diff options
| author |   Zqiang <qiang1.zhang@intel.com> | 2022-04-04 07:59:32 +0800 |
|---|---|---|
| committer |   Paul E. McKenney <paulmck@kernel.org> | 2022-04-20 16:51:11 -0700 |
| commit | f596e2ce1c0f250bb3ecc179f611be37e862635f (patch) | |
| tree | f1757f53792ddda51ed036798280e74eeda820ee /kernel/rcu | |
| parent | rcu_sync: Fix comment to properly reflect rcu_sync_exit() behavior (diff) | |
| download | linux-dev-f596e2ce1c0f250bb3ecc179f611be37e862635f.tar.xz linux-dev-f596e2ce1c0f250bb3ecc179f611be37e862635f.zip | |
rcu: Use IRQ_WORK_INIT_HARD() to avoid rcu_read_unlock() hangs
When booting kernels built with both CONFIG_RCU_STRICT_GRACE_PERIOD=y
and CONFIG_PREEMPT_RT=y, the rcu_read_unlock_special() function's
invocation of irq_work_queue_on() the init_irq_work() causes the
rcu_preempt_deferred_qs_handler() function to work execute in SCHED_FIFO
irq_work kthreads. Because rcu_read_unlock_special() is invoked on each
rcu_read_unlock() in such kernels, the amount of work just keeps piling
up, resulting in a boot-time hang.
This commit therefore avoids this hang by using IRQ_WORK_INIT_HARD()
instead of init_irq_work(), but only in kernels built with both
CONFIG_PREEMPT_RT=y and CONFIG_RCU_STRICT_GRACE_PERIOD=y.
Signed-off-by: Zqiang <qiang1.zhang@intel.com>
Signed-off-by: Paul E. McKenney <paulmck@kernel.org>
Diffstat (limited to 'kernel/rcu')
| -rw-r--r-- | kernel/rcu/tree_plugin.h | 8 |
1 files changed, 7 insertions, 1 deletions
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index 5c23aceecd62..2a3715419073 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -661,7 +661,13 @@ static void rcu_read_unlock_special(struct task_struct *t) expboost && !rdp->defer_qs_iw_pending && cpu_online(rdp->cpu)) { // Get scheduler to re-evaluate and call hooks. // If !IRQ_WORK, FQS scan will eventually IPI. - init_irq_work(&rdp->defer_qs_iw, rcu_preempt_deferred_qs_handler); + if (IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD) && + IS_ENABLED(CONFIG_PREEMPT_RT)) + rdp->defer_qs_iw = IRQ_WORK_INIT_HARD( + rcu_preempt_deferred_qs_handler); + else + init_irq_work(&rdp->defer_qs_iw, + rcu_preempt_deferred_qs_handler); rdp->defer_qs_iw_pending = true; irq_work_queue_on(&rdp->defer_qs_iw, rdp->cpu); } |