All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jan Kara <jack@suse.cz>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
	akpm@linux-foundation.org, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org,
	Cong Wang <xiyou.wangcong@gmail.com>,
	Dave Hansen <dave.hansen@intel.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Mel Gorman <mgorman@suse.de>, Michal Hocko <mhocko@kernel.org>,
	Petr Mladek <pmladek@suse.com>,
	Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	"yuwang.yuwang" <yuwang.yuwang@alibaba-inc.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Jan Kara <jack@suse.cz>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Subject: Re: [PATCH v2] printk: Add console owner and waiter logic to load balance console writes
Date: Fri, 3 Nov 2017 11:19:53 +0100	[thread overview]
Message-ID: <20171103101953.GA5280@quack2.suse.cz> (raw)
In-Reply-To: <20171102130605.05e987e8@gandalf.local.home>

Hi,

On Thu 02-11-17 13:06:05, Steven Rostedt wrote:
> +			if (spin) {
> +				/* We spin waiting for the owner to release us */
> +				spin_acquire(&console_owner_dep_map, 0, 0, _THIS_IP_);
> +				/* Owner will clear console_waiter on hand off */
> +				while (!READ_ONCE(console_waiter))
> +					cpu_relax();

Hum, what prevents us from rescheduling here? And what if the process
stored in console_owner is scheduled out? Both seem to be possible with
CONFIG_PREEMPT kernel? Unless I'm missing something you will need to
disable preemption in some places...

Other than that I like the simplicity of your approach.

								Honza

> +
> +				spin_release(&console_owner_dep_map, 1, _THIS_IP_);
> +				printk_safe_exit_irqrestore(flags);
> +
> +				/*
> +				 * The owner passed the console lock to us.
> +				 * Since we did not spin on console lock, annotate
> +				 * this as a trylock. Otherwise lockdep will
> +				 * complain.
> +				 */
> +				mutex_acquire(&console_lock_dep_map, 0, 1, _THIS_IP_);
> +				console_unlock();
> +				printk_safe_enter_irqsave(flags);
> +			}
> +			printk_safe_exit_irqrestore(flags);
> +
> +		}
>  	}
>  
>  	return printed_len;
> @@ -2141,6 +2196,7 @@ void console_unlock(void)
>  	static u64 seen_seq;
>  	unsigned long flags;
>  	bool wake_klogd = false;
> +	bool waiter = false;
>  	bool do_cond_resched, retry;
>  
>  	if (console_suspended) {
> @@ -2215,6 +2271,20 @@ skip:
>  			goto skip;
>  		}
>  
> +		/*
> +		 * While actively printing out messages, if another printk()
> +		 * were to occur on another CPU, it may wait for this one to
> +		 * finish. This task can not be preempted if there is a
> +		 * waiter waiting to take over.
> +		 */
> +
> +		/* The waiter may spin on us after this */
> +		spin_acquire(&console_owner_dep_map, 0, 0, _THIS_IP_);
> +
> +		raw_spin_lock(&console_owner_lock);
> +		console_owner = current;
> +		raw_spin_unlock(&console_owner_lock);
> +
>  		len += msg_print_text(msg, false, text + len, sizeof(text) - len);
>  		if (nr_ext_console_drivers) {
>  			ext_len = msg_print_ext_header(ext_text,
> @@ -2232,11 +2302,48 @@ skip:
>  		stop_critical_timings();	/* don't trace print latency */
>  		call_console_drivers(ext_text, ext_len, text, len);
>  		start_critical_timings();
> +
> +		raw_spin_lock(&console_owner_lock);
> +		waiter = console_waiter;
> +		console_owner = NULL;
> +		raw_spin_unlock(&console_owner_lock);
> +
> +		/*
> +		 * If there is a waiter waiting for us, then pass the
> +		 * rest of the work load over to that waiter.
> +		 */
> +		if (waiter)
> +			break;
> +
> +		/* There was no waiter, and nothing will spin on us here */
> +		spin_release(&console_owner_dep_map, 1, _THIS_IP_);
> +
>  		printk_safe_exit_irqrestore(flags);
>  
>  		if (do_cond_resched)
>  			cond_resched();
>  	}
> +
> +	/*
> +	 * If there is an active waiter waiting on the console_lock.
> +	 * Pass off the printing to the waiter, and the waiter
> +	 * will continue printing on its CPU, and when all writing
> +	 * has finished, the last printer will wake up klogd.
> +	 */
> +	if (waiter) {
> +		WRITE_ONCE(console_waiter, false);
> +		/* The waiter is now free to continue */
> +		spin_release(&console_owner_dep_map, 1, _THIS_IP_);
> +		/*
> +		 * Hand off console_lock to waiter. The waiter will perform
> +		 * the up(). After this, the waiter is the console_lock owner.
> +		 */
> +		mutex_release(&console_lock_dep_map, 1, _THIS_IP_);
> +		printk_safe_exit_irqrestore(flags);
> +		/* Note, if waiter is set, logbuf_lock is not held */
> +		return;
> +	}
> +
>  	console_locked = 0;
>  
>  	/* Release the exclusive_console once it is used */
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

