rcu: Make rcu_read_unlock_special() expedite strict grace periods
authorPaul E. McKenney <paulmck@kernel.org>
Wed, 27 Jan 2021 21:57:16 +0000 (13:57 -0800)
committerPaul E. McKenney <paulmck@kernel.org>
Mon, 8 Mar 2021 22:21:41 +0000 (14:21 -0800)
commit7308e0240410d3644c9d7cc6263079a58e3effeb
treea6586d7044ef9e1bac4c3779a87b75d119874717
parent5e59fba573e64cffc3a7a3113fff2336d652f45a
rcu: Make rcu_read_unlock_special() expedite strict grace periods

In kernels built with CONFIG_RCU_STRICT_GRACE_PERIOD=y, every grace
period is an expedited grace period.  However, rcu_read_unlock_special()
does not treat them that way, instead allowing the deferred quiescent
state to be reported whenever.  This commit therefore adds a check of
this Kconfig option that causes rcu_read_unlock_special() to treat all
grace periods as expedited for CONFIG_RCU_STRICT_GRACE_PERIOD=y kernels.

Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
kernel/rcu/tree_plugin.h