Re: [PATCH v5 2/2] printk: console: Support console-specific loglevels

[Chris Down]

Thanks for the detailed feedback! All comments not explicitly replied to are 
acked and will be updated in v6.

Petr Mladek writes:
> > @@ -2293,7 +2429,14 @@ asmlinkage int vprintk_emit(int facility, int level,
> >  		in_sched = true;
> >  	}
> >
> > -	printk_delay(level);
> > +	cookie = console_srcu_read_lock();
> > +	for_each_console_srcu(con) {
> > +		if (!suppress_message_printing(level, con)) {
> > +			printk_delay();
>
> This would repeat the delay for each registered and non-supressed console.
> But it should be called only once when there is at least one
> non-suppressed console.
>
> > +			break;
> > +		}
> > +	}
> > +	console_srcu_read_unlock(cookie);

Doesn't the `break' below avoid that already? Maybe I'm missing something?

[...]

> > +static int printk_sysctl_deprecated(struct ctl_table *table, int write,
> > +				    void *buffer, size_t *lenp, loff_t *ppos)
> > +{
> > +	int res = proc_dointvec(table, write, buffer, lenp, ppos);
> > +
> > +	if (write)
> > +		pr_warn_once(
> > +			"printk: The kernel.printk sysctl is deprecated. Consider using kernel.console_loglevel or kernel.default_message_loglevel instead.\n");
>
> Nit: Please, put it on the same line as pr_warn_once("printk: The kernel....
>     The small shift left did not help much ;-)

I think this is done by clang-format with kernel default config, since my 
preference is single line as well. Maybe worth discussing changing the config?

> How does this sounds, please?

Sounds great! Thank you again for the very detailed feedback :-)