WARNING: multiple messages have this Message-ID (diff)
From: Jan Kara <jack@suse.cz>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
	akpm@linux-foundation.org, linux-mm@kvack.org,
	linux-kernel@vger.kernel.org,
	Cong Wang <xiyou.wangcong@gmail.com>,
	Dave Hansen <dave.hansen@intel.com>,
	Johannes Weiner <hannes@cmpxchg.org>,
	Mel Gorman <mgorman@suse.de>, Michal Hocko <mhocko@kernel.org>,
	Petr Mladek <pmladek@suse.com>,
	Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
	Vlastimil Babka <vbabka@suse.cz>,
	"yuwang.yuwang" <yuwang.yuwang@alibaba-inc.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Jan Kara <jack@suse.cz>,
	Mathieu Desnoyers <mathieu.desnoyers@efficios.com>
Subject: Re: [PATCH v2] printk: Add console owner and waiter logic to load balance console writes
Date: Fri, 3 Nov 2017 11:19:53 +0100	[thread overview]
Message-ID: <20171103101953.GA5280@quack2.suse.cz> (raw)
In-Reply-To: <20171102130605.05e987e8@gandalf.local.home>

Hi,

On Thu 02-11-17 13:06:05, Steven Rostedt wrote:
> +			if (spin) {
> +				/* We spin waiting for the owner to release us */
> +				spin_acquire(&console_owner_dep_map, 0, 0, _THIS_IP_);
> +				/* Owner will clear console_waiter on hand off */
> +				while (!READ_ONCE(console_waiter))
> +					cpu_relax();

Hum, what prevents us from rescheduling here? And what if the process
stored in console_owner is scheduled out? Both seem to be possible with
CONFIG_PREEMPT kernel? Unless I'm missing something you will need to
disable preemption in some places...

Other than that I like the simplicity of your approach.

								Honza

> +
> +				spin_release(&console_owner_dep_map, 1, _THIS_IP_);
> +				printk_safe_exit_irqrestore(flags);
> +
> +				/*
> +				 * The owner passed the console lock to us.
> +				 * Since we did not spin on console lock, annotate
> +				 * this as a trylock. Otherwise lockdep will
> +				 * complain.
> +				 */
> +				mutex_acquire(&console_lock_dep_map, 0, 1, _THIS_IP_);
> +				console_unlock();
> +				printk_safe_enter_irqsave(flags);
> +			}
> +			printk_safe_exit_irqrestore(flags);
> +
> +		}
>  	}
>  
>  	return printed_len;
> @@ -2141,6 +2196,7 @@ void console_unlock(void)
>  	static u64 seen_seq;
>  	unsigned long flags;
>  	bool wake_klogd = false;
> +	bool waiter = false;
>  	bool do_cond_resched, retry;
>  
>  	if (console_suspended) {
> @@ -2215,6 +2271,20 @@ skip:
>  			goto skip;
>  		}
>  
> +		/*
> +		 * While actively printing out messages, if another printk()
> +		 * were to occur on another CPU, it may wait for this one to
> +		 * finish. This task can not be preempted if there is a
> +		 * waiter waiting to take over.
> +		 */
> +
> +		/* The waiter may spin on us after this */
> +		spin_acquire(&console_owner_dep_map, 0, 0, _THIS_IP_);
> +
> +		raw_spin_lock(&console_owner_lock);
> +		console_owner = current;
> +		raw_spin_unlock(&console_owner_lock);
> +
>  		len += msg_print_text(msg, false, text + len, sizeof(text) - len);
>  		if (nr_ext_console_drivers) {
>  			ext_len = msg_print_ext_header(ext_text,
> @@ -2232,11 +2302,48 @@ skip:
>  		stop_critical_timings();	/* don't trace print latency */
>  		call_console_drivers(ext_text, ext_len, text, len);
>  		start_critical_timings();
> +
> +		raw_spin_lock(&console_owner_lock);
> +		waiter = console_waiter;
> +		console_owner = NULL;
> +		raw_spin_unlock(&console_owner_lock);
> +
> +		/*
> +		 * If there is a waiter waiting for us, then pass the
> +		 * rest of the work load over to that waiter.
> +		 */
> +		if (waiter)
> +			break;
> +
> +		/* There was no waiter, and nothing will spin on us here */
> +		spin_release(&console_owner_dep_map, 1, _THIS_IP_);
> +
>  		printk_safe_exit_irqrestore(flags);
>  
>  		if (do_cond_resched)
>  			cond_resched();
>  	}
> +
> +	/*
> +	 * If there is an active waiter waiting on the console_lock.
> +	 * Pass off the printing to the waiter, and the waiter
> +	 * will continue printing on its CPU, and when all writing
> +	 * has finished, the last printer will wake up klogd.
> +	 */
> +	if (waiter) {
> +		WRITE_ONCE(console_waiter, false);
> +		/* The waiter is now free to continue */
> +		spin_release(&console_owner_dep_map, 1, _THIS_IP_);
> +		/*
> +		 * Hand off console_lock to waiter. The waiter will perform
> +		 * the up(). After this, the waiter is the console_lock owner.
> +		 */
> +		mutex_release(&console_lock_dep_map, 1, _THIS_IP_);
> +		printk_safe_exit_irqrestore(flags);
> +		/* Note, if waiter is set, logbuf_lock is not held */
> +		return;
> +	}
> +
>  	console_locked = 0;
>  
>  	/* Release the exclusive_console once it is used */
-- 
Jan Kara <jack@suse.com>
SUSE Labs, CR

