ksummit.lists.linux.dev archive mirror
 help / color / mirror / Atom feed
From: Steven Rostedt <rostedt@goodmis.org>
To: Josh Triplett <josh@joshtriplett.org>
Cc: "ksummit-discuss@lists.linuxfoundation.org"
	<ksummit-discuss@lists.linuxfoundation.org>
Subject: Re: [Ksummit-discuss] [TECH TOPIC] asynchronous printk
Date: Wed, 27 Jul 2016 23:05:32 -0400	[thread overview]
Message-ID: <20160727230532.2982b40d@grimm.local.home> (raw)
In-Reply-To: <20160721142127.GA23563@x>

On Thu, 21 Jul 2016 07:21:28 -0700
Josh Triplett <josh@joshtriplett.org> wrote:
> 
> Given the discussion in this thread about wanting to eliminate
> earlyprintk in favor of a unified mechanism that works as early as
> possible, how would this case work on the system I described above where
> earlyprintk works and console doesn't?

No no no!!! Please do not eliminate earlyprintk(). It is critical to
the PREEMPT_RT project. It requires no sleeping locks, and can be used
in pretty much any context (with slight modifications). printk() is a
beast, and in -rt it calls sleeping locks. That is, if interrupts are
disabled, we prevent printk from print anything. (Note PREEMPT_RT
keeps interrupts enabled much more than the normal kernel does). If a
bug happens where interrupts are disabled, we fall back to earlyprintk
(and tracing), to debug it.

-- Steve

  parent reply	other threads:[~2016-07-28  3:23 UTC|newest]

Thread overview: 60+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-07-19  3:47 Sergey Senozhatsky
2016-07-19  3:56 ` Viresh Kumar
2016-07-19  6:17 ` Hannes Reinecke
2016-07-19  6:49   ` Josh Triplett
2016-07-19  7:02     ` Hannes Reinecke
2016-07-19  7:11       ` Geert Uytterhoeven
2016-07-20  6:02         ` Jan Kara
2016-07-20 22:54       ` Josh Triplett
2016-07-21  0:46         ` Sergey Senozhatsky
2016-07-21  1:12           ` Josh Triplett
2016-07-19  7:33   ` Sergey Senozhatsky
2016-07-19  7:38     ` Hannes Reinecke
2016-07-19  7:46       ` Sergey Senozhatsky
2016-07-19  8:02         ` Hannes Reinecke
2016-07-19  8:23           ` Sergey Senozhatsky
2016-07-21 10:36           ` David Woodhouse
2016-07-21 12:31             ` Jan Kara
2016-07-28  2:55             ` Steven Rostedt
2016-07-20  6:09       ` Jan Kara
2016-07-19  7:46   ` Christian Borntraeger
2016-07-19  7:53     ` Christian Borntraeger
2016-07-19 13:55       ` Jan Kara
2016-07-28  2:59         ` Steven Rostedt
2016-07-28  4:12           ` Sergey Senozhatsky
2016-07-28 13:02             ` Steven Rostedt
2016-07-20  3:35   ` Wangnan (F)
2016-07-21  1:16     ` Andy Lutomirski
2016-07-21  1:52       ` Wangnan (F)
2016-07-21  5:59       ` Hannes Reinecke
2016-07-21 10:31         ` David Woodhouse
2016-07-21 11:19           ` Josh Triplett
2016-07-21 11:59             ` David Woodhouse
2016-07-21 14:21               ` Josh Triplett
2016-07-21 14:40                 ` David Woodhouse
2016-07-28  3:05                 ` Steven Rostedt [this message]
2016-08-02 11:59               ` Petr Mladek
2016-07-21 15:05           ` Andy Lutomirski
2016-07-26 14:40             ` David Woodhouse
2016-07-26 15:44               ` Benjamin Herrenschmidt
2016-07-26 21:00               ` Andy Lutomirski
2016-07-27  0:03                 ` David Woodhouse
2016-07-27  1:16                   ` Sergey Senozhatsky
2016-07-21 10:28       ` David Woodhouse
2016-07-19 14:45 ` James Bottomley
2016-07-19 14:55   ` Sergey Senozhatsky
2016-07-19 17:58     ` James Bottomley
2016-07-19 18:24       ` Viresh Kumar
2016-07-20  2:08       ` Sergey Senozhatsky
2016-07-20  6:14     ` Jan Kara
2016-09-21  4:41 ` Sergey Senozhatsky
2016-10-31  6:54   ` Sergey Senozhatsky
2016-10-31 13:56     ` Theodore Ts'o
2016-10-31 13:59       ` Jiri Kosina
2016-10-31 14:56       ` [Ksummit-discuss] [TECH TOPIC] printk considered harmful (was: [TECH TOPIC] asynchronous printk) Sergey Senozhatsky
2016-10-31 16:18         ` Theodore Ts'o
2016-10-31 18:21           ` Sergey Senozhatsky
2016-10-31 18:26             ` [Ksummit-discuss] [TECH TOPIC] printk considered harmful Hannes Reinecke
2016-10-31 20:28           ` [Ksummit-discuss] [TECH TOPIC] printk considered harmful (was: [TECH TOPIC] asynchronous printk) Jan Kara
2016-11-01 12:27             ` [Ksummit-discuss] [TECH TOPIC] printk considered harmful Hannes Reinecke
2016-11-01 17:50         ` [Ksummit-discuss] [TECH TOPIC] printk considered harmful (was: [TECH TOPIC] asynchronous printk) Sergey Senozhatsky

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20160727230532.2982b40d@grimm.local.home \
    --to=rostedt@goodmis.org \
    --cc=josh@joshtriplett.org \
    --cc=ksummit-discuss@lists.linuxfoundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox