On 14.03.2012 18:29, Mark Brown wrote:
> On Tue, Mar 13, 2012 at 05:54:38PM +0100, Karol Lewandowski wrote:
>
>> - replace s3c24xx_i2c_type enum with plain unsigned int that can
>> hold not only device type but also hw revision-specific quirks
>
> Would it not be clearer to just have explicit flags for the quirks (eg,
> as a set of bitfield flags)?
That would work on runtime but we also need to initialize this
somehow. The way it was done today on non-dt platforms is via platform
device variants, i.e. (taken from 3rd patch):
@@ -1128,6 +1161,9 @@ static struct platform_device_id s3c24xx_driver_ids[] = {
}, {
.name = "s3c2440-i2c",
.driver_data = TYPE_S3C2440,
+ }, {
+ .name = "s3c2440-hdmiphy-i2c",
+ .driver_data = TYPE_S3C2440 | FLAG_HDMIPHY | FLAG_NO_GPIO,
Ability to address above scenario was sole motivation for this change.
Without it one would need either need separate type (e.g.
TYPE_S3C2440_HDMIPHY) or setting flags based just on device name.
Introducing separate type (TYPE_S3C2440_HDMIPHY) has been our original
attempt to solve this issue. However, this required adding explicit
checks to driver code all over the place (if (type == S3C2400 ||
type == S3c2440_HDMIPHY).
Thus, I felt that sqeezing quirks into type is a bit cleaner approach.
Regards,
--
Karol Lewandowski | Samsung Poland R&D Center | Linux/Platform
--
To unsubscribe from this list: send the line "unsubscribe linux-i2c" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
