From 316706beeb9c64804201fd12d6777b9c638c4a16 Mon Sep 17 00:00:00 2001 Message-Id: <316706beeb9c64804201fd12d6777b9c638c4a16.1594742966.git.zanussi@kernel.org> In-Reply-To: <832f7d97d6b989a5b4860dd2dbec58ad6ad5ab81.1594742966.git.zanussi@kernel.org> References: <832f7d97d6b989a5b4860dd2dbec58ad6ad5ab81.1594742966.git.zanussi@kernel.org> From: Mike Galbraith Date: Sun, 19 Aug 2018 08:28:35 +0200 Subject: [PATCH 229/329] sched: Allow pinned user tasks to be awakened to the CPU they pinned Origin: https://www.kernel.org/pub/linux/kernel/projects/rt/4.19/older/patches-4.19.132-rt59.tar.xz Since commit 7af443ee16976 ("sched/core: Require cpu_active() in select_task_rq(), for user tasks") select_fallback_rq() will BUG() if the CPU to which a task has pinned itself and pinned becomes !cpu_active() while it slept. The task will continue running on the to-be-removed CPU and will remove itself from the CPU during takedown_cpu() (while cpuhp_pin_lock will be acquired) and move to another CPU based on its mask after the migrate_disable() section has been left. Cc: stable-rt@vger.kernel.org Signed-off-by: Mike Galbraith Signed-off-by: Sebastian Andrzej Siewior --- kernel/sched/core.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 1abd32af7918..173557e65689 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -904,7 +904,7 @@ static inline bool is_cpu_allowed(struct task_struct *p, int cpu) if (!cpumask_test_cpu(cpu, p->cpus_ptr)) return false; - if (is_per_cpu_kthread(p)) + if (is_per_cpu_kthread(p) || __migrate_disabled(p)) return cpu_online(cpu); return cpu_active(cpu); -- 2.17.1