public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Stefan Richter <stefanr@s5r6.in-berlin.de>
To: Ingo Molnar <mingo@elte.hu>
Cc: Sam Ravnborg <sam@ravnborg.org>,
	Manish Katiyar <mkatiyar@gmail.com>,
	LKML <linux-kernel@vger.kernel.org>,
	kernel-janitors@vger.kernel.org
Subject: Re: [PATCH] Remove errors caught by checkpatch.pl in	kernel/kallsyms.c
Date: Mon, 16 Feb 2009 16:22:41 +0100	[thread overview]
Message-ID: <499984C1.6020004@s5r6.in-berlin.de> (raw)
In-Reply-To: <20090216141917.GA8981@elte.hu>

Ingo Molnar wrote:
> We routinely mention Sparse, lockdep, Coverity, Coccinelle, kmemleak, 
> ftrace, kmemcheck and other tools as well when it motives to fix a bug 
> or uncleanliness. [...] It is absolutely fine to
> mention checkpatch when it catches uncleanliness in code that already 
> got merged. I dont understand your point.

I wrote "don't mention checkpatch" but I really meant "think about what
the effect of the patch is and describe this".

It's not really a hard problem to mention checkpatch --- it is a problem
to make it the main point or, like in this case, the only point of the
changelog.  (Furthermore, it is also a problem to do something routinely
*if* doing it does not make sense.  There routinely appear coccinelle
metapatch sources in changelogs.  That does not make sense at all, and
doing it routinely is not a justification in itself.)

So, "don't mention checkpatch" is simply a rule of thumb; read it as "I
mentioned checkpatch in the changelog --- wait, I have possibly written
a changelog that is besides the point; I should think about it once
more".  :-)

Now, when this particular patch is updated to get a good changelog, then
the title could become e.g.:
	kernel/kallsyms: change initcall level; adjust whitespace
and anything more than that is just fluff and wasted electrons. Actually
the changelog should rather contain a note on why device_initcall is
supposed to be the correct initcall level.

Fixes due to reports from sparse, lockdep, coverity, coccinelle, etc.
are the in this respect the same as fixes due to reports from
checkpatch:  Patch titles should for example be
  - "fix potential deadlock..."
  - "fix use-after free..."
  - "use XYZ helper..."
  - "adjust whitespace..."
and *not* something like "fix lockdep backtrace" or whatever.

A difference would be a patch title like "add sparse annotations"
because this is indeed about what the patch does, not by which means it
was created.

Why do I make a fuzz?  Well, because many of our changelogs really suck
and we need to become better in general.
-- 
Stefan Richter
-=====-=-=== -=-= -==-=
http://arcgraph.de/sr/

  reply	other threads:[~2009-02-16 15:22 UTC|newest]

Thread overview: 20+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2009-02-15 18:34 [PATCH] Remove errors caught by checkpatch.pl in kernel/kallsyms.c Manish Katiyar
2009-02-15 18:47 ` Sam Ravnborg
2009-02-15 18:47   ` Manish Katiyar
2009-02-16 13:07   ` Stefan Richter
2009-02-16 13:28     ` Ingo Molnar
2009-02-16 14:00       ` Stefan Richter
2009-02-16 14:19         ` Ingo Molnar
2009-02-16 15:22           ` Stefan Richter [this message]
2009-02-16 15:41             ` Manish Katiyar
2009-02-16 15:50             ` Ingo Molnar
2009-02-16 16:13               ` Stefan Richter
2009-02-16 17:12                 ` Ingo Molnar
2009-02-16 18:04                   ` Stefan Richter
2009-02-16 16:13               ` Al Viro
2009-02-16 17:11                 ` Ingo Molnar
2009-02-16 14:28       ` Paolo Ciarrocchi
2009-02-16 16:17       ` Julia Lawall
2009-02-16 16:35         ` Stefan Richter
2009-02-16 17:21           ` Ingo Molnar
2009-02-16 17:15         ` Ingo Molnar

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=499984C1.6020004@s5r6.in-berlin.de \
    --to=stefanr@s5r6.in-berlin.de \
    --cc=kernel-janitors@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=mkatiyar@gmail.com \
    --cc=sam@ravnborg.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox