From mboxrd@z Thu Jan 1 00:00:00 1970 From: Greg KH Subject: Re: linux-next: build failure after merge of the driver-core tree Date: Wed, 17 Feb 2010 20:27:11 -0800 Message-ID: <20100218042711.GC11649@kroah.com> References: <20100216173926.b7a4d25c.sfr@canb.auug.org.au> <20100216100100.55f5922f@hyperion.delvare> <20100217045616.GA2923@kroah.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Return-path: Content-Disposition: inline In-Reply-To: <20100217045616.GA2923@kroah.com> Sender: linux-kernel-owner@vger.kernel.org To: Jean Delvare Cc: Stephen Rothwell , linux-next@vger.kernel.org, linux-kernel@vger.kernel.org, Thomas Gleixner List-Id: linux-next.vger.kernel.org On Tue, Feb 16, 2010 at 08:56:16PM -0800, Greg KH wrote: > On Tue, Feb 16, 2010 at 10:01:00AM +0100, Jean Delvare wrote: > > Hi Stephen, > > > > On Tue, 16 Feb 2010 17:39:26 +1100, Stephen Rothwell wrote: > > > After merging the driver-core tree, today's linux-next build (x86_64 > > > allmodconfig) failed like this: > > > > > > drivers/i2c/i2c-smbus.c:58: error: 'struct device' has no member named 'sem' > > > drivers/i2c/i2c-smbus.c:66: error: 'struct device' has no member named 'sem' > > > > > > Caused by commit 0819a881db29059d113fc5f019f7c489dea5937d from the > > > driver-core tree interacting with commit > > > 38f1e8aedf0a238d56fbcd3660fc140b50dbc89a ("i2c: Add SMBus alert support") > > > from the i2c tree. > > > > > > I have applied the following patch for today and will use it as a merge > > > fixup for the driver-core tree while necessary. > > > > Thanks for the heads up and the fix. Greg, is there an API I am > > supposed to use instead of accessing the device's semaphore/mutex > > directly? > > I will create a device_lock() and device_unlock() api to solve this > problem. That will fix the issue for when we try to change this from a > semaphore to a mutex (like the patch in linux-next does), but we need to > revert it due to lockdep issues that are being reported incorrectly. > > For now, your code is fine, I'll change over your code in the .34 tree > when the new api goes in. I'll also drop this mutex patch in a day as > well, because of the problems it has caused. The functions are now in my tree, I'll fix up your code in the .34 merge period. thanks, greg k-h