Messages in this thread Patch in this message | | | From | "Paul E. McKenney" <> | Subject | [PATCH rcu 01/10] rcu: Fix rcu_read_unlock_strict() strict QS reporting | Date | Wed, 31 Aug 2022 11:07:56 -0700 |
| |
From: Zqiang <qiang1.zhang@intel.com>
Kernels built with CONFIG_PREEMPT=n and CONFIG_RCU_STRICT_GRACE_PERIOD=y report the quiescent state directly from the outermost rcu_read_unlock(). However, the current CPU's rcu_data structure's ->cpu_no_qs.b.norm might still be set, in which case rcu_report_qs_rdp() will exit early, thus failing to report quiescent state.
This commit therefore causes rcu_read_unlock_strict() to clear CPU's rcu_data structure's ->cpu_no_qs.b.norm field before invoking rcu_report_qs_rdp().
Signed-off-by: Zqiang <qiang1.zhang@intel.com> Signed-off-by: Paul E. McKenney <paulmck@kernel.org> --- kernel/rcu/tree_plugin.h | 1 + 1 file changed, 1 insertion(+)
diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h index 438ecae6bd7e7..86772c95ed0ae 100644 --- a/kernel/rcu/tree_plugin.h +++ b/kernel/rcu/tree_plugin.h @@ -824,6 +824,7 @@ void rcu_read_unlock_strict(void) if (irqs_disabled() || preempt_count() || !rcu_state.gp_kthread) return; rdp = this_cpu_ptr(&rcu_data); + rdp->cpu_no_qs.b.norm = false; rcu_report_qs_rdp(rdp); udelay(rcu_unlock_delay); } -- 2.31.1.189.g2e36527f23
| |