All of lore.kernel.org
 help / color / mirror / Atom feed
From: Vlastimil Babka <vbabka-AlSwsSmVLrQ@public.gmane.org>
To: Sebastian Andrzej Siewior
	<bigeasy-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org>,
	linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org
Cc: Thomas Gleixner <tglx-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org>,
	Andrew Morton
	<akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org>,
	Luis Chamberlain <mcgrof-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
	Kees Cook <keescook-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org>,
	Iurii Zaikin <yzaikin-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
	Mel Gorman
	<mgorman-3eNAlZScCAx27rWaFMvyedHuzzzSOjJt@public.gmane.org>,
	Linux API <linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>
Subject: Re: [PATCH] mm/compaction: Disable compact_unevictable_allowed on RT
Date: Wed, 15 Jan 2020 23:04:19 +0100	[thread overview]
Message-ID: <4cf4507b-0632-34e6-5985-df933559af9f@suse.cz> (raw)
In-Reply-To: <20200115161035.893221-1-bigeasy-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org>

On 1/15/2020 5:10 PM, Sebastian Andrzej Siewior wrote:
> Since commit
>     5bbe3547aa3ba ("mm: allow compaction of unevictable pages")
> 
> it is allowed to examine mlocked pages and compact them by default.
> On -RT even minor pagefaults are problematic because it may take a few
> 100us to resolve them and until then the task is blocked.

Fine, this makes sense on RT I guess. There might be some trade-off for
high-order allocation latencies though. We could perhaps migrate such mlocked
pages to pages allocated without __GFP_MOVABLE during the mlock() to at least
somewhat prevent them being scattered all over the zones. For MCL_FUTURE,
allocate them as unmovable from the beginning. But that can wait until issues
are reported.
I assume you have similar solution for NUMA balancing and whatever else can
cause minor faults?

> Make compact_unevictable_allowed = 0 default and remove it from /proc on
> RT.

Removing it is maybe going too far in terms of RT kernel differences confusing
users? Change the default sure, perhaps making it read-only, but removing?

> Link: https://lore.kernel.org/linux-mm/20190710144138.qyn4tuttdq6h7kqx-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org/

In any case the sysctl Documentation/ should be updated? And perhaps also the
mlock manpage as you noted in the older thread above?

Thanks,
Vlastimil

> Signed-off-by: Sebastian Andrzej Siewior <bigeasy-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org>
> ---
>  kernel/sysctl.c | 3 ++-
>  mm/compaction.c | 4 ++++
>  2 files changed, 6 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/sysctl.c b/kernel/sysctl.c
> index 70665934d53e2..d08bd51a0fbc3 100644
> --- a/kernel/sysctl.c
> +++ b/kernel/sysctl.c
> @@ -1488,6 +1488,7 @@ static struct ctl_table vm_table[] = {
>  		.extra1		= &min_extfrag_threshold,
>  		.extra2		= &max_extfrag_threshold,
>  	},
> +#ifndef CONFIG_PREEMPT_RT
>  	{
>  		.procname	= "compact_unevictable_allowed",
>  		.data		= &sysctl_compact_unevictable_allowed,
> @@ -1497,7 +1498,7 @@ static struct ctl_table vm_table[] = {
>  		.extra1		= SYSCTL_ZERO,
>  		.extra2		= SYSCTL_ONE,
>  	},
> -
> +#endif
>  #endif /* CONFIG_COMPACTION */
>  	{
>  		.procname	= "min_free_kbytes",
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 672d3c78c6abf..b2c804c35ae56 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -1590,7 +1590,11 @@ typedef enum {
>   * Allow userspace to control policy on scanning the unevictable LRU for
>   * compactable pages.
>   */
> +#ifdef CONFIG_PREEMPT_RT
> +#define sysctl_compact_unevictable_allowed 0
> +#else
>  int sysctl_compact_unevictable_allowed __read_mostly = 1;
> +#endif
>  
>  static inline void
>  update_fast_start_pfn(struct compact_control *cc, unsigned long pfn)
> 

WARNING: multiple messages have this Message-ID (diff)
From: Vlastimil Babka <vbabka@suse.cz>
To: Sebastian Andrzej Siewior <bigeasy@linutronix.de>, linux-mm@kvack.org
Cc: Thomas Gleixner <tglx@linutronix.de>,
	Andrew Morton <akpm@linux-foundation.org>,
	Luis Chamberlain <mcgrof@kernel.org>,
	Kees Cook <keescook@chromium.org>,
	Iurii Zaikin <yzaikin@google.com>,
	Mel Gorman <mgorman@techsingularity.net>,
	Linux API <linux-api@vger.kernel.org>
Subject: Re: [PATCH] mm/compaction: Disable compact_unevictable_allowed on RT
Date: Wed, 15 Jan 2020 23:04:19 +0100	[thread overview]
Message-ID: <4cf4507b-0632-34e6-5985-df933559af9f@suse.cz> (raw)
In-Reply-To: <20200115161035.893221-1-bigeasy@linutronix.de>

On 1/15/2020 5:10 PM, Sebastian Andrzej Siewior wrote:
> Since commit
>     5bbe3547aa3ba ("mm: allow compaction of unevictable pages")
> 
> it is allowed to examine mlocked pages and compact them by default.
> On -RT even minor pagefaults are problematic because it may take a few
> 100us to resolve them and until then the task is blocked.

