From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753627AbbDBOpF (ORCPT ); Thu, 2 Apr 2015 10:45:05 -0400 Received: from mail-lb0-f175.google.com ([209.85.217.175]:36502 "EHLO mail-lb0-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752679AbbDBOpD (ORCPT ); Thu, 2 Apr 2015 10:45:03 -0400 MIME-Version: 1.0 In-Reply-To: <551D5352.8060305@nod.at> References: <1427984026-31100-1-git-send-email-richard@nod.at> <551D5352.8060305@nod.at> Date: Thu, 2 Apr 2015 11:45:00 -0300 Message-ID: Subject: Re: [PATCH] mtd: Add simple read disturb test From: Fabio Estevam To: Richard Weinberger Cc: Brian Norris , "linux-mtd@lists.infradead.org" , David Woodhouse , linux-kernel Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, Apr 2, 2015 at 11:33 AM, Richard Weinberger wrote: > Why? Free()ing a NULL pointer is perfectly fine. > What did I miss? :) If the first 'iobuf = kmalloc(mtd->erasesize, GFP_KERNEL);' fails then you jump to the out label where you call 5 kfree() and then return the error. It would be much better just to return the error immediately in this case and add one label for each allocation error, so that it only kfree the previous successful allocations.