Messages in this thread | | | Date | Fri, 22 Mar 2013 17:40:25 +0530 | Subject | Re: [PATCH] cpufreq: stats: do cpufreq_cpu_put() corresponding to cpufreq_cpu_get | From | Viresh Kumar <> |
| |
On 22 March 2013 17:42, Rafael J. Wysocki <rjw@sisk.pl> wrote: > Which would be useful to write in the changelog, wouldn't it?
Hmm..
copy-paste with gmail is also broken, so find it attached too.
New change log, no change in patch and you can trust me on that :)
----------x-------------x--------
From 034e5ac4cccd09872592a46decd38d5f24047f10 Mon Sep 17 00:00:00 2001 Message-Id: <034e5ac4cccd09872592a46decd38d5f24047f10.1363954124.git.viresh.kumar@linaro.org> From: Viresh Kumar <viresh.kumar@linaro.org> Date: Fri, 22 Mar 2013 15:15:48 +0530 Subject: [PATCH] cpufreq: stats: do cpufreq_cpu_put() corresponding to cpufreq_cpu_get
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.
We will get a hang if somehow cpufreq_driver_unregister() is called. And that can happen when we compile our driver as module and insmod/rmmod it.
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org> [unhandled content-type:application/octet-stream] | |