Messages in this thread | | | Date | Tue, 7 Apr 2015 13:18:56 +0200 | From | Ingo Molnar <> | Subject | Re: about the flood of trivial patches and the Code of Conduct (was: Re: [PATCH 19/25] sched: Use bool function return values of true/false not 1/0) |
| |
* Greg Kroah-Hartman <gregkh@linuxfoundation.org> wrote:
> On Tue, Apr 07, 2015 at 11:12:46AM +0200, Ingo Molnar wrote: > > Pointing out this truth and protecting against such abusive flood of > > trivial patches is not against the code of conduct I signed. > > I totally agree, it's not "against" the code of conflict that I > helped write. > > Joe, you know better than to send trivial stuff to maintainers who > don't want it. Send it through the trivial maintainer for > subsystems that have expressed annoyance at this, it's not the first > time this has happened.
I argue that they should not be sent _at all_ in such cases, not even via the trivial tree: firstly because typically I'll pick up the bits from the trivial tree as well, and secondly because most of the arguments I listed against bulk trivial commits (weaker bisectability, taking up reviewer bandwidth, taking up Git space, etc.) still stand.
Frankly IMHO such a */25 series could be a net negative contribution when coming from a kernel contributor who has written 2000+ trivial patches already...
> Some maintainers, like me, are fine with your types of patches, I'd > stick to those subsystems if you like doing this type of work.
So sending trivial patches for things like totally unreadable code in say drivers/staging/ is probably OK, as they materially transform the code and make it more maintainable.
For the rest it can be more harmful than beneficial, for the reasons I outlined.
Thanks,
Ingo
| |