lkml.org 
[lkml]   [2018]   [Apr]   [25]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Subject[PATCH v1] kthread/smpboot: Serialize kthread parking against wakeup
Date
The control cpu thread which initiates hotplug calls kthread_park()
for hotplug thread and sets KTHREAD_SHOULD_PARK. After this control
thread wakes up the hotplug thread. There is a chance that wakeup
code sees the hotplug thread (running on AP core) in INTERRUPTIBLE
state, but sets its state to RUNNING after hotplug thread has entered
kthread_parkme() and changed its state to TASK_PARKED. This can result
in panic later on in kthread_unpark(), as it sees KTHREAD_IS_PARKED
flag set but fails to rebind the kthread, due to it being not in
TASK_PARKED state. Fix this, by serializing wakeup state change,
against state change before parking the kthread.

Below is the possible race:

Control thread Hotplug Thread

kthread_park()
set KTHREAD_SHOULD_PARK
smpboot_thread_fn
set_current_state(TASK_INTERRUPTIBLE);
kthread_parkme

wake_up_process()

raw_spin_lock_irqsave(&p->pi_lock, flags);
if (!(p->state & state)) -> this will fail
goto out;

__kthread_parkme
__set_current_state(TASK_PARKED);

if (p->on_rq && ttwu_remote(p, wake_flags))
ttwu_remote()
p->state = TASK_RUNNING;
schedule();

So to avoid this race, take pi_lock to serial state changes.

Suggested-by: Pavankumar Kondeti <pkondeti@codeaurora.org>
Co-developed-by: Neeraj Upadhyay <neeraju@codeaurora.org>
Signed-off-by: Neeraj Upadhyay <neeraju@codeaurora.org>
Signed-off-by: Gaurav Kohli <gkohli@codeaurora.org>
---

Changes since V1:
- Add comment to explain need of pi_lock.

diff --git a/kernel/smpboot.c b/kernel/smpboot.c
index 5043e74..c5c5184 100644
--- a/kernel/smpboot.c
+++ b/kernel/smpboot.c
@@ -122,7 +122,45 @@ static int smpboot_thread_fn(void *data)
}

if (kthread_should_park()) {
+ /*
+ * Serialize against wakeup. If we take the lock first,
+ * wakeup is skipped. If we run later, we observe,
+ * TASK_RUNNING update from wakeup path, before moving
+ * forward. This helps avoid the race, where wakeup
+ * observes TASK_INTERRUPTIBLE, and also observes
+ * the TASK_PARKED in kthread_parkme() before updating
+ * task state to TASK_RUNNING. In this case, kthread
+ * gets parked in TASK_RUNNING state. This results
+ * in panic later on in kthread_unpark(), as it sees
+ * KTHREAD_IS_PARKED flag set but fails to rebind the
+ * kthread, due to it being not in TASK_PARKED state.
+ *
+ * Control thread Hotplug Thread
+ *
+ * kthread_park()
+ * set KTHREAD_SHOULD_PARK
+ * smpboot_thread_fn()
+ * set_current_state(
+ * TASK_INTERRUPTIBLE);
+ * kthread_parkme()
+ *
+ * wake_up_process()
+ *
+ * raw_spin_lock_irqsave(&p->pi_lock, flags);
+ * if (!(p->state & state)) __set_current_state(
+ * goto out; TASK_RUNNING);
+ *
+ * __set_current_state(
+ * TASK_PARKED);
+ *
+ * if (p->on_rq && ttwu_remote(p, wake_flags))
+ * ttwu_remote()
+ * p->state = TASK_RUNNING;
+ * schedule();
+ */
+ raw_spin_lock(&current->pi_lock);
__set_current_state(TASK_RUNNING);
+ raw_spin_unlock(&current->pi_lock);
preempt_enable();
if (ht->park && td->status == HP_THREAD_ACTIVE) {
BUG_ON(td->cpu != smp_processor_id());
--
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc. is a member of the Code Aurora Forum,
a Linux Foundation Collaborative Project.
\
 
 \ /
  Last update: 2018-04-25 10:34    [W:0.097 / U:1.372 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site