Re: [RFC PATCH] printk: console: Allow each console to have its own loglevel

From: Petr Mladek <pmladek@suse.com>
To: Chris Down <chris@chrisdown.name>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	linux-kernel@vger.kernel.org, kernel-team@fb.com
Subject: Re: [RFC PATCH] printk: console: Allow each console to have its own loglevel
Date: Tue, 24 May 2022 11:19:28 +0200	[thread overview]
Message-ID: <YoyjIFI0LTvUHGQr@alley> (raw)
In-Reply-To: <YoVM+KbdyJm8RSSr@chrisdown.name>

On Wed 2022-05-18 20:46:00, Chris Down wrote:
> Greg Kroah-Hartman writes:
> > > +static int printk_sysctl_deprecated(struct ctl_table *table, int write,
> > > +				    void __user *buffer, size_t *lenp,
> > > +				    loff_t *ppos)
> > > +{
> > > +	int res = proc_dointvec(table, write, buffer, lenp, ppos);
> > > +
> > > +	if (write)
> > > +		pr_warn_ratelimited(
> > > +			"printk: The kernel.printk sysctl is deprecated and will be removed soon. Use kernel.force_console_loglevel, kernel.default_message_loglevel, kernel.minimum_console_loglevel, or kernel.default_console_loglevel instead.\n"
> > 
> > Please define "soon".
> 
> Petr, what do you think about the timebounds here? :-)

Good question. Honestly, I do not have experience with deprecating
such interfaces. Any advice would be helpful.

We might investigate the kernel log whether and how some sysctl
interface was deprecated.

I would personally keep the interface between 10-20 kernel
releases.

Best Regards,
Petr

PS: I still have to look deeper at the patchset. It might take
    some more days. It seems that nature is not the only one who
    woke up after the winter.