From: "Paul E. McKenney" paulmck@kernel.org

commit 4d60b475f858ebdb06c1339f01a890f287b5e587 upstream.

The rcu_cpu_starting() and rcu_report_dead() functions transition the current CPU between online and offline state from an RCU perspective. Unfortunately, this means that the rcu_cpu_starting() function's lock acquisition and the rcu_report_dead() function's lock releases happen while the CPU is offline from an RCU perspective, which can result in lockdep-RCU splats about using RCU from an offline CPU. And this situation can also result in too-short grace periods, especially in guest OSes that are subject to vCPU preemption.

This commit therefore uses sequence-count-like synchronization to forgive use of RCU while RCU thinks a CPU is offline across the full extent of the rcu_cpu_starting() and rcu_report_dead() function's lock acquisitions and releases.

One approach would have been to use the actual sequence-count primitives provided by the Linux kernel. Unfortunately, the resulting code looks completely broken and wrong, and is likely to result in patches that break RCU in an attempt to address this appearance of broken wrongness. Plus there is no net savings in lines of code, given the additional explicit memory barriers required.

Therefore, this sequence count is instead implemented by a new ->ofl_seq field in the rcu_node structure. If this counter's value is an odd number, RCU forgives RCU read-side critical sections on other CPUs covered by the same rcu_node structure, even if those CPUs are offline from an RCU perspective. In addition, if a given leaf rcu_node structure's ->ofl_seq counter value is an odd number, rcu_gp_init() delays starting the grace period until that counter value changes.

[ paulmck: Apply Peter Zijlstra feedback. ] Signed-off-by: Paul E. McKenney paulmck@kernel.org Cc: stable@vger.kernel.org # 5.10.x Signed-off-by: Joel Fernandes (Google) joel@joelfernandes.org --- kernel/rcu/tree.c | 21 ++++++++++++++++++++- kernel/rcu/tree.h | 1 + 2 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c index 3fe7c75c371b..9cce4e13af41 100644 --- a/kernel/rcu/tree.c +++ b/kernel/rcu/tree.c @@ -1157,7 +1157,7 @@ bool rcu_lockdep_current_cpu_online(void) preempt_disable_notrace(); rdp = this_cpu_ptr(&rcu_data); rnp = rdp->mynode; - if (rdp->grpmask & rcu_rnp_online_cpus(rnp)) + if (rdp->grpmask & rcu_rnp_online_cpus(rnp) || READ_ONCE(rnp->ofl_seq) & 0x1) ret = true; preempt_enable_notrace(); return ret; @@ -1724,6 +1724,7 @@ static void rcu_strict_gp_boundary(void *unused) */ static bool rcu_gp_init(void) { + unsigned long firstseq; unsigned long flags; unsigned long oldmask; unsigned long mask; @@ -1767,6 +1768,12 @@ static bool rcu_gp_init(void) */ rcu_state.gp_state = RCU_GP_ONOFF; rcu_for_each_leaf_node(rnp) { + smp_mb(); // Pair with barriers used when updating ->ofl_seq to odd values. + firstseq = READ_ONCE(rnp->ofl_seq); + if (firstseq & 0x1) + while (firstseq == READ_ONCE(rnp->ofl_seq)) + schedule_timeout_idle(1); // Can't wake unless RCU is watching. + smp_mb(); // Pair with barriers used when updating ->ofl_seq to even values. raw_spin_lock(&rcu_state.ofl_lock); raw_spin_lock_irq_rcu_node(rnp); if (rnp->qsmaskinit == rnp->qsmaskinitnext && @@ -4107,6 +4114,9 @@ void rcu_cpu_starting(unsigned int cpu)

rnp = rdp->mynode; mask = rdp->grpmask; + WRITE_ONCE(rnp->ofl_seq, rnp->ofl_seq + 1); + WARN_ON_ONCE(!(rnp->ofl_seq & 0x1)); + smp_mb(); // Pair with rcu_gp_cleanup()'s ->ofl_seq barrier(). raw_spin_lock_irqsave_rcu_node(rnp, flags); WRITE_ONCE(rnp->qsmaskinitnext, rnp->qsmaskinitnext | mask); newcpu = !(rnp->expmaskinitnext & mask); @@ -4124,6 +4134,9 @@ void rcu_cpu_starting(unsigned int cpu) } else { raw_spin_unlock_irqrestore_rcu_node(rnp, flags); } + smp_mb(); // Pair with rcu_gp_cleanup()'s ->ofl_seq barrier(). + WRITE_ONCE(rnp->ofl_seq, rnp->ofl_seq + 1); + WARN_ON_ONCE(rnp->ofl_seq & 0x1); smp_mb(); /* Ensure RCU read-side usage follows above initialization. */ }

@@ -4150,6 +4163,9 @@ void rcu_report_dead(unsigned int cpu)

/* Remove outgoing CPU from mask in the leaf rcu_node structure. */ mask = rdp->grpmask; + WRITE_ONCE(rnp->ofl_seq, rnp->ofl_seq + 1); + WARN_ON_ONCE(!(rnp->ofl_seq & 0x1)); + smp_mb(); // Pair with rcu_gp_cleanup()'s ->ofl_seq barrier(). raw_spin_lock(&rcu_state.ofl_lock); raw_spin_lock_irqsave_rcu_node(rnp, flags); /* Enforce GP memory-order guarantee. */ rdp->rcu_ofl_gp_seq = READ_ONCE(rcu_state.gp_seq); @@ -4162,6 +4178,9 @@ void rcu_report_dead(unsigned int cpu) WRITE_ONCE(rnp->qsmaskinitnext, rnp->qsmaskinitnext & ~mask); raw_spin_unlock_irqrestore_rcu_node(rnp, flags); raw_spin_unlock(&rcu_state.ofl_lock); + smp_mb(); // Pair with rcu_gp_cleanup()'s ->ofl_seq barrier(). + WRITE_ONCE(rnp->ofl_seq, rnp->ofl_seq + 1); + WARN_ON_ONCE(rnp->ofl_seq & 0x1);

rdp->cpu_started = false; } diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h index e4f66b8f7c47..6e8c77729a47 100644 --- a/kernel/rcu/tree.h +++ b/kernel/rcu/tree.h @@ -56,6 +56,7 @@ struct rcu_node { /* Initialized from ->qsmaskinitnext at the */ /* beginning of each grace period. */ unsigned long qsmaskinitnext; + unsigned long ofl_seq; /* CPU-hotplug operation sequence count. */ /* Online CPUs for next grace period. */ unsigned long expmask; /* CPUs or groups that need to check in */ /* to allow the current expedited GP */