Re: [patch net-next v4 06/10] net: sched: introduce helpers to work with filter chains

[Joe Perches <joe@perches.com>]

On Wed, 2017-05-17 at 14:44 +0200, Jiri Pirko wrote:
> Wed, May 17, 2017 at 02:39:05PM CEST, jhs@mojatatu.com wrote:
> > On 17-05-17 08:25 AM, Jiri Pirko wrote:
> > > Wed, May 17, 2017 at 02:18:00PM CEST, jhs@mojatatu.com wrote:
> > > > On 17-05-17 05:07 AM, Jiri Pirko wrote:
> > > > > From: Jiri Pirko <jiri@mellanox.com>
> > > > > 
> > > > > Introduce struct tcf_chain object and set of helpers around it. Wraps up
> > > > > insertion, deletion and search in the filter chain.
> > > > > 
> > > > > Signed-off-by: Jiri Pirko <jiri@mellanox.com>
> > > > > ---
> > > > 
> > > > [..]
> > > > > +
> > > > > +static void
> > > > > +tcf_chain_filter_chain_ptr_set(struct tcf_chain *chain,
> > > > > +			       struct tcf_proto __rcu **p_filter_chain)
> > > > > +
> > > > 
> > > > What are the rules for this? Common coding style is:
> > > > static void tcf_chain_filter_chain_ptr_set(struct tcf_chain *chain,
> > > >                                           struct tcf_proto ..
> > > 
> > > When this would not fit 80 cols (this case), you need to wrap the
> > > text in front of the function name. That is exacly what I did.
> > > 
> > 
> > That i understand.
> > The question is: what does scripture dictate on conflict?
> > Should a function signature always follow coding style and
> > allow for exceeding 80 chars or the 80 chars rules trumps?
> 
> Definitelly 80 chars rules trumps here.

Disagree.
80 columns is just a "strongly preferred" limit.

Clarity for a human reader trumps everything else.