Re: [RFC PATCH v4 0/9] printk: new ringbuffer implementation

From: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
To: Peter Zijlstra <peterz@infradead.org>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>,
	Petr Mladek <pmladek@suse.com>,
	John Ogness <john.ogness@linutronix.de>,
	Andrea Parri <andrea.parri@amarulasolutions.com>,
	Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
	Steven Rostedt <rostedt@goodmis.org>,
	Brendan Higgins <brendanhiggins@google.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	linux-kernel@vger.kernel.org
Subject: Re: [RFC PATCH v4 0/9] printk: new ringbuffer implementation
Date: Fri, 6 Sep 2019 22:44:39 +0900	[thread overview]
Message-ID: <20190906134439.GA9590@tigerII.localdomain> (raw)
In-Reply-To: <20190906104948.GX2349@hirez.programming.kicks-ass.net>

On (09/06/19 12:49), Peter Zijlstra wrote:
> On Fri, Sep 06, 2019 at 07:09:43PM +0900, Sergey Senozhatsky wrote:
> 
> > ---
> > diff --git a/kernel/printk/printk_safe.c b/kernel/printk/printk_safe.c
> > index 139c310049b1..9c73eb6259ce 100644
> > --- a/kernel/printk/printk_safe.c
> > +++ b/kernel/printk/printk_safe.c
> > @@ -103,7 +103,10 @@ static __printf(2, 0) int printk_safe_log_store(struct printk_safe_seq_buf *s,
> >         if (atomic_cmpxchg(&s->len, len, len + add) != len)
> >                 goto again;
> >  
> > -       queue_flush_work(s);
> > +       if (early_console)
> > +               early_console->write(early_console, s->buffer + len, add);
> > +       else
> > +               queue_flush_work(s);
> >         return add;
> >  }
> 
> You've not been following along, that generates absolutely unreadable
> garbage.

This was more of a joke/reference to "Those NMI buffers are a trainwreck
and need to die a horrible death". Of course this needs a re-entrant cpu
lock to serialize access to atomic/early consoles. But here is one more
missing thing - we need atomic/early consoles on a separate, sort of
immutable, list. And probably forbid any modifications of such console
drivers, (PM, etc.) If we can do this then we don't need to take console_sem
while we iterate that list, which removes sched/timekeeping locks out
of the fast printk() path.

We, at the same time, don't have that many options on systems without
atomic/early consoles. Move printing to NMI (e.g. up to X pending logbug
lines per NMI)? Move printing to IPI (again, up to X pending logbuf lines
per IPI)? printk() softirqs?

	-ss