Re: [PATCH v3 0/4] Introduce the latent_entropy gcc plugin

[Kees Cook]

On Wed, Jun 15, 2016 at 1:39 PM, Emese Revfy <re.emese@xxxxxxxxx> wrote:
> On Wed, 15 Jun 2016 11:55:44 -0700
> Kees Cook <keescook@xxxxxxxxxxxx> wrote:
>
>>  The limit on the length of lines is 80 columns and this is a strongly
>>  preferred limit.
>
> I think the code looks worse when it is truncated to 80 columns but
> I'll do it and resend the patches.

Yup, I understand your concerns, but since we're optimizing for
readability by a larger audience that has agreed to the guidelines in
CodingStyle, this is what we get. :)

One area I'm unclear on with kernel coding style, though, is if
splitting all the stuff prior to function name onto a separate line is
"acceptable", since that solves most of the long lines where
__latent_entropy has been added. For example, I don't know which is
better:

All on one line (gmail may split this, but my intention is all one line):

static __latent_entropy void rcu_process_callbacks(struct
softirq_action *unused)

Types and attributes on a separate line:

static __latent_entropy void
rcu_process_callbacks(struct softirq_action *unused)

All arguments on the next line:

static __latent_entropy void rcu_process_callbacks(
                                                          struct
softirq_action *unused)


Greg, do you have a better sense of how to split (or not split) these
kinds of long lines?

-Kees

-- 
Kees Cook
Chrome OS & Brillo Security