Re: [RFC][PATCH 4/5 v2] printk: Add per_cpu printk func to allow printk to be diverted

["Petr Mládek" <pmladek@suse.cz>]

On Fri 2014-06-27 10:39:33, Steven Rostedt wrote:
> On Fri, 27 Jun 2014 16:20:25 +0200
> Petr Mládek <pmladek@suse.cz> wrote:
> 
> 
> > >  	va_start(args, fmt);
> > > -	r = vprintk_emit(0, -1, NULL, 0, fmt, args);
> > > +	preempt_disable();
> > 
> > I think that it is too late to disable the preemption here.
> > It has to be done by the printk() caller if it wants to be sure
> > that the requested function is used.
> 
> That's only if the printk() caller cares. But it would be nice that we
> run the printk_func for the CPU that vprintk_func() is on, thus the
> preempt_disable() is required. (in -rt, this would turn into a
> migrate_disable()).

It makes sense. Thanks for explanation.

Best Regards,
Petr
 
> -- Steve
> 
> > 
> > > +	vprintk_func = this_cpu_read(printk_func);
> > > +	r = vprintk_func(fmt, args);
> > > +	preempt_enable();
> > >  	va_end(args);
> > >  
> > >  	return r;
> > 
> > Best Regards,
> > Petr
>