rcu: Fix rcu_read_unlock() deadloop due to IRQ work
Linux
Affected versions
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
Linux
| Product | Affected | Fixed in |
|---|---|---|
| Linux Linux |
|
≥ 0864f057b050bc6dd68106b3185e02db5140012d · < b41642c87716bbd09797b1e4ea7d904f06c39b7b| No fix yet |
| Linux Linux |
|
|
In the Linux kernel, the following vulnerability has been resolved:
rcu: Fix rcu_read_unlock() deadloop due to IRQ work
During rcu_read_unlock_special(), if this happens during irq_exit(), we can lockup if an IPI is issued. This is because the IPI itself triggers the irq_exit() path causing a recursive lock up.
This is precisely what Xiongfeng found when invoking a BPF program on the trace_tick_stop() tracepoint As shown in the trace below. Fix by managing the irq_work state correctly.
irq_exit() __irq_exit_rcu() /* in_hardirq() returns false after this */ preempt_count_sub(HARDIRQ_OFFSET) tick_irq_exit() tick_nohz_irq_exit() tick_nohz_stop_sched_tick() trace_tick_stop() /* a bpf prog is hooked on this trace point */ __bpf_trace_tick_stop() bpf_trace_run2() rcu_read_unlock_special() /* will send a IPI to itself */ irq_work_queue_on(&rdp->defer_qs_iw, rdp->cpu);
A simple reproducer can also be obtained by doing the following in tick_irq_exit(). It will hang on boot without the patch:
static inline void tick_irq_exit(void) { + rcu_read_lock(); + WRITE_ONCE(current->rcu_read_unlock_special.b.need_qs, true); + rcu_read_unlock(); +
[neeraj: Apply Frederic's suggested fix for PREEMPT_RT]
From the CVE record · Linux
The Linux kernel can lock up in rcu_read_unlock_special() if it issues an IPI during irq_exit(), causing a recursive IRQ-exit path. This was observed when a BPF program was hooked to the trace_tick_stop() tracepoint. The input does not specify attacker access requirements or whether user interaction is needed.
Anyone running the Linux kernel should check whether their version includes the RCU fix; affected versions are not specified.