PM / OPP: Pass opp_table to dev_pm_opp_put_regulator()
authorStephen Boyd <sboyd@codeaurora.org>
Wed, 30 Nov 2016 10:51:25 +0000 (16:21 +0530)
committerRafael J. Wysocki <rafael.j.wysocki@intel.com>
Wed, 30 Nov 2016 21:41:28 +0000 (22:41 +0100)
commit91291d9ad92faa65a56a9a19d658d8049b78d3d4
tree85aaf772e49e3663b1595fc21b57f93c61834ffd
parent349aa92e81bd14552d8d2335aff490f306038603
PM / OPP: Pass opp_table to dev_pm_opp_put_regulator()

Joonyoung Shim reported an interesting problem on his ARM octa-core
Odoroid-XU3 platform. During system suspend, dev_pm_opp_put_regulator()
was failing for a struct device for which dev_pm_opp_set_regulator() is
called earlier.

This happened because an earlier call to
dev_pm_opp_of_cpumask_remove_table() function (from cpufreq-dt.c file)
removed all the entries from opp_table->dev_list apart from the last CPU
device in the cpumask of CPUs sharing the OPP.

But both dev_pm_opp_set_regulator() and dev_pm_opp_put_regulator()
routines get CPU device for the first CPU in the cpumask. And so the OPP
core failed to find the OPP table for the struct device.

This patch attempts to fix this problem by returning a pointer to the
opp_table from dev_pm_opp_set_regulator() and using that as the
parameter to dev_pm_opp_put_regulator(). This ensures that the
dev_pm_opp_put_regulator() doesn't fail to find the opp table.

Note that similar design problem also exists with other
dev_pm_opp_put_*() APIs, but those aren't used currently by anyone and
so we don't need to update them for now.

Cc: 4.4+ <stable@vger.kernel.org> # 4.4+
Reported-by: Joonyoung Shim <jy0922.shim@samsung.com>
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
[ Viresh: Wrote commit log and tested on exynos 5250 ]
Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
drivers/base/power/opp/core.c
drivers/cpufreq/cpufreq-dt.c
include/linux/pm_opp.h