[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20171030134007.GA2846@bombadil.infradead.org>
Date: Mon, 30 Oct 2017 06:40:07 -0700
From: Matthew Wilcox <willy@...radead.org>
To: Masahiro Yamada <yamada.masahiro@...ionext.com>
Cc: Jonathan Corbet <corbet@....net>,
"open list:DOCUMENTATION" <linux-doc@...r.kernel.org>,
Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
Michal Marek <mmarek@...e.com>,
Linux Kbuild mailing list <linux-kbuild@...r.kernel.org>,
Matthew Wilcox <mawilcox@...rosoft.com>
Subject: Re: [PATCH] Check all .c files for bad kernel-doc comments
On Mon, Oct 30, 2017 at 12:40:20PM +0900, Masahiro Yamada wrote:
> 2017-10-28 4:41 GMT+09:00 Matthew Wilcox <willy@...radead.org>:
> > Implement a '-none' output mode for kernel-doc which will only output
> > warning messages, and suppresses the warning message about there being
> > no kernel-doc in the file. Add it to the rule to build .o files from .c
> > files, so it will check all .c files that have been modified.
> >
> > Adds about 1300 warnings to my build, but will hopefully discourage
> > people from introducing more kerneldoc mistakes.
>
> Basically, I think this is good,
> but it is controversial to sprinkle warnings by default.
Yes, it is. I just got three nastygrams from 01.org ;-)
But if it's not turned on by default, then people aren't going to notice
when they introduce new warnings. I think it needs to be up to someone
like Andrew or Linus to decide when to add these warnings by default.
Maybe we should do something like you have below for now, then work on
cleaning up a few hundred of these warnings, then enable this by default?
Thanks for looking at this!
> Maybe,
>
> ifeq ($(KBUILD_ENABLE_EXTRA_GCC_CHECKS),)
> cmd_checkdoc = $(srctree)/scripts/kernel-doc -none $< ;
> endif
>
>
> so that this is checked only when W=... is given?
Powered by blists - more mailing lists