From mboxrd@z Thu Jan 1 00:00:00 1970 From: Paul Mundt Date: Thu, 11 Dec 2008 00:59:29 +0000 Subject: Re: [PATCH 01/05] clocksource: add read2() callback and warning Message-Id: <20081211005929.GA9377@linux-sh.org> List-Id: References: <20081210151243.32515.61254.sendpatchset@rx1.opensource.se> In-Reply-To: <20081210151243.32515.61254.sendpatchset@rx1.opensource.se> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: linux-sh@vger.kernel.org On Wed, Dec 10, 2008 at 09:47:40PM +0100, Thomas Gleixner wrote: > On Wed, 10 Dec 2008, john stultz wrote: > > I do somewhat agree with Thomas that maybe just fixing all the in tree > > clocksources instead of creating the dual interface is the way to go. > > But I recognize as an internal interface we may want to deprecate it and > > let out of tree users have some time to convert. So I could go either > > way. > > Come on, the function takes now an argument with no further changes to > the function code itself. > > It's not a complex change which involves refactoring of code. > I agree, we've done this with kmem_cache_create() and friends plenty of times, which are far more invasive, yet have never been a problem in converting in one shot. The incremental move approach seems to mostly work for modules where people are primarily concerned about out-of-tree code. As modules aren't really a factor here that basically leaves out-of-tree code as a consideration, which is not something that should be stopping us from making these API changes. It's not a difficult change to make to out-of-tree code in any event.