Fine, this makes sense on RT I guess. There might be some trade-off for
high-order allocation latencies though. We could perhaps migrate such mlocked
pages to pages allocated without __GFP_MOVABLE during the mlock() to at least
somewhat prevent them being scattered all over the zones. For MCL_FUTURE,
allocate them as unmovable from the beginning. But that can wait until issues
are reported.
I assume you have similar solution for NUMA balancing and whatever else can
cause minor faults?

> Make compact_unevictable_allowed = 0 default and remove it from /proc on
> RT.

Removing it is maybe going too far in terms of RT kernel differences confusing
users? Change the default sure, perhaps making it read-only, but removing?

> Link: https://lore.kernel.org/linux-mm/20190710144138.qyn4tuttdq6h7kqx@linutronix.de/

In any case the sysctl Documentation/ should be updated? And perhaps also the
mlock manpage as you noted in the older thread above?

Thanks,
Vlastimil

> Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
> ---
>  kernel/sysctl.c | 3 ++-
>  mm/compaction.c | 4 ++++
>  2 files changed, 6 insertions(+), 1 deletion(-)
> 
> diff --git a/kernel/sysctl.c b/kernel/sysctl.c
> index 70665934d53e2..d08bd51a0fbc3 100644
> --- a/kernel/sysctl.c
> +++ b/kernel/sysctl.c
> @@ -1488,6 +1488,7 @@ static struct ctl_table vm_table[] = {
>  		.extra1		= &min_extfrag_threshold,
>  		.extra2		= &max_extfrag_threshold,
>  	},
> +#ifndef CONFIG_PREEMPT_RT
>  	{
>  		.procname	= "compact_unevictable_allowed",
>  		.data		= &sysctl_compact_unevictable_allowed,
> @@ -1497,7 +1498,7 @@ static struct ctl_table vm_table[] = {
>  		.extra1		= SYSCTL_ZERO,
>  		.extra2		= SYSCTL_ONE,
>  	},
> -
> +#endif
>  #endif /* CONFIG_COMPACTION */
>  	{
>  		.procname	= "min_free_kbytes",
> diff --git a/mm/compaction.c b/mm/compaction.c
> index 672d3c78c6abf..b2c804c35ae56 100644
> --- a/mm/compaction.c
> +++ b/mm/compaction.c
> @@ -1590,7 +1590,11 @@ typedef enum {
>   * Allow userspace to control policy on scanning the unevictable LRU for
>   * compactable pages.
>   */
> +#ifdef CONFIG_PREEMPT_RT
> +#define sysctl_compact_unevictable_allowed 0
> +#else
>  int sysctl_compact_unevictable_allowed __read_mostly = 1;
> +#endif
>  
>  static inline void
>  update_fast_start_pfn(struct compact_control *cc, unsigned long pfn)
> 



  parent reply	other threads:[~2020-01-15 22:04 UTC|newest]

Thread overview: 19+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-01-15 16:10 [PATCH] mm/compaction: Disable compact_unevictable_allowed on RT Sebastian Andrzej Siewior
     [not found] ` <20200115161035.893221-1-bigeasy-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org>
2020-01-15 22:04   ` Vlastimil Babka [this message]
2020-01-15 22:04     ` Vlastimil Babka
2020-01-16 10:22     ` Sebastian Andrzej Siewior
2020-03-02 17:35     ` [PATCH v2] " Sebastian Andrzej Siewior
2020-03-02 21:25       ` Andrew Morton
2020-03-03 17:59         ` Sebastian Andrzej Siewior
2020-03-03 20:20           ` [PATCH 1/2] =?UTF-8?q?mm/compaction:=20Really=20limit=20compact?= =?UTF-8?q?=5Funevictable=5Fallowed=20to=200=E2=80=A61?= Sebastian Andrzej Siewior
2020-03-03 20:22             ` [PATCH 2/2 v3] mm/compaction: Disable compact_unevictable_allowed on RT Sebastian Andrzej Siewior
2020-03-03 23:56               ` Andrew Morton
2020-03-04  8:19                 ` Vlastimil Babka
2020-03-04  9:27                   ` Sebastian Andrzej Siewior
2020-03-19 16:39                 ` [PATCH 2/2 v4] " Sebastian Andrzej Siewior
2020-03-19 16:49                   ` Vlastimil Babka
2020-03-19 16:55                     ` [PATCH 2/2 v5] " Sebastian Andrzej Siewior
2020-03-04  8:18               ` [PATCH 2/2 v3] " Vlastimil Babka
2020-03-04  9:25                 ` Sebastian Andrzej Siewior
2020-03-04  9:11               ` Mel Gorman
2020-03-04  8:12             ` [PATCH 1/2] mm/compaction: Really limit compact_unevictable_allowed to 0…1 Vlastimil Babka

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=4cf4507b-0632-34e6-5985-df933559af9f@suse.cz \
    --to=vbabka-alswssmvlrq@public.gmane.org \
    --cc=akpm-de/tnXTf+JLsfHDXvbKv3WD2FQJk+8+b@public.gmane.org \
    --cc=bigeasy-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org \
    --cc=keescook-F7+t8E8rja9g9hUCZPvPmw@public.gmane.org \
    --cc=linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
    --cc=mcgrof-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
    --cc=mgorman-3eNAlZScCAx27rWaFMvyedHuzzzSOjJt@public.gmane.org \
    --cc=tglx-hfZtesqFncYOwBW4kG4KsQ@public.gmane.org \
    --cc=yzaikin-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.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 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.