sched_ext: Fix deadlock between scx_root_disable() and concurrent forks

scx_root_disable() enters SCX_DISABLING before it grabs scx_enable_mutex to
clear __scx_switched_all and scx_switching_all. task_should_scx() short-circuits on DISABLING,
so forks in that window land on fair while next_active_class() still skips
fair - the new tasks stall.

This can deadlock the disable path itself: scx_alloc_and_add_sched() runs
under scx_enable_mutex and creates a helper kthread; if that new kthread is
one of the stalled fair tasks, the mutex holder waits forever and
scx_root_disable() can never make progress. Only sub-sched support exposes
this, since sub-sched enables are the only path where
scx_alloc_and_add_sched() can race the root's disable.

Move the DISABLING check after @scx_switching_all. @scx_switching_all
serves as a proxy for __scx_switched_all, so while it's set, forks keep
going to scx. Once cleared, DISABLING applies normally.

v2: Reword in-source comment and description. (Andrea)

Fixes: 337ec00b1d9c ("sched_ext: Implement cgroup sub-sched enabling and disabling")
Signed-off-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Andrea Righi <arighi@nvidia.com>

1 files changed, 21 insertions, 1 deletions

diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index a6d0a93d8174..547ca398f646 100644
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -4946,10 +4946,30 @@ static const struct kset_uevent_ops scx_uevent_ops = {
  */
 bool task_should_scx(int policy)
 {
-	if (!scx_enabled() || unlikely(scx_enable_state() == SCX_DISABLING))
+	/* if disabled, nothing should be on it */
+	if (!scx_enabled())
 		return false;
+
+	/* scx is taking over all SCHED_OTHER and SCHED_EXT tasks */
 	if (READ_ONCE(scx_switching_all))
 		return true;
+
+	/*
+	 * scx is tearing down - keep new SCHED_EXT tasks out.
+	 *
+	 * Must come after scx_switching_all test, which serves as a proxy
+	 * for __scx_switched_all. While __scx_switched_all is set, we must
+	 * return true via the branch above: a fork routed to fair would
+	 * stall because next_active_class() skips fair.
+	 *
+	 * This can develop into a deadlock - scx holds scx_enable_mutex across
+	 * kthread_create() in scx_alloc_and_add_sched(); if the new kthread is
+	 * the stalled task, the disable path can never grab the mutex to clear
+	 * scx_switching_all.
+	 */
+	if (unlikely(scx_enable_state() == SCX_DISABLING))
+		return false;
+
 	return policy == SCHED_EXT;
 }