From: Petr Mladek <pmladek@suse.com>
To: John Ogness <john.ogness@linutronix.de>
Cc: Sergey Senozhatsky <senozhatsky@chromium.org>,
Steven Rostedt <rostedt@goodmis.org>,
Thomas Gleixner <tglx@linutronix.de>,
linux-kernel@vger.kernel.org,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Subject: Re: two locations: was: Re: [PATCH printk v1 03/13] printk: use percpu flag instead of cpu_online()
Date: Fri, 4 Mar 2022 17:14:14 +0100 [thread overview]
Message-ID: <YiI61ppIW+FsrzPu@alley> (raw)
In-Reply-To: <87v8wwh0jw.fsf@jogness.linutronix.de>
On Wed 2022-03-02 15:55:23, John Ogness wrote:
> On 2022-02-16, Petr Mladek <pmladek@suse.com> wrote:
> >> diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
> >> index d1b773823d63..b346e60e9e51 100644
> >> --- a/kernel/printk/printk.c
> >> +++ b/kernel/printk/printk.c
> >> @@ -2577,11 +2577,11 @@ static int have_callable_console(void)
> >> *
> >> * Console drivers may assume that per-cpu resources have been allocated. So
> >> * unless they're explicitly marked as being able to cope (CON_ANYTIME) don't
> >> - * call them until this CPU is officially up.
> >> + * call them until per-cpu resources have been allocated.
> >> */
> >> static inline int can_use_console(void)
> >> {
> >> - return cpu_online(raw_smp_processor_id()) || have_callable_console();
> >> + return (printk_percpu_data_ready() || have_callable_console());
> >> }
> >
> > cpu_online(raw_smp_processor_id()) check is used also in
> > call_console_drivers(). The same logic should be used in both
> > locations.
> >
> > I found this when reviewing 6th patch that replaced both checks
> > with a single one.
> >
> > Note that I am still not sure if this change is correct at all.
> > It will allow to always call the console during CPU hotplug
> > and I am not sure if it is safe. IMHO, it might cause problems when
> > a console driver uses, for example, CPU-bound workqueues.
>
> You are correct. We must take hotplug into account for !CON_ANYTIME
> consoles. There may be some hotplug callbacks that make memory
> unavailable for the console.
>
> However, I will add the use of printk_percpu_data_ready() in the
> check. !CON_ANYTIME consoles also should not be called until the per-cpu
> areas are ready. For example, it would be bad if a console queued
> irq_work before per-cpu areas are setup (cpu_online() is true during
> this time).
>
> One of my main concerns was that raw_smp_processor_id() was used for the
> check. It is conceptually wrong to exclude certain consoles based on a
> current CPU when migration is still enabled. I understand that the use
> of can_use_console() is an optimization to avoid doing extra work where
> there are no consoles available. But the task could be preemptible there
> and _conceptually_, could get moved to another CPU before its write()
> callback is called. The cpu_online() check belongs in code where
> preemption is disabled.
>
> If the context is preemptible, I do not think it will ever see
> !cpu_online(). So I think if the cpu_online() check is limited to
> unlocking when console_trylock() was used, it will be correct.
This would require calling console_lock()/console_unlock()
in a hotplug code when cpu_online() already returns false.
Do I get it correctly?
I agree that it should not happen. console_lock() must be called in
a preemptible context. And CPU should not be in a pre-emptible
context when cpu_online() returns false. To be honest, I did
not check the code. It just does not make much sense.
> In the current implementation of printk(), it would be odd to do this
> conditional check (perhaps by passing @do_cond_resched to
> can_use_console()). But my series does significant refactoring and
> actually does need to distinguish between console_lock() and
> console_trylock() due to the kthreads and supporting the handover. So it
> should work well that the cpu_online() check for !CON_ANYTIME is only
> performed when !preemptible.
>
> Regardless, my v2 will keep cpu_online() checks since they are necessary
> for hotplug support.
Yes, I would do it to stay on the safe side.
