Re: [PATCH v4 5/7] xen/riscv: introduce and initialize SBI RFENCE extension

[Jan Beulich <jbeulich@xxxxxxxx>]

On 16.08.2024 14:06, oleksii.kurochko@xxxxxxxxx wrote:
> On Tue, 2024-08-13 at 11:34 +0200, Jan Beulich wrote:
>>>   
>>> +static unsigned long sbi_spec_version = SBI_SPEC_VERSION_DEFAULT;
>>> +static unsigned long sbi_fw_id, sbi_fw_version;
>>
>> __ro_after_init for perhaps all three of them?
>>
>> Considering SBI_SPEC_VERSION_{MAJOR,MINOR}_MASK, at least the first
>> of them also doesn't need to be unsigned long, but could be unsigned
>> int?
> 
> sbi_spec_version can be really unsigned int as according to the spec
> only 32 bits are used:
> ```
>    struct sbiret sbi_get_spec_version(void);
>    Returns the current SBI specification version. This function must
>    always succeed. The minor number
>    of the SBI specification is encoded in the low 24 bits, with the
>    major number encoded in the next 7
>    bits. Bit 31 must be 0 and is reserved for future expansion.
> ```
> 
> For sbi_fw_id, sbi_fw_version it is not mentioned the same thing, so we
> can just assume ( despite of the fact that now this values are very
> small and it is unlikely to be bigger the UINT_MAX ) that it will be
> always fit in uint32_t.
> 
> But I think it would be better to leave unsigned long for everyone as
> according to the specification this functions returns sbiret structure
> which consist of 2 longs ( error and value ) and it is good idea to
> follow the specification.

Okay with me (for these two) then.

Jan