Re: share/private/slave a subtree - define vs enum

From: Pekka Enberg <penberg@cs.helsinki.fi>
To: Roman Zippel <zippel@linux-m68k.org>
Cc: Bryan Henderson <hbryan@us.ibm.com>,
	Andrew Morton <akpm@osdl.org>,
	bfields@fieldses.org, linux-fsdevel@vger.kernel.org,
	linux-kernel@vger.kernel.org, linuxram@us.ibm.com,
	mike@waychison.com, Miklos Szeredi <miklos@szeredi.hu>,
	Alexander Viro <viro@parcelfarce.linux.theplanet.co.uk>
Subject: Re: share/private/slave a subtree - define vs enum
Date: Sun, 10 Jul 2005 21:21:42 +0300	[thread overview]
Message-ID: <1121019702.20821.17.camel@localhost> (raw)
In-Reply-To: <Pine.LNX.4.61.0507082154090.3728@scrub.home>

Hi Roman,

At some point in time, I wrote:
> > Roman, it is not as if I get to decide for the patch submitters. I
> > comment on any issues _I_ have with the patch and the authors fix
> > whatever they want (or what the maintainers ask for).

On Fri, 2005-07-08 at 21:59 +0200, Roman Zippel wrote:
> The point of a review is to comment on things that _need_ fixing. Less 
> experienced hackers take this a requirement for their drivers to be 
> included.

Hmm. So we disagree on that issue as well. I think the point of review
is to improve code and help others conform with the existing coding
style which is why I find it strange that you're suggesting me to limit
my comments to a subset you agree with.

Would you please be so kind to define your criteria for things that
"need fixing" so we could see if can reach some sort of an agreement on
this. My list is roughly as follows:

  - Erroneous use of kernel API
  - Bad coding style
  - Layering violations
  - Duplicate code
  - Hard to read code

			Pekka