Best Regards,
Petr
next prev parent reply other threads:[~2022-03-04 16:14 UTC|newest]
Thread overview: 87+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-07 19:43 [PATCH printk v1 00/13] implement threaded console printing John Ogness
2022-02-07 19:43 ` [PATCH printk v1 01/13] printk: rename cpulock functions John Ogness
2022-02-11 12:44 ` Petr Mladek
2022-02-11 14:42 ` John Ogness
2022-02-11 20:57 ` Steven Rostedt
2022-02-11 21:04 ` Peter Zijlstra
2022-02-15 9:32 ` Petr Mladek
2022-02-15 9:13 ` Petr Mladek
2022-02-14 6:49 ` Sergey Senozhatsky
2022-02-14 9:45 ` John Ogness
2022-02-15 9:29 ` Petr Mladek
2022-02-16 3:27 ` Sergey Senozhatsky
2022-02-17 14:34 ` John Ogness
2022-02-07 19:43 ` [PATCH printk v1 02/13] printk: cpu sync always disable interrupts John Ogness
2022-02-11 12:58 ` Petr Mladek
2022-02-14 6:36 ` Sergey Senozhatsky
2022-02-07 19:43 ` [PATCH printk v1 03/13] printk: use percpu flag instead of cpu_online() John Ogness
2022-02-11 16:05 ` Petr Mladek
2022-02-14 7:08 ` Sergey Senozhatsky
2022-02-14 7:35 ` Sergey Senozhatsky
2022-02-15 10:38 ` Petr Mladek
2022-02-16 3:29 ` Sergey Senozhatsky
2022-03-02 14:21 ` John Ogness
2022-03-04 15:56 ` Petr Mladek
2022-03-05 17:05 ` Jason A. Donenfeld
2022-03-07 16:14 ` Petr Mladek
2022-02-16 13:58 ` two locations: was: " Petr Mladek
2022-03-02 14:49 ` John Ogness
2022-03-04 16:14 ` Petr Mladek [this message]
2022-03-07 10:06 ` John Ogness
2022-03-08 16:08 ` Petr Mladek
2022-02-07 19:43 ` [PATCH printk v1 04/13] printk: get caller_id/timestamp after migration disable John Ogness
2022-02-15 5:53 ` Sergey Senozhatsky
2022-02-15 11:56 ` Petr Mladek
2022-02-07 19:43 ` [PATCH printk v1 05/13] printk: call boot_delay_msec() in printk_delay() John Ogness
2022-02-15 5:58 ` Sergey Senozhatsky
2022-02-15 14:59 ` Petr Mladek
2022-02-16 3:21 ` Sergey Senozhatsky
2022-02-15 15:03 ` Petr Mladek
2022-02-07 19:43 ` [PATCH printk v1 06/13] printk: refactor and rework printing logic John Ogness
2022-02-16 15:43 ` Petr Mladek
2022-03-02 16:10 ` John Ogness
2022-02-07 19:43 ` [PATCH printk v1 07/13] printk: move buffer definitions into console_emit_next_record() caller John Ogness
2022-02-16 16:10 ` Petr Mladek
2022-03-02 16:25 ` John Ogness
2022-02-07 19:43 ` [PATCH printk v1 08/13] printk: add pr_flush() John Ogness
2022-02-17 10:11 ` Petr Mladek
2022-03-02 17:23 ` John Ogness
2022-03-04 13:24 ` Petr Mladek
2022-02-07 19:43 ` [PATCH printk v1 09/13] printk: add functions to allow direct printing John Ogness
2022-02-17 12:52 ` Petr Mladek
2022-02-18 9:00 ` David Laight
2022-02-18 12:52 ` Petr Mladek
2022-03-03 14:37 ` John Ogness
2022-02-07 19:43 ` [PATCH printk v1 10/13] printk: add kthread console printers John Ogness
2022-02-18 9:00 ` early start: was: " Petr Mladek
2022-02-18 9:04 ` start&stop: " Petr Mladek
2022-02-18 9:08 ` main loop: " Petr Mladek
2022-02-18 9:12 ` wake_up_all: " Petr Mladek
2022-02-07 19:43 ` [PATCH printk v1 11/13] printk: reimplement console_lock for proper kthread support John Ogness
2022-02-18 16:20 ` Petr Mladek
2022-02-18 21:41 ` John Ogness
2022-02-18 22:03 ` John Ogness
2022-02-22 11:42 ` Petr Mladek
2022-02-23 17:20 ` John Ogness
2022-02-24 8:27 ` Petr Mladek
2022-02-23 10:19 ` Petr Mladek
2022-03-09 13:56 ` John Ogness
2022-03-10 14:34 ` Petr Mladek
2022-03-10 16:08 ` John Ogness
2022-03-11 10:26 ` Petr Mladek
2022-03-11 13:28 ` John Ogness
2022-03-11 16:17 ` Petr Mladek
2022-03-11 22:21 ` John Ogness
2022-03-14 14:08 ` Petr Mladek
2022-03-14 14:43 ` John Ogness
2022-03-14 15:53 ` Petr Mladek
2022-03-11 18:41 ` Petr Mladek
2022-02-07 19:43 ` [PATCH printk v1 12/13] printk: remove @console_locked John Ogness
2022-02-23 12:17 ` Petr Mladek
2022-02-07 19:43 ` [PATCH printk v1 13/13] console: introduce CON_MIGHT_SLEEP for vt John Ogness
2022-02-23 13:37 ` Petr Mladek
2022-02-23 18:31 ` Greg Kroah-Hartman
[not found] ` <20220208083620.2736-1-hdanton@sina.com>
2022-02-08 11:08 ` [PATCH printk v1 10/13] printk: add kthread console printers John Ogness
2022-02-08 14:53 ` Petr Mladek
2022-02-14 6:12 ` Sergey Senozhatsky
2022-02-14 10:02 ` Petr Mladek
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=YiI61ppIW+FsrzPu@alley \
--to=pmladek@suse.com \
--cc=gregkh@linuxfoundation.org \
--cc=john.ogness@linutronix.de \
--cc=linux-kernel@vger.kernel.org \
--cc=rostedt@goodmis.org \
--cc=senozhatsky@chromium.org \
--cc=tglx@linutronix.de \
/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;
as well as URLs for NNTP newsgroup(s).