From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rob Herring Subject: Re: [PATCH 05/18] cpuidle: make a single register function for all Date: Wed, 10 Apr 2013 09:59:45 -0500 Message-ID: <51657E61.9010701@gmail.com> References: <1365603743-5618-1-git-send-email-daniel.lezcano@linaro.org> <1365603743-5618-6-git-send-email-daniel.lezcano@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Return-path: Received: from mail-qc0-f178.google.com ([209.85.216.178]:40260 "EHLO mail-qc0-f178.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S934984Ab3DJO7s (ORCPT ); Wed, 10 Apr 2013 10:59:48 -0400 Received: by mail-qc0-f178.google.com with SMTP id d10so232483qca.37 for ; Wed, 10 Apr 2013 07:59:48 -0700 (PDT) In-Reply-To: <1365603743-5618-6-git-send-email-daniel.lezcano@linaro.org> Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: Daniel Lezcano Cc: rjw@sisk.pl, andrew@lunn.ch, khilman@deeprootsystems.com, nsekhar@ti.com, josephl@nvidia.com, kgene.kim@samsung.com, patches@linaro.org, magnus.damm@gmail.com, tony@atomide.com, plagnioj@jcrosoft.com, linaro-kernel@lists.linaro.org, jason@lakedaemon.net, swarren@wwwdotorg.org, rob.herring@calxeda.com, horms@verge.net.au, linux@maxim.org.za, linux-arm-kernel@lists.infradead.org, deepthi@linux.vnet.ibm.com, jkosina@suse.cz, linux-pm@vger.kernel.org, lethal@linux-sh.org, kernel@pengutronix.de On 04/10/2013 09:22 AM, Daniel Lezcano wrote: > The usual scheme to initialize a cpuidle driver on a SMP is: > > cpuidle_register_driver(drv); > for_each_possible_cpu(cpu) { > device = &per_cpu(cpuidle_dev, cpu); > cpuidle_register_device(device); > } > > This code is duplicated in each cpuidle driver. > > On UP systems, it is done this way: > > cpuidle_register_driver(drv); > device = &per_cpu(cpuidle_dev, cpu); > cpuidle_register_device(device); > > On UP, the macro 'for_each_cpu' does one iteration: > > #define for_each_cpu(cpu, mask) \ > for ((cpu) = 0; (cpu) < 1; (cpu)++, (void)mask) > > Hence, the initialization loop is the same for UP than SMP. > > Beside, we saw different bugs / mis-initialization / return code unchecked in > the different drivers, the code is duplicated including bugs. After fixing all > these ones, it appears the initialization pattern is the same for everyone. > > Let's add a wrapper function doing this initialization with a cpumask parameter > for the coupled idle states and use it for all the drivers. > > That will save a lot of LOC, consolidate the code, and the modifications in the > future could be done in a single place. Another benefit is the consolidation of > the cpuidle_device variable which is now in the cpuidle framework and no longer > spread accross the different arch specific drivers. > > Signed-off-by: Daniel Lezcano > --- Perhaps my ack was too quick... > +int cpuidle_register(struct cpuidle_driver *drv, > + const struct cpumask *const coupled_cpus) > +{ > + int ret, cpu; > + struct cpuidle_device *device; > + > + ret = cpuidle_register_driver(drv); > + if (ret) { > + printk(KERN_ERR "failed to register cpuidle driver\n"); > + return ret; > + } > + > + for_each_possible_cpu(cpu) { > + device = &per_cpu(cpuidle_dev, cpu); > + device->cpu = cpu; > +#ifdef CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED > + device->coupled_cpus = *coupled_cpus; This needs a NULL check. CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED is most likely going to be enabled with multi-platform kernels and this will crash. Also, I think typically struct copies have a note that it is copy. Rob