linux-kernel - [PATCH 4/6] timers/migration: Assert that hotplug preparing CPU is part of stable active hierarchy

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20251024132536.39841-5-frederic@kernel.org>
Date: Fri, 24 Oct 2025 15:25:34 +0200
From: Frederic Weisbecker <frederic@...nel.org>
To: Thomas Gleixner <tglx@...utronix.de>
Cc: LKML <linux-kernel@...r.kernel.org>,
	Frederic Weisbecker <frederic@...nel.org>,
	Anna-Maria Behnsen <anna-maria@...utronix.de>
Subject: [PATCH 4/6] timers/migration: Assert that hotplug preparing CPU is part of stable active hierarchy

The CPU doing the prepare work for a remote target must be online from
the tree point of view and its hierarchy must be active, otherwise
propagating its active state up to the new root branch would be either
incorrect or racy.

Assert those conditions with more sanity checks.

Signed-off-by: Frederic Weisbecker <frederic@...nel.org>
---
 kernel/time/timer_migration.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/kernel/time/timer_migration.c b/kernel/time/timer_migration.c
index 49635a2b7ee2..bddd816faaeb 100644
--- a/kernel/time/timer_migration.c
+++ b/kernel/time/timer_migration.c
@@ -1703,6 +1703,7 @@ static int tmigr_setup_groups(unsigned int cpu, unsigned int node,
 
 	if (activate) {
 		struct tmigr_walk data;
+		union tmigr_state state;
 
 		/*
 		 * To prevent inconsistent states, active children need to be active in
@@ -1726,6 +1727,8 @@ static int tmigr_setup_groups(unsigned int cpu, unsigned int node,
 		 *   the new childmask and parent to subsequent walkers through this
 		 *   @child. Therefore propagate active state unconditionally.
 		 */
+		state.state = atomic_read(&start->migr_state);
+		WARN_ON_ONCE(!state.active);
 		WARN_ON_ONCE(!start->parent);
 		data.childmask = start->groupmask;
 		__walk_groups_from(tmigr_active_up, &data, start, start->parent);
@@ -1768,6 +1771,11 @@ static int tmigr_add_cpu(unsigned int cpu)
 		 * active or not) and/or release an uninitialized childmask.
 		 */
 		WARN_ON_ONCE(cpu == raw_smp_processor_id());
+		/*
+		 * The (likely) current CPU is expected to be online in the hierarchy,
+		 * otherwise the old root may not be active as expected.
+		 */
+		WARN_ON_ONCE(!per_cpu_ptr(&tmigr_cpu, raw_smp_processor_id())->online);
 		ret = tmigr_setup_groups(-1, old_root->numa_node, old_root, true);
 	}
 
-- 
2.51.0