Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Message-ID: <f68b7797aa73452d99508bdaf2801b3d141e7a69.camel@perches.com>
Subject: Re: clean up kernel_{read,write} & friends v2
From:   Joe Perches <joe@perches.com>
To:     Linus Torvalds <torvalds@linux-foundation.org>,
        Christoph Hellwig <hch@lst.de>
Cc:     Al Viro <viro@zeniv.linux.org.uk>, Ian Kent <raven@themaw.net>,
        David Howells <dhowells@redhat.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        linux-fsdevel <linux-fsdevel@vger.kernel.org>,
        LSM List <linux-security-module@vger.kernel.org>,
        NetFilter <netfilter-devel@vger.kernel.org>
Date:   Thu, 28 May 2020 12:22:08 -0700
In-Reply-To: <CAHk-=wj3iGQqjpvc+gf6+C29Jo4COj6OQQFzdY0h5qvYKTdCow@mail.gmail.com>
References: <20200528054043.621510-1-hch@lst.de>
         <CAHk-=wj3iGQqjpvc+gf6+C29Jo4COj6OQQFzdY0h5qvYKTdCow@mail.gmail.com>
Content-Type: text/plain; charset="ISO-8859-1"
User-Agent: Evolution 3.36.2-0ubuntu1 
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Thu, 2020-05-28 at 11:51 -0700, Linus Torvalds wrote:
> On Wed, May 27, 2020 at 10:40 PM Christoph Hellwig <hch@lst.de> wrote:
> > this series fixes a few issues and cleans up the helpers that read from
> > or write to kernel space buffers, and ensures that we don't change the
> > address limit if we are using the ->read_iter and ->write_iter methods
> > that don't need the changed address limit.
> 
> Apart from the "please don't mix irrelevant whitespace changes with
> other changes" comment, this looks fine to me.
> 
> And a rant related to that change: I'm really inclined to remove the
> checkpatch check for 80 columns entirely, but it shouldn't have been
> triggering for old lines even now.
> 
> Or maybe make it check for something more reasonable, like 100 characters.
> 
> I find it ironic and annoying how "checkpatch" warns about that silly
> legacy limit, when checkpatch itself then on the very next few lines
> has a line that is 124 columns wide

Yeah.  perl ain't c.

And this discussion has been had many times.

Here's one from 2009
https://lkml.org/lkml/2009/12/15/490

Another from 2012
https://lkml.org/lkml/2012/2/5/141

Line lengths checks are normally pretty silly.

Hard limits at 80 really don't work well, especially with
some of the 25+ character length identifiers used today.

I think a line length warning at 132 is generally reasonable
but it could depend on complexity and identifier lengths.

> And yes, that 124 character line has a good reason for it. But that's
> kind of the point. There are lots of perfectly fine reasons for longer
> lines.
> 
> I'd much rather check for "no deep indentation" or "no unnecessarily
> complex conditionals" or other issues that are more likely to be
> _real_ problems.

That deep indentation test already exists at 6 tabs.
Maybe it should be 5 instead.  Or maybe even 4, but
that's a pretty easy to need and common use case.

Tab depth use in the kernel is more or less

$ git grep -Poh '^\t+(if|do|while|for|switch)\b' | \
  sed -r 's/\w+//g' | \
  awk '{print length($0);}' | \
  sort | uniq -c | sort -rn
 903993 1
 339059 2
  89334 3
  18216 4
   3282 5
    605 6
    148 7
     36 8
      4 9
      1 10

> But do we really have 80x25 terminals any more that
> we'd care about?

trivial btw: VT100s were 80x24 or 132x24, PCs were 80x25