cpufreq: Drop policy locking from cpufreq_policy_is_good_for_eas()

Policy locking was added to cpufreq_policy_is_good_for_eas() by commit
4854649 ("cpufreq/sched: Move cpufreq-specific EAS checks to
cpufreq") to address a theoretical race condition, but it turned out to
introduce a circular locking dependency between the policy rwsem and
sched_domains_mutex via cpuset_mutex. This leads to a board lockup on
OdroidN2 that is based on the ARM64 Amlogic Meson SoC.

Drop the policy locking from cpufreq_policy_is_good_for_eas() to address
this issue.

Fixes: 4854649 ("cpufreq/sched: Move cpufreq-specific EAS checks to cpufreq")
Closes: https://lore.kernel.org/linux-pm/[email protected]/
Reported-by: Marek Szyprowski <[email protected]>
Tested-by: Marek Szyprowski <[email protected]>
Signed-off-by: Rafael J. Wysocki <[email protected]>
Link: https://patch.msgid.link/[email protected]

Author: rafaeljw

drivers/cpufreq/cpufreq.c

@@ -3067,8 +3067,6 @@ static bool cpufreq_policy_is_good_for_eas(unsigned int cpu)
 		return false;
 	}
 
-	guard(cpufreq_policy_read)(policy);
-
 	return sugov_is_governor(policy);
 }