From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Rafael J. Wysocki" Subject: Re: [PATCH] cpufreq: stats: do cpufreq_cpu_put() corresponding to cpufreq_cpu_get Date: Fri, 22 Mar 2013 12:47:35 +0100 Message-ID: <3472441.hhXSbXj7X9@vostro.rjw.lan> References: <234906745c34fb72b0c450d897f261a22ed18d40.1363947023.git.viresh.kumar@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset="utf-8" Content-Transfer-Encoding: 7Bit Return-path: In-Reply-To: Sender: cpufreq-owner@vger.kernel.org To: Viresh Kumar Cc: Amit Kucheria , Robin Randhawa , linux-pm@vger.kernel.org, Liviu Dudau , linux-kernel@vger.kernel.org, cpufreq@vger.kernel.org, Steve Bannister , Arvind Chauhan , Lists linaro-kernel , Charles Garcia-Tobin List-Id: linux-pm@vger.kernel.org On Friday, March 22, 2013 04:47:29 PM Viresh Kumar wrote: > On 22 March 2013 16:42, Amit Kucheria wrote: > > On Fri, Mar 22, 2013 at 3:43 PM, Viresh Kumar wrote: > >> In cpufreq_stats_free_sysfs() we aren't balancing calls to cpufreq_cpu_get() > >> with cpufreq_cpu_put(). This will never let us have ref count to policy->kobj as > >> zero. > > > > Rafael, > > > > Since this prevents booting on our hardware (we unregister and > > re-register the cpufreq driver to account for virtual cores), will > > this be considered as a hotfix for 3.9? > > > >> Signed-off-by: Viresh Kumar > > > > Acked-by: Amit Kucheria > > Sorry i forgot to mention, this should be pushed for next rc. Well, -rc5 is a realistic target. Thanks, Rafael -- I speak only for myself. Rafael J. Wysocki, Intel Open Source Technology Center.