Re: [PATCH] Cleanup: replace prefered with preferred

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Wed, Oct 23, 2019 at 08:40:59AM -0700, Mark Salyzyn wrote:
> On 10/23/19 4:56 AM, Jarkko Sakkinen wrote:
> > On Tue, Oct 22, 2019 at 02:41:45PM -0700, Mark Salyzyn wrote:
> > > Replace all occurrences of prefered with preferred to make future
> > > checkpatch.pl's happy.  A few places the incorrect spelling is
> > > matched with the correct spelling to preserve existing user space API.
> > > 
> > > Signed-off-by: Mark Salyzyn <salyzyn@xxxxxxxxxxx>
> > I'd fix such things when the code is otherwise change and scope this
> > patch only to Documentation/. There is no pragmatic benefit of doing
> > this for the code.
> > 
> > /Jarkko
> 
> The pragmatic benefit comes with the use of an ABI/API checker (which is a
> 'distro' thing, not a top of tree kernel thing) produces its map which is
> typically required to be co-located in the same tree as the kernel
> repository. Quite a few ABI/API update checkins result in a checkpatch.pl
> complaint about the misspelled elements being (re-)recorded due to
> proximity. We have a separate task to improve how it is tracked in Android
> to reduce milepost marker changes that result in sweeping changes to the
> database which would reduce the occurrences.
> 
> I will split this between pure and inert documentation/comments for now,
> with a followup later for the code portion which understandably is more
> controversial.
> 
> Cleanup is the least appreciated part of kernel maintenance ;-}.
> 
> Sincerely -- Mark Salyzyn

I'm a strong believer of "evolutionary" approach. Patch sets for the
most part (everything in the end has to be considered case by case, not
a strict rule) should have some functional changes involved.

What I do require for the parts that I maintain is that any new change
will result cleaner code base than the one that existed before that
change was applied. Again, there are some exceptions to this e.g.
circulating a firmware bug but this is my driving guideline as a
maintainer.

Doing cleanups just for cleanups can sometimes add unnecessary merge
conflicts when backporting patches to stable kernels. Thus, if you are
doing just a cleanup you should have extremely good reasons to do so.

/Jarkko



[Index of Archives]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Linux Audio Users]     [Photo]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]

  Powered by Linux