Re: [PATCH] drivers/base: error handling fixes

[Jeff Garzik <jeff@garzik.org>]

Heiko Carstens wrote:
>>>>> static int __cpuinit topology_cpu_callback(struct notifier_block *nfb,
>>>>> @@ -112,17 +110,18 @@ static int __cpuinit topology_cpu_callba
>>>>> {
>>>>> 	unsigned int cpu = (unsigned long)hcpu;
>>>>> 	struct sys_device *sys_dev;
>>>>> +	int rc = 0;
>>>>>  	sys_dev = get_cpu_sysdev(cpu);
>>>>> 	switch (action) {
>>>>> 	case CPU_ONLINE:
>>>>> -		topology_add_dev(sys_dev);
>>>>> +		rc = topology_add_dev(sys_dev);
>>>>> 		break;
>>>>> 	case CPU_DEAD:
>>>>> 		topology_remove_dev(sys_dev);
>>>>> 		break;
>>>>> 	}
>>>>> -	return NOTIFY_OK;
>>>>> +	return rc ? NOTIFY_BAD : NOTIFY_OK;
>>>>> }
>>>> Wouldn't that also require that _cpu_up checked the return code when
>>>> doing CPU_ONLINE notification (and clean up on error)?
>>> After all code that gets a CPU_ONLINE notification is not supposed to fail.
>>> For allocating resources while bringing up a cpu CPU_UP_PREPARE is supposed
>>> to be used. That one is allowed to fail.
>> It's a bug no matter how you look at it... I just lessen the impact.  :)
>>
>> If someone wants to provide a better fix, let's see the patch...
> 
> If sysfs_remove_group() would also work for non-created (-existent) groups
> then the patch below would work. Unfortunately that is not the case. So one
> would have to remember if sysfs_create_group() was done and succeeded before
> calling sysfs_remove_group()...
> There must be an easier way.
> 
> diff --git a/drivers/base/topology.c b/drivers/base/topology.c
> index 3ef9d51..d0056c3 100644
> --- a/drivers/base/topology.c
> +++ b/drivers/base/topology.c

ACK