Re: [PATCH v5 10/11] platform/x86/amd/hsmp: Change devm_kzalloc() to devm_kcalloc()

["Hegde, Suma" <Suma.Hegde@amd.com>]

On 1/31/2024 4:02 PM, Ilpo Järvinen wrote:
> Caution: This message originated from an External Source. Use proper caution when opening attachments, clicking links, or responding.
>
>
> On Mon, 29 Jan 2024, Hegde, Suma wrote:
>
>> On 1/29/2024 6:14 PM, Ilpo Järvinen wrote:
>>> Caution: This message originated from an External Source. Use proper caution
>>> when opening attachments, clicking links, or responding.
>>>
>>>
>>> On Thu, 25 Jan 2024, Ilpo Järvinen wrote:
>>>
>>>> On Sat, 6 Jan 2024, Suma Hegde wrote:
>>>>
>>>>> Use the standard array allocation variant of devm memory allocation
>>>>> APIs.
>>>>>
>>>>> Signed-off-by: Suma Hegde <suma.hegde@amd.com>
>>>>> Signed-off-by: Naveen Krishna Chatradhi <nchatrad@amd.com>
>>>> I decided to apply all but this patch 10/11 into review-ilpo. I fixed the
>>>> parenthesis issue I mentioned in one of the patches while applying.
>>>>
>>>> Please check the comment I made against this patch and respin this one.
>>>>
>>>> I also noticed while applying one other extra parenthesis case in patch 5
>>>> but since it was not added, I didn't go to tweak it now myself, but just
>>>> you know.
>>> Hi Suma,
>>>
>>> There are number of issues and warnings due to these patches including
>>> one build failure due to lack of ACPI in the config (I think), can you
>>> please take a look at them.
>> Hi Ilpo,
>>
>> I have pushed patch with fixes for smatch error and warnings.
>>
>> For the CONFIG_ACPI=n build failure, I have added "depends on ACPI" for hsmp
>> driver and pushed patch for that.
> Hi,
>
> I've folded your fixes into the relevant patches now.
>
>> But we support NON-ACPI probing also, there may be x86 platforms with ACPI
>> disabled, is there a previous reference of how this can be handled
>>
>> without making it dependent on ACPI in Kconfig?
> Given you have quite much code that relates to ACPI case, perhaps
> creating hsmp-acpi.c wouldn't be a bad idea so you can make that file
> depend on ACPI without polluting the hsmp.c code with #ifdefs.


Thanks Ilpo for the suggestion. I will address Han's and Greg's comments 
and later will work on splitting the ACPI code into separate file.


Regards,

Suma

> --
>   i.