Re: [PATCH 3/3] iio: accel: bmc150: Get mount-matrix from ACPI

[Hans de Goede <hdegoede@redhat.com>]

Hi,

On 11/25/20 12:07 PM, Andy Shevchenko wrote:
> On Wed, Nov 25, 2020 at 10:37 AM Hans de Goede <hdegoede@redhat.com> wrote:
>>
>> bmc150 accelerometers with an ACPI hardware-id of BOSC0200 have an ACPI
>> method providing their mount-matrix, add support for retrieving this.
> 
> ...
> 
>> +       if (strcmp(dev_name(dev), "i2c-BOSC0200:base") == 0)
>> +               alt_name = "ROMK";
>> +       else
>> +               alt_name = "ROMS";
>> +
>> +       if (acpi_has_method(adev->handle, "ROTM"))
>> +               name = "ROTM";
> 
> My gosh, it's a third method of this...
> 
> ...
> 
>> +       elements = obj->package.elements;
>> +       for (i = 0; i < 3; i++) {
>> +               if (elements[i].type != ACPI_TYPE_STRING)
>> +                       goto unknown_format;
>> +
>> +               str = elements[i].string.pointer;
>> +               if (sscanf(str, "%d %d %d", &val[0], &val[1], &val[2]) != 3)
>> +                       goto unknown_format;
>> +
>> +               for (j = 0; j < 3; j++) {
>> +                       switch (val[j]) {
>> +                       case -1: str = "-1"; break;
>> +                       case 0:  str = "0";  break;
>> +                       case 1:  str = "1";  break;
>> +                       default: goto unknown_format;
>> +                       }
>> +                       orientation->rotation[i * 3 + j] = str;
>> +               }
>> +       }
> 
> I'm wondering if we can come up with some common code out of this and
> existing apply_acpi_orientation().

Honestly they are all different enough that I don't think it is worth
the trouble (I did take a look at this, but it did not seem feasible
without creating horrible code).

Regards,

Hans