On Sun, 29 Nov 2009 18:01:19 +0200
Dan Carpenter <error27@xxxxxxxxx> wrote:
> There was a potential locking issue found by smatch static checker.
>
> drivers/pci/hotplug/ibmphp_hpc.c
> 889 up (&semOperations);
> 890 msleep(POLL_INTERVAL_SEC * 1000);
> 891
> 892 if (kthread_should_stop())
> 893 break;
>
> We break here with semOperations released.
>
> 894
> 895 down (&semOperations);
> 896
> 897 if (poll_count >= POLL_LATCH_CNT) {
> 898 poll_count = 0;
> 899 poll_state = POLL_SLOTS;
> 900 } else
> 901 poll_state =
> POLL_LATCH_REGISTER; 902 break;
>
> We break here with semOperations held.
>
> 903 }
> 904 /* give up the hardware semaphore */
>
> We release semOperations here a second time which may be wrong since
> it was already released.
>
> 905 up (&semOperations);
Thanks, I committed what looks like an obvious fix (jump over the up in
the kthread_should_stop() case), but I have no test hardware, so I'll
trust someone will scream if it breaks something.
--
Jesse Barnes, Intel Open Source Technology Center
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
