From mboxrd@z Thu Jan 1 00:00:00 1970 From: Mark Brown Date: Mon, 26 Dec 2011 12:20:11 +0000 Subject: Re: [PATCH 1/9] drivers/mfd/88pm860x-i2c.c: introduce missing kfree Message-Id: <20111226122011.GG8722@opensource.wolfsonmicro.com> List-Id: References: <1324661974-17281-1-git-send-email-julia@diku.dk> <20111226120855.GB17760@sirena.org.uk> In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Julia Lawall Cc: Samuel Ortiz , kernel-janitors@vger.kernel.org, linux-kernel@vger.kernel.org On Mon, Dec 26, 2011 at 01:18:54PM +0100, Julia Lawall wrote: > On Mon, 26 Dec 2011, Mark Brown wrote: > >With this and probably a bunch of the other corrections in this series a > >conversion to devm_kazlloc() would be a more complete fix as it prevents > >people introducing similar missing cleanup paths in future. Not sure if > >spatch can generate that automatically though... > I saw a patch for that recently, and looked into it a little bit, > but I wasn't sure what should be done. What are the criteria for > using devm_kzalloc? It's good for any allocation that lasts for the lifetime of the device - most driver data allocated in probe() fits well, for example.