From owner-svn-src-head@FreeBSD.ORG  Wed Dec 21 23:27:54 2011
Return-Path: <owner-svn-src-head@FreeBSD.ORG>
Delivered-To: svn-src-head@freebsd.org
Received: by hub.freebsd.org (Postfix, from userid 1233)
	id A486F1065675; Wed, 21 Dec 2011 23:27:54 +0000 (UTC)
Date: Wed, 21 Dec 2011 23:27:54 +0000
From: Alexander Best <arundel@freebsd.org>
To: John Baldwin <jhb@freebsd.org>
Message-ID: <20111221232754.GA51331@freebsd.org>
References: <201112211716.pBLHGhDH078507@svn.freebsd.org>
	<CAJ-Vmont2wOF=TkT1CmZt4Gp0iOc7ZYmfQpjW=8j1=F5DQGNwQ@mail.gmail.com>
	<201112211700.42772.jhb@freebsd.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <201112211700.42772.jhb@freebsd.org>
Cc: svn-src-head@freebsd.org, Adrian Chadd <adrian@freebsd.org>,
	src-committers@freebsd.org, Dimitry Andric <dim@freebsd.org>,
	svn-src-all@freebsd.org
Subject: Re: svn commit: r228785 - in head/sys/dev/ath/ath_hal: ar5210 ar5211
X-BeenThere: svn-src-head@freebsd.org
X-Mailman-Version: 2.1.5
Precedence: list
List-Id: SVN commit messages for the src tree for head/-current
	<svn-src-head.freebsd.org>
List-Unsubscribe: <http://lists.freebsd.org/mailman/listinfo/svn-src-head>,
	<mailto:svn-src-head-request@freebsd.org?subject=unsubscribe>
List-Archive: <http://lists.freebsd.org/pipermail/svn-src-head>
List-Post: <mailto:svn-src-head@freebsd.org>
List-Help: <mailto:svn-src-head-request@freebsd.org?subject=help>
List-Subscribe: <http://lists.freebsd.org/mailman/listinfo/svn-src-head>,
	<mailto:svn-src-head-request@freebsd.org?subject=subscribe>
X-List-Received-Date: Wed, 21 Dec 2011 23:27:54 -0000

On Wed Dec 21 11, John Baldwin wrote:
> On Wednesday, December 21, 2011 4:52:04 pm Adrian Chadd wrote:
> > Erm, why did you do this without first getting clearance from someone
> > who has the hardware to test it?
> > 
> > Just because it looks obviously wrong to you, doesn't at all mean that
> > it's "wrong". It's quite possible that the driver _requires_ those
> > bits to be written to the hardware as 0.
> > 
> > 
> > I'd appreciate it if would please revert this and other ath/hal
> > changes until I've had time to research them and test them out.
> 
> I agree it should be reviewed, but if you are seriously depending on
> the fact that the shifted values are beyond the edge of the word boundary
> and so the result "wraps" to zero, then I'd question the sanity of your code.

i disagree.

the commits should stay. after all this is HEAD. this way all developers
running HEAD and with the appropriate ath hardware will test the changes. if
dim@ really broke something, people will notice.

the changes should *not* be MFC'ed. but if no one complains until 10.0-RELEASE,
it's very unlikely he broke something.

plus...does it make a different, if you test HEAD with the changes present or
test HEAD with the changes integrated via patches?

one way or the other you can test the changes and *if* something broke, the
commits can be reverted.

cheers.
alex

> 
> -- 
> John Baldwin