:19 PM, Peter Maydell wrote:
> On 8 July 2016 at 17:06, Cédric Le Goater <address@hidden> wrote:
>> Hello,
>>
>> Here are a couple of cleanups and a very basic memory controller model
>> providing a configuration register. This is needed to determine the
>> RAM size of the SOC, only used by U-Boot as of today.
>>
>> Ultimately, we would want to add more support for U-Boot and be able
>> to boot from flash 0 but the device model of the m25p80 object is not
>> designed to use a memory region. And so, it is difficult to define a
>> rom device for the flash backend.
>>
>> Some initial patches were proposed but they are not being friendly
>> with the object interface. Work in progress.
> 
> Hi; the release process is now at the point where I wouldn't really
> want to put this into 2.7; I can be persuaded if you have a solid
> justification, though.

Here is my take to it  :

* 1/5 hw/misc: fix typo in Aspeed SCU hw-strap2 property name
  
  That's a minor fix

* 2/5 ast2400: replace aspeed_smc_is_implemented()

  That is not a fix but it removes some ugliness in the code, which 
  you had noticed, and which prevents us from fixing possible issues 
  nicely. So this is a preventive cleanup :)

* 3/5 ast2400: pretend DMAs are done for U-boot

  This is a work around fix. U-boot[1] uses DMAs to calibrate the 
  SPI flash module. It will spin until the DMA transfers are done.

* 4/5 ast2400: externalize revision numbers

  This is minor fix de-duplicating code and a prereq for the next
  patch.

* 5/5  ast2400: add a memory controller device model

  That's an enhancement. It adds a basic device model for the memory
  controller. This is needed by some version of the SDK relying on
  the configuration register to know how much RAM the SOC has. Older
  versions used a hardcoded value.  


The first patches should not be too much of a problem regarding the 
soft freeze. The last can wait 2.8 if you prefer, we need more work 
to fully support u-boot anyhow. I would be happy to have review 
though :) The model is very much like the SCU.

Thanks,

C.

[1] not supported yet in qemu because of the lack of a rom device 
    memory region needed to boot from the flash module.





reply via email to

[Prev in Thread] Current Thread [Next in Thread]