From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Wed, 18 Jun 2008 05:04:52 -0700 (PDT) Received: from cuda.sgi.com ([192.48.176.15]) by oss.sgi.com (8.12.11.20060308/8.12.11/SuSE Linux 0.7) with ESMTP id m5IC4n83017683 for ; Wed, 18 Jun 2008 05:04:49 -0700 Date: Wed, 18 Jun 2008 14:05:36 +0200 From: Christoph Hellwig Subject: Re: [PATCH] kill INDUCE_IO_ERROR Message-ID: <20080618120535.GA32691@lst.de> References: <20080523062323.GA32637@lst.de> <20080616101412.GA17939@lst.de> <485892D7.4080000@sgi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <485892D7.4080000@sgi.com> Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: Donald Douwsma Cc: Christoph Hellwig , xfs@oss.sgi.com On Wed, Jun 18, 2008 at 02:45:11PM +1000, Donald Douwsma wrote: > Christoph Hellwig wrote: > >On Fri, May 23, 2008 at 08:23:23AM +0200, Christoph Hellwig wrote: > >>All the error injection is already enabled through ifdef DEBUG, so kill > >>the never set second cpp symbol to activate it without the rest of the > >>debugging infrastructure. > > > >Can I get a review on this absolutely trivial patch? > > I'd like to keep INDUCE_IO_ERROR around so we have a way to introduce > ioerrors in release builds during testing. > > We could add it to xfs.h with the commented out QUOTADEBUG so people > know we still want to use it. cpp symbols not exposed by Kconfig will never get tested. If you thing it's really useful for release builds (which don't quite agree with) we should just enable it unconditionally. It's just in a few places, and any branch predictor worth it's name should take care of the additional branches without visible cost.