Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1501977 > unrolled thread

Re: [PATCH v3 1/8] drivers:input:tsc2007: add new common binding names, pre-calibration, flipping and rotation

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH v3 1/8] drivers:input:tsc2007: add new common binding names, pre-calibration, flipping and rotation "H. Nikolaus Schaller" <hns@goldelico.com> - 2016-10-17 16:00 +0200

#1501977 — Re: [PATCH v3 1/8] drivers:input:tsc2007: add new common binding names, pre-calibration, flipping and rotation

[Multipart message — attachments visible in raw view] — view raw

Hi Sebastian,
I did wait for the 4.9-rc1 merge window to be closed until I bring up this patch
set again.

> Am 30.09.2016 um 16:40 schrieb H. Nikolaus Schaller <hns@goldelico.com>:
>> 
>> I do agree with Rob, that the ti,min/max-x/y should become common,
>> though. Also I would do s/minimum value/minimum raw value/g.
> 
> Yes, that is a good idea!
> 
> To me it seems as if "minimum-raw-value-x", "maximum-raw-value-y" could be good.
> 
> Or is the string too long (then we can drop the "-value" or the "imum")? It
> might blow up the DTB and string constant for finding the property?
> 
> If we come to some consensus on this, I can update the patch set.

Now when updating the patch set, I found that using ti,min/max-x/y was not really
a new invention. It is already upstream for a while for the ads7846/2046 driver!

This means: by introducing these properties to the tsc2007 as well we just extend the
already defined use of ti,min/max-x/y.

Simply changing that as well to use some minimum-raw-value-x etc. would break ads7846/2046
boards not in mainline or need ugly code to handle property alias names.

And since they are raw data and quite chip-specific, I think there is no urgent need
for harmonization of raw device data. While using the common bindings for output
scaling and fuzz, swap etc. is very important and helpful for board designers.

So I would prefer if we can keep it as is for the moment. And make property name
harmonization and evolution a separate task and discussion.

I will submit v4 in a couple of minutes. It contains some other small fixes as
suggested by the previous discussion.

BR and thanks,
Nikolaus

[toc] | [standalone]

Back to top | Article view | linux.kernel

csiph-web