[BUG] stmmac: A possible sleep-in-atomic bug in stmmac_suspend

[Jia-Ju Bai]

According to stmmac_main.c, the driver may sleep under a spinlock,
and the function call path is:
stmmac_suspend (acquire the spinlock)
  stmmac_disable_all_queues
    napi_disable
      might_sleep --> may sleep
      msleep --> may sleep

This bug is found by my static analysis tool and my code review.


Thanks,
Jia-Ju Bai