On Mon, 2007-08-27 at 11:30 +0400, Pavel Emelyanov wrote:
> From: Alexey Dobriyan <adobriyan@openvz.org>
> 
> One of the easiest things to isolate is the pid printed in kernel
> log. There was a patch, that made this for arch-independent code,
> this one makes so for arch/xxx files.

Perhaps it would be worthwhile to standardize the printk
task information content? Maybe a format knob?

Maybe TASK_FMT/TASK_ARG(task)

#define TASK_FMT "%s"
#define TASK_ARG(tsk) ({char __buf[MAX_TASK_INFO]; print_task(tsk, buf, sizeof(__buf)); __buf})

char* print_task(const struct task_struct* tsk, char* buf, size_t size)
{
	char* p = buf;
	int task_knob = get_task_knob();
	*p++ = '[';
	if (task_knob & feature)
		p += snprintf(p, size - (p - buf), fmt, feature);
	[...]
	*p++ = ']';
	return buf;
}
EXPORT_SYMBOL(print_task)


_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers

On Aug 27 2007 11:01, Joe Perches wrote:
>On Mon, 2007-08-27 at 11:30 +0400, Pavel Emelyanov wrote:
>> From: Alexey Dobriyan <adobriyan@openvz.org>
>> 
>> One of the easiest things to isolate is the pid printed in kernel
>> log. There was a patch, that made this for arch-independent code,
>> this one makes so for arch/xxx files.
>
>Perhaps it would be worthwhile to standardize the printk
>task information content? Maybe a format knob?
>
>Maybe TASK_FMT/TASK_ARG(task)
>
>#define TASK_FMT "%s"
>#define TASK_ARG(tsk) ({char __buf[MAX_TASK_INFO]; print_task(tsk, buf, sizeof(__buf)); __buf})

Except that __buf goes out of scope before printk() is called...



	Jan
-- 
_______________________________________________
Containers mailing list
Containers@lists.linux-foundation.org
https://lists.linux-foundation.org/mailman/listinfo/containers