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  parent reply	other threads:[~2017-11-03 10:19 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-26 11:28 [PATCH] mm: don't warn about allocations which stall for too long Tetsuo Handa
2017-10-26 11:28 ` Tetsuo Handa
2017-10-26 11:41 ` Michal Hocko
2017-10-26 11:41   ` Michal Hocko
2017-11-08 10:30   ` peter enderborg
2017-11-09  8:52     ` Michal Hocko
2017-11-09  8:52       ` Michal Hocko
2017-11-09  9:34       ` peter enderborg
2017-11-09 10:09         ` Michal Hocko
2017-11-09 10:09           ` Michal Hocko
2017-11-09 10:19           ` Tetsuo Handa
2017-11-09 10:19             ` Tetsuo Handa
2017-10-26 14:37 ` Johannes Weiner
2017-10-26 14:37   ` Johannes Weiner
2017-10-31 19:32 ` Steven Rostedt
2017-11-01  8:30   ` Vlastimil Babka
2017-11-01  8:30     ` Vlastimil Babka
2017-11-01 13:38     ` Petr Mladek
2017-11-01 13:38       ` Petr Mladek
2017-11-01 15:36       ` Steven Rostedt
2017-11-01 15:36         ` Steven Rostedt
2017-11-02 11:46         ` Petr Mladek
2017-11-02 11:46           ` Petr Mladek
2017-11-02 14:49           ` Steven Rostedt
2017-11-02 14:49             ` Steven Rostedt
2017-11-01 15:33     ` Steven Rostedt
2017-11-01 15:33       ` Steven Rostedt
2017-11-01 17:42       ` Vlastimil Babka
2017-11-01 17:42         ` Vlastimil Babka
2017-11-01 17:54         ` Steven Rostedt
2017-11-01 17:54           ` Steven Rostedt
2017-11-02  8:53   ` Sergey Senozhatsky
2017-11-02  8:53     ` Sergey Senozhatsky
2017-11-02  9:14     ` Sergey Senozhatsky
2017-11-02  9:14       ` Sergey Senozhatsky
2017-11-02 14:55     ` Steven Rostedt
2017-11-02 14:55       ` Steven Rostedt
2017-11-02 12:55   ` Michal Hocko
2017-11-02 12:55     ` Michal Hocko
2017-11-02 15:56 ` Steven Rostedt
2017-11-02 15:56   ` Steven Rostedt
2017-11-02 17:06   ` [PATCH v2] printk: Add console owner and waiter logic to load balance console writes Steven Rostedt
2017-11-02 17:06     ` Steven Rostedt
2017-11-02 17:10     ` Steven Rostedt
2017-11-02 17:10       ` Steven Rostedt
2017-11-02 17:38     ` Steven Rostedt
2017-11-02 17:38       ` Steven Rostedt
2017-11-03 10:19     ` Jan Kara [this message]
2017-11-03 10:19       ` Jan Kara
2017-11-03 11:18       ` Steven Rostedt
2017-11-03 11:18         ` Steven Rostedt

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=20171103101953.GA5280@quack2.suse.cz \
    --to=jack@suse.cz \
    --cc=akpm@linux-foundation.org \
    --cc=dave.hansen@intel.com \
    --cc=hannes@cmpxchg.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mathieu.desnoyers@efficios.com \
    --cc=mgorman@suse.de \
    --cc=mhocko@kernel.org \
    --cc=penguin-kernel@I-love.SAKURA.ne.jp \
    --cc=peterz@infradead.org \
    --cc=pmladek@suse.com \
    --cc=rostedt@goodmis.org \
    --cc=sergey.senozhatsky@gmail.com \
    --cc=torvalds@linux-foundation.org \
    --cc=vbabka@suse.cz \
    --cc=xiyou.wangcong@gmail.com \
    --cc=yuwang.yuwang@alibaba-inc.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.