@@ -16,6 +16,15 @@ DECLARE_STATIC_KEY_FALSE(freezer_active);
extern bool pm_freezing; /* PM freezing in effect */
extern bool pm_nosig_freezing; /* PM nosig freezing in effect */
+/*
+ * Check whether the status and locks are normal
+ * when the task is frozen
+ */
+struct task_freeze_check {
+ unsigned int state;
+ int lockdep_depth;
+};
+
/*
* Timeout for stopping processes
*/
@@ -108,6 +108,7 @@ static void fake_signal_wake_up(struct task_struct *p)
static int __set_task_frozen(struct task_struct *p, void *arg)
{
unsigned int state = READ_ONCE(p->__state);
+ struct task_freeze_check *p_check = arg;
if (p->on_rq)
return 0;
@@ -118,30 +119,37 @@ static int __set_task_frozen(struct task_struct *p, void *arg)
if (!(state & (TASK_FREEZABLE | __TASK_STOPPED | __TASK_TRACED)))
return 0;
- /*
- * Only TASK_NORMAL can be augmented with TASK_FREEZABLE, since they
- * can suffer spurious wakeups.
- */
- if (state & TASK_FREEZABLE)
- WARN_ON_ONCE(!(state & TASK_NORMAL));
-
-#ifdef CONFIG_LOCKDEP
- /*
- * It's dangerous to freeze with locks held; there be dragons there.
- */
- if (!(state & __TASK_FREEZABLE_UNSAFE))
- WARN_ON_ONCE(debug_locks && p->lockdep_depth);
-#endif
-
p->saved_state = p->__state;
WRITE_ONCE(p->__state, TASK_FROZEN);
+ p_check->state = p->__state;
+ p_check->lockdep_depth = p->lockdep_depth;
return TASK_FROZEN;
}
static bool __freeze_task(struct task_struct *p)
{
+ struct task_freeze_check p_check;
+ unsigned int ret;
/* TASK_FREEZABLE|TASK_STOPPED|TASK_TRACED -> TASK_FROZEN */
- return task_call_func(p, __set_task_frozen, NULL);
+ ret = task_call_func(p, __set_task_frozen, &p_check);
+ if (ret) {
+ /*
+ * Only TASK_NORMAL can be augmented with TASK_FREEZABLE, since they
+ * can suffer spurious wakeups.
+ */
+ if (p_check.state & TASK_FREEZABLE)
+ WARN_ON_ONCE(!(p_check.state & TASK_NORMAL));
+
+#ifdef CONFIG_LOCKDEP
+ /*
+ * It's dangerous to freeze with locks held; there be dragons there.
+ */
+ if (!(p_check.state & __TASK_FREEZABLE_UNSAFE))
+ WARN_ON_ONCE(debug_locks && p_check.lockdep_depth);
+#endif
+ }
+ return ret;
+
}
/**
It is dangerous to output warnings in task_call_func, which may lead to the risk of deadlock. task_call_func uses p->pi_lock, and the serial port output may call try_to_wake_up to wake up the write buff, which also uses p->pi_lock. WARNING: possible circular locking dependency detected 6.7.0-rc4 #28 Not tainted ------------------------------------------------------ sh/475 is trying to acquire lock: ffff800082b17f20 ((console_sem).lock){-.-.}-{2:2}, at: down_trylock+0x18/0x48 but task is already holding lock: ffff0000c582dde0 (&p->pi_lock){-.-.}-{2:2}, at: task_call_func+0x40/0x124 which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&p->pi_lock){-.-.}-{2:2}: _raw_spin_lock_irqsave+0x68/0xd0 try_to_wake_up+0x5c/0x820 wake_up_process+0x18/0x24 __up.isra.0+0x40/0x4c up+0x5c/0x78 console_unlock+0x124/0x138 vt_move_to_console+0x48/0xb8 pm_restore_console+0x44/0x5c pm_suspend+0x2f0/0x688 state_store+0x8c/0x118 kobj_attr_store+0x18/0x2c sysfs_kf_write+0x4c/0x78 kernfs_fop_write_iter+0x120/0x1b4 vfs_write+0x3b4/0x558 ksys_write+0x6c/0xfc __arm64_sys_write+0x1c/0x28 invoke_syscall+0x44/0x104 el0_svc_common.constprop.0+0xc0/0xe0 do_el0_svc+0x1c/0x28 el0_svc+0x50/0xec el0t_64_sync_handler+0xc0/0xc4 el0t_64_sync+0x190/0x194 -> #0 ((console_sem).lock){-.-.}-{2:2}: __lock_acquire+0x1248/0x1ab4 lock_acquire+0x120/0x308 _raw_spin_lock_irqsave+0x68/0xd0 down_trylock+0x18/0x48 __down_trylock_console_sem+0x38/0xc4 console_trylock+0x34/0x78 vprintk_emit+0x124/0x3a4 vprintk_default+0x38/0x44 vprintk+0xb0/0xc0 _printk+0x60/0x88 report_bug+0x208/0x270 bug_handler+0x24/0x6c brk_handler+0x70/0xd4 do_debug_exception+0x9c/0x16c el1_dbg+0x74/0x90 el1h_64_sync_handler+0xc8/0xe4 el1h_64_sync+0x64/0x68 __set_task_frozen+0x64/0xac task_call_func+0xa0/0x124 freeze_task+0xb4/0x10c try_to_freeze_tasks+0xd8/0x3fc freeze_processes+0xd4/0xe4 pm_suspend+0x21c/0x688 state_store+0x8c/0x118 kobj_attr_store+0x18/0x2c sysfs_kf_write+0x4c/0x78 kernfs_fop_write_iter+0x120/0x1b4 vfs_write+0x3b4/0x558 ksys_write+0x6c/0xfc __arm64_sys_write+0x1c/0x28 invoke_syscall+0x44/0x104 el0_svc_common.constprop.0+0xc0/0xe0 do_el0_svc+0x1c/0x28 el0_svc+0x50/0xec el0t_64_sync_handler+0xc0/0xc4 el0t_64_sync+0x190/0x194 other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&p->pi_lock); lock((console_sem).lock); lock(&p->pi_lock); *** DEADLOCK *** 7 locks held by sh/475: #0: ffff0000c7245420 (sb_writers#5){.+.+}-{0:0}, at: ksys_write+0x6c/0xfc #1: ffff0000c313a890 (&of->mutex){+.+.}-{3:3}, at: kernfs_fop_write_iter+0xf0/0x1b4 #2: ffff0000c0d3e0b8 (kn->active#48){.+.+}-{0:0}, at: kernfs_fop_write_iter+0xf8/0x1b4 #3: ffff800082b11530 (system_transition_mutex){+.+.}-{3:3}, at: pm_suspend+0xb4/0x688 #4: ffff800082ae6058 (tasklist_lock){.+.+}-{2:2}, at: try_to_freeze_tasks+0x88/0x3fc #5: ffff800082b93f10 (freezer_lock){....}-{2:2}, at: freeze_task+0x3c/0x10c #6: ffff0000c582dde0 (&p->pi_lock){-.-.}-{2:2}, at: task_call_func+0x40/0x124 Fixes: f5d39b020809 ("freezer,sched: Rewrite core freezer logic") Signed-off-by: Xiaolei Wang <xiaolei.wang@windriver.com> --- include/linux/freezer.h | 9 +++++++++ kernel/freezer.c | 40 ++++++++++++++++++++++++---------------- 2 files changed, 33 insertions(+), 16 deletions(-)