Re: [PATCH 2/7] platform/x86: think-lmi: Use ACPI object when extracting strings

["Mark Pearson" <mpearson-lenovo@squebb.ca>]

Hi Armin

On Sun, Feb 9, 2025, at 7:31 PM, Armin Wolf wrote:
> Am 03.02.25 um 19:23 schrieb Armin Wolf:
>
>> Move the ACPI buffer handling out of tlmi_extract_output_string()
>> and instead pass the unpacked ACPI object to prepare for future
>> changes.
>
> Hi,
>
> i was hoping that maybe the driver maintainer could take a look at this patch
> and give some feedback.
>
My apologies - because of this patch (and a couple of others) I've just realised I never updated the MAINTAINERS file.
I have been mothballing the markpearson@lenovo.com address as it's a nightmare to use and switched to using my personal email domain instead. It seems my email filters aren't flagging these messages the way they are supposed to be - I have to figure that out :(

> Thanks,
> Armin Wolf
>
>> Signed-off-by: Armin Wolf <W_Armin@gmx.de>
>> ---
>>   drivers/platform/x86/think-lmi.c | 38 +++++++++++++++++---------------
>>   1 file changed, 20 insertions(+), 18 deletions(-)
>>
>> diff --git a/drivers/platform/x86/think-lmi.c b/drivers/platform/x86/think-lmi.c
>> index 323316ac6783..2c94a4af9a1d 100644
>> --- a/drivers/platform/x86/think-lmi.c
>> +++ b/drivers/platform/x86/think-lmi.c
>> @@ -262,16 +262,11 @@ static int tlmi_simple_call(const char *guid, const char *arg)
>>   	return 0;
>>   }
>>
>> -/* Extract output string from WMI return buffer */
>> -static int tlmi_extract_output_string(const struct acpi_buffer *output,
>> -				      char **string)
>> +/* Extract output string from WMI return value */
>> +static int tlmi_extract_output_string(union acpi_object *obj, char **string)
>>   {
>> -	const union acpi_object *obj;
>>   	char *s;
>>
>> -	obj = output->pointer;
>> -	if (!obj)
>> -		return -ENOMEM;
>>   	if (obj->type != ACPI_TYPE_STRING || !obj->string.pointer)
>>   		return -EIO;
>>
>> @@ -352,17 +347,21 @@ static int tlmi_opcode_setting(char *setting, const char *value)
>>   static int tlmi_setting(int item, char **value, const char *guid_string)
>>   {
>>   	struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
>> +	union acpi_object *obj;
>>   	acpi_status status;
>>   	int ret;
>>
>>   	status = wmi_query_block(guid_string, item, &output);
>> -	if (ACPI_FAILURE(status)) {
>> -		kfree(output.pointer);
>> +	if (ACPI_FAILURE(status))
>>   		return -EIO;
>> -	}
>>
>> -	ret = tlmi_extract_output_string(&output, value);
>> -	kfree(output.pointer);
>> +	obj = output.pointer;
>> +	if (!obj)
>> +		return -ENODATA;
>> +
>> +	ret = tlmi_extract_output_string(obj, value);
>> +	kfree(obj);
>> +
>>   	return ret;
>>   }
>>
>> @@ -370,19 +369,22 @@ static int tlmi_get_bios_selections(const char *item, char **value)
>>   {
>>   	const struct acpi_buffer input = { strlen(item), (char *)item };
>>   	struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL };
>> +	union acpi_object *obj;
>>   	acpi_status status;
>>   	int ret;
>>
>>   	status = wmi_evaluate_method(LENOVO_GET_BIOS_SELECTIONS_GUID,
>>   				     0, 0, &input, &output);
>> -
>> -	if (ACPI_FAILURE(status)) {
>> -		kfree(output.pointer);
>> +	if (ACPI_FAILURE(status))
>>   		return -EIO;
>> -	}
>>
>> -	ret = tlmi_extract_output_string(&output, value);
>> -	kfree(output.pointer);
>> +	obj = output.pointer;
>> +	if (!obj)
>> +		return -ENODATA;
>> +
>> +	ret = tlmi_extract_output_string(obj, value);
>> +	kfree(obj);
>> +
>>   	return ret;
>>   }
>>
>> --
>> 2.39.5
>>
>>
Changes look good to me. If you can hold on a bit I'll see if I can test them on a few platforms to make sure no surprises.

Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca>

Mark