Re: [PATCH 7/6] ipc/sem.c: add a printk_once for semctl(GETNCNT/GETZCNT)

[Manfred Spraul <manfred@colorfullife.com>]

Hi Joe,

On 05/25/2014 08:39 PM, Joe Perches wrote:
> On Sun, 2014-05-25 at 20:21 +0200, Manfred Spraul wrote:
> +	 */
> +	printk_once(KERN_INFO "semctl(GETNCNT/GETZCNT) is since 3.16 Single " \
> +				"Unix Specification compliant.\n" \
> +				"The task %d triggered the difference, " \
> +				"watch for misbehavior.", current->pid);
> Unnecessary line continuations.
> Missing terminating newline after "misbehavior"
> Ideally coalesced or broken at linebreaks like:
>
> 	pr_info_once("semctl(GETNCNT/GETZCNT) is Single Unix Specification compliant since kernel v3.16\n"
> 		     "Task %d triggered the difference, watch for misbehavior\n",
> 		     current->pid);
Thanks. I'll try to remember to really run checkpatch instead of 
assuming what it might report.
>>   	if (sop->sem_num != semnum)
>>   		return 0;
>>   
> Should the printk_once (which could be pr_info_once or _ratelimited
> or maybe even emitted at KERN_DEBUG) be done only when
> the return is 1?
>
To fully check if there is a difference would mean that the old code and 
the new code run in parallel.

The code might trigger slightly too often, but since there are zero 
known users of GETZCNT / GETNCNT simplicity wins.

--
     Manfred