Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753421Ab2EMUkI (ORCPT ); Sun, 13 May 2012 16:40:08 -0400 Received: from mail-ob0-f174.google.com ([209.85.214.174]:56450 "EHLO mail-ob0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751258Ab2EMUkG (ORCPT ); Sun, 13 May 2012 16:40:06 -0400 Date: Sun, 13 May 2012 13:38:36 -0700 From: Anton Vorontsov To: Dan Carpenter Cc: Greg Kroah-Hartman , Kees Cook , Colin Cross , devel@driverdev.osuosl.org, linaro-kernel@lists.linaro.org, Arnd Bergmann , patches@linaro.org, Marco Stornelli , Stephen Boyd , linux-kernel@vger.kernel.org, arve@android.com, Jesper Juhl , John Stultz , Rebecca Schultz Zavin , WANG Cong , Andrew Morton , kernel-team@android.com, Thomas Meyer Subject: Re: [PATCH 02/11] persistent_ram: Fix buffer size clamping during writes Message-ID: <20120513203835.GA30082@lizard> References: <20120512001506.GA8653@lizard> <20120512001717.GB14782@lizard> <20120513165601.GC16984@mwanda> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20120513165601.GC16984@mwanda> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1577 Lines: 38 On Sun, May 13, 2012 at 07:56:01PM +0300, Dan Carpenter wrote: > On Fri, May 11, 2012 at 05:17:17PM -0700, Anton Vorontsov wrote: > > This is a longstanding bug, almost unnoticeable when calling > > persistent_ram_write() for small buffers. > > > > But when called for large data buffers, the write routine behaves > > incorrectly, as the size may never update: instead of clamping > > the size to the maximum buffer size, buffer_size_add_clamp() returns > > an error (which is never checked by the write routine, btw). > > > > To fix this, we now use buffer_size_add() that actually clamps the > > size to the max value. > > > > Also remove buffer_size_add_clamp(), it is no longer needed. > > > > Say if you did notice it, what would that look like? It's just that > something gets lost instead of written to the screen right? Yep. Suppose the ring buffer size is 4096 bytes, when somebody tries to write a data in a 2000 bytes chunk, the first write will succeed (buffer size will be 2000), but the second now 3000-bytes write will left the size equal to 2000, instead of clamping it to 4096. When we had a large buffer but a small writes (e.g. ram_console usage scenario), this is almost unnoticeable. But when we started using large writes the bug showed up. Thanks, -- Anton Vorontsov Email: cbouatmailru@gmail.com -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/