Re: Btrfs: fix num_workers_starting bug and other bugs in async thread

[Chris Mason <chris.mason@oracle.com>]

On Fri, Dec 23, 2011 at 01:44:53PM +0300, Dan Carpenter wrote:
> Hi Josef,
> 
> Smatch complains about this change introduces a double unlock.
> 
> fs/btrfs/async-thread.c +608 find_worker(49) error: double unlock 'spin_lock:&workers->lock'
> 
>    579                          spin_unlock_irqrestore(&workers->lock, flags);
>                                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Thanks Dan, fixing.

> We unlock here.
> 
>    580                          /* we're below the limit, start another worker */
>    581                          ret = __btrfs_start_workers(workers);
>    582                          if (ret)
>    583                                  goto fallback;
>    584                          goto again;
>    585                  }
>    586          }
>    587          goto found;
>    588  
>    589  fallback:
>    590          fallback = NULL;
>    591          /*
>    592           * we have failed to find any workers, just
>    593           * return the first one we can find.
>    594           */
>    595          if (!list_empty(&workers->worker_list))
>    596                  fallback = workers->worker_list.next;
>    597          if (!list_empty(&workers->idle_list))
>    598                  fallback = workers->idle_list.next;
>    599          BUG_ON(!fallback);
>    600          worker = list_entry(fallback,
>    601                    struct btrfs_worker_thread, worker_list);
>    602  found:
>    603          /*
>    604           * this makes sure the worker doesn't exit before it is placed
>    605           * onto a busy/idle list
>    606           */
>    607          atomic_inc(&worker->num_pending);
>    608          spin_unlock_irqrestore(&workers->lock, flags);
>                 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> And again here.
> 
> Btw, does find_worker() ever get called with IRQs disabled?  If so then
> __btrfs_start_workers() enables them.  Maybe that function should use
> spin_lock_irqsave() instead of spin_lock_irq().

Patching this too.

-chris