[Intel-gfx] [PATCH v2 4/6] drm: scrambling support in drm layer

[Jani Nikula]

On Wed, 08 Feb 2017, Jose Abreu <Jose.Abreu at synopsys.com> wrote:
> Hi Jani,
>
>
>
> On 07-02-2017 15:09, Jani Nikula wrote:
>> On Tue, 07 Feb 2017, Jose Abreu <Jose.Abreu at synopsys.com> wrote:
>>> Hi Jani,
>>>
>>>
>>> On 07-02-2017 13:35, Jani Nikula wrote:
>>>> On Tue, 07 Feb 2017, Jose Abreu <Jose.Abreu at synopsys.com> wrote:
>>>>>> +bool drm_scdc_check_scrambling_status(struct i2c_adapter *adapter)
>>>>>> +{
>>>>>> +	u8 status;
>>>>>> +	int ret;
>>>>>> +
>>>>>> +	ret = drm_scdc_readb(adapter, SCDC_SCRAMBLER_STATUS, &status);
>>>>>> +	if (ret < 0) {
>>>>>> +		DRM_ERROR("Failed to read scrambling status, error %d\n", ret);
>>>>>> +		return false;
>>>>>> +	}
>>>>>> +
>>>>>> +	return status & SCDC_SCRAMBLING_STATUS;
>>>>> "return (status & SCDC_SCRAMBLING_STATUS) > 0;" ?
>>>> What's the point in that?
>>>>
>>>> BR,
>>>> Jani.
>>>>
>>>>
>>> Sorry, I didn't see the SCDC_SCRAMBLING_STATUS is BIT(0). Anyway,
>>> my intention was to return either 1 or 0 or else the value of the
>>> "and" would be returned. I think in x86 the bool is char, what
>>> could happen if SCDC_SCRAMBLING_STATUS was BIT(>7)? Does the cast
>>> work as expected?
>> The implicit type conversion works just fine.
>
> Hmm, are you sure? I'm reading this thread:
> http://yarchive.net/comp/linux/bool.html (see Linus last answer).

I think you're confusing ABI with what C guarantees. I don't think that
thread has any relevance here.

BR,
Jani.


>
> (This is just for curiosity anyway).
>
> Best regards,
> Jose Miguel Abreu
>
>>
>> BR,
>> Jani.
>>
>>> Best regards,
>>> Jose Miguel Abreu
>>>
>

-- 
Jani Nikula, Intel Open Source Technology Center