Re: [Ksummit-discuss] [CORE TOPIC] checkpatch/Codingstyle and trivial patch spam

[Jonathan Cameron <jic23@jic23.retrosnub.co.uk>]

On 14 September 2016 18:11:07 BST, Alexandre Belloni <alexandre.belloni@free-electrons.com> wrote:
>On 13/09/2016 at 23:27:12 -0700, Joe Perches wrote :
>> On Wed, 2016-09-14 at 07:57 +0200, Julia Lawall wrote:
>> > What types of changes are unacceptable?
>> 
>> It's a mixed bag.
>> 
>> Some maintainers reject all "style/whitespace changes".
>> Some maintainers reject global consistency patches like
>> int -> bool conversions.
>> Some maintainers reject literal -> #define changes like
>> 1 -> true and 0 -> false for booleans.
>> 
>> Some of those maintainers are IMO misguided.
>
>On my side, I usually take that kind of changes only when they come
>with
>other substantial changes, especially when the code hasn't been touched
>for a while.
>
>The other thing that I find annoying are people using Coccinnelle or
>any
>other static code analysis tool and sending patches without saying how
>they found the alleged bug. Sometimes, this results in pointless
>cleanups that haven't been tested by the patch author.

As a maintainer who gets a lot of patches from newbies (partly as I maintain a fair
set of staging patches but also as we have a steady stream of new drivers in
IIO) I am quite happy to take 'good' whitespace cleanups.

Most of the time I can find something deeper to point out in the driver and get
them moving up the food chain.

Personally I think it is worth the pain in recent driver code at least as the
 advantage of pulling someone new in outweighs the costs.

Probably helps that IIO has a fair sized group of reviewers many of whom help
out with advice to newbies.  Hence I have it easy :)

Just thought I'd add a positive viewpoint.

Jonathan
-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.