Re: [PATCH v4 2/8] ACPI: property: Parse _CRS CSI-2 descriptor

[Sakari Ailus <sakari.ailus@linux.intel.com>]

Hi Andy,

Thanks for the review.

On Wed, Feb 08, 2023 at 06:43:36PM +0200, Andy Shevchenko wrote:
> On Wed, Feb 08, 2023 at 05:28:01PM +0200, Sakari Ailus wrote:
> > Parse newly added ACPI _CRS CSI-2 descriptor for CSI-2 and camera
> > configuration. For now, only figure out where the descriptor is present in
> > order to allow adding information from it to related devices.
> 
> Nit-picks below that may be ignored. Up to you, guys.
> 
> ...
> 
> > +#define NO_CSI2_PORT (~1U)
> 
> A bit unclear why this value. Is it bitfield? Then GENMASK() would be better.
> Is it a plain value with a type limit? Then (UINT_MAX - 1) probably the best.

This value is used to signify that a port node isn't yet allocated for a
CSI-2 port. I can change this to UINT_MAX - 1.

> 
> ...
> 
> > +	ads->nodeptrs = (void *)(ads->nodes +
> > +				 ports_count * 2 + 1);
> 
> Now this fits one line.

Yes.

> 
> ...
> 
> > +	handle_refs = kcalloc(csi2_all.handle_count + 1, sizeof(*handle_refs),
> > +			      GFP_KERNEL);
> > +	if (!handle_refs) {
> > +		acpi_handle_debug(handle, "no memory for %zu handle refs\n",
> > +				  csi2_all.handle_count + 1);
> > +		return;
> > +	}
> 
> In a code above you used "1 + foo" approach if I'm not mistaken. Why here is
> the difference?

The last entry is a guardian (NULL handle). 1 + number of lanes is used
where the first lane is the clock lane.

> 
> ...
> 
> In some code comments which I removed before remarking you forgot the
> grammatical period.

Some comments aren't proper sentences. I'll see which could be improved by
adding a period.

-- 
Regards,

Sakari Ailus