diff options
author | Sebastian Andrzej Siewior <bigeasy@linutronix.de> | 2020-03-31 23:18:49 +0300 |
---|---|---|
committer | Thomas Gleixner <tglx@linutronix.de> | 2020-04-01 14:20:14 +0300 |
commit | 73d20564e0dcae003e0d79977f044d5e57496304 (patch) | |
tree | 10e1f109985f6da880de947d076646bd33fac2e9 /kernel | |
parent | 42595ce90b9d4a6b9d8c5a1ea78da4eeaf7e086a (diff) | |
download | linux-73d20564e0dcae003e0d79977f044d5e57496304.tar.xz |
hrtimer: Don't dereference the hrtimer pointer after the callback
A hrtimer can be released in its callback, but lockdep_hrtimer_exit()
dereferences the pointer after the callback returns, i.e. a potential use
after free.
Retrieve the context in which the hrtimer expires before the callback is
invoked and use it in lockdep_hrtimer_exit().
Fixes: 40db173965c0 ("lockdep: Add hrtimer context tracing bits")
Reported-by: syzbot+62c155c276e580cfb606@syzkaller.appspotmail.com
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20200331201849.fkp2siy3vcdqvqlz@linutronix.de
Diffstat (limited to 'kernel')
-rw-r--r-- | kernel/time/hrtimer.c | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c index d0a5ba37aff4..d89da1c7e005 100644 --- a/kernel/time/hrtimer.c +++ b/kernel/time/hrtimer.c @@ -1480,6 +1480,7 @@ static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base, unsigned long flags) __must_hold(&cpu_base->lock) { enum hrtimer_restart (*fn)(struct hrtimer *); + bool expires_in_hardirq; int restart; lockdep_assert_held(&cpu_base->lock); @@ -1514,11 +1515,11 @@ static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base, */ raw_spin_unlock_irqrestore(&cpu_base->lock, flags); trace_hrtimer_expire_entry(timer, now); - lockdep_hrtimer_enter(timer); + expires_in_hardirq = lockdep_hrtimer_enter(timer); restart = fn(timer); - lockdep_hrtimer_exit(timer); + lockdep_hrtimer_exit(expires_in_hardirq); trace_hrtimer_expire_exit(timer); raw_spin_lock_irq(&cpu_base->lock); |