Re: [PATCH v2 04/11] hw/arm/aspeed: Use the IEC binary prefix definitions

["Philippe Mathieu-Daudé" <philmd@linaro.org>]

On 18/1/23 08:32, Cédric Le Goater wrote:
> On 1/18/23 07:53, Joel Stanley wrote:
>> On Fri, 30 Dec 2022 at 11:35, Philippe Mathieu-Daudé 
>> <philmd@linaro.org> wrote:
>>>
>>> IEC binary prefixes ease code review: the unit is explicit.
>>
>> I strongly prefer the existing code; it tells you the size without
>> having to do maths.

I guess it depends on one's mindset and culture / education,
maybe I'm too young for the full hexadecimal world and am more
custom to decimal notation with binary prefixes.

0x16400 is just another magic number for me, while 89 * KiB is a size.

> you mean that it matches better with the address space representation
> in the code and the 'info mtree' output ? If so, I agree. We can keep
> this patch out, it is not fundamental.
> 
> The hex representation of values has its advantages compared to the
> macros because hex is generally what you get in debug outputs and
> it is easier to compare and manipulate.  Some Linux dev feel the
> same.

>>> aspeed_soc_ast2600_class_init(ObjectClass *oc, void *data)
>>>       sc->name         = "ast2600-a3";
>>>       sc->cpu_type     = ARM_CPU_TYPE_NAME("cortex-a7");
>>>       sc->silicon_rev  = AST2600_A3_SILICON_REV;
>>> -    sc->sram_size    = 0x16400;
>>> +    sc->sram_size    = 89 * KiB;
To keep everybody happy I'll respin using:

          sc->sram_size    = 0x16400; /* 89 * KiB */

Thanks,

Phil.