All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrea Arcangeli <aarcange@redhat.com>
To: Tejun Heo <tj@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>,
	"Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com>,
	"Rafael J. Wysocki" <rjw@suse.com>,
	linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org,
	Jiri Slaby <jirislaby@gmail.com>,
	linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH] thp: reduce khugepaged freezing latency
Date: Wed, 9 Nov 2011 20:40:48 +0100	[thread overview]
Message-ID: <20111109194047.GO5075@redhat.com> (raw)
In-Reply-To: <20111109183447.GG1260@google.com>

On Wed, Nov 09, 2011 at 10:34:47AM -0800, Tejun Heo wrote:
> know.  My instinct tells me to strongly recommend use of
> wait_event_freezable_timeout() and run away.  :)

Passing false wasn't so appealing to me but ok. Jiri can you test this
with some suspend? (beware builds but untested)

===
From: Andrea Arcangeli <aarcange@redhat.com>
Subject: thp: reduce khugepaged freezing latency

Use wait_event_freezable_timeout() instead of
schedule_timeout_interruptible() to avoid missing freezer wakeups. A
try_to_freeze() would have been needed in the
khugepaged_alloc_hugepage tight loop too in case of the allocation
failing repeatedly, and wait_event_freezable_timeout will provide it
too.

khugepaged would still freeze just fine by trying again the next
minute but it's better if it freezes immediately.

Reported-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
---
 mm/huge_memory.c |   14 ++++----------
 1 files changed, 4 insertions(+), 10 deletions(-)

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 4298aba..fd925d0 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2259,12 +2259,9 @@ static void khugepaged_do_scan(struct page **hpage)
 
 static void khugepaged_alloc_sleep(void)
 {
-	DEFINE_WAIT(wait);
-	add_wait_queue(&khugepaged_wait, &wait);
-	schedule_timeout_interruptible(
-		msecs_to_jiffies(
-			khugepaged_alloc_sleep_millisecs));
-	remove_wait_queue(&khugepaged_wait, &wait);
+	wait_event_freezable_timeout(khugepaged_wait, false,
+				     msecs_to_jiffies(
+					     khugepaged_alloc_sleep_millisecs));
 }
 
 #ifndef CONFIG_NUMA
@@ -2313,14 +2310,11 @@ static void khugepaged_loop(void)
 		if (unlikely(kthread_should_stop()))
 			break;
 		if (khugepaged_has_work()) {
-			DEFINE_WAIT(wait);
 			if (!khugepaged_scan_sleep_millisecs)
 				continue;
-			add_wait_queue(&khugepaged_wait, &wait);
-			schedule_timeout_interruptible(
+			wait_event_freezable_timeout(khugepaged_wait, false,
 				msecs_to_jiffies(
 					khugepaged_scan_sleep_millisecs));
-			remove_wait_queue(&khugepaged_wait, &wait);
 		} else if (khugepaged_enabled())
 			wait_event_freezable(khugepaged_wait,
 					     khugepaged_wait_event());

WARNING: multiple messages have this Message-ID (diff)
From: Andrea Arcangeli <aarcange@redhat.com>
To: Tejun Heo <tj@kernel.org>
Cc: Oleg Nesterov <oleg@redhat.com>,
	"Srivatsa S. Bhat" <srivatsa.bhat@linux.vnet.ibm.com>,
	"Rafael J. Wysocki" <rjw@suse.com>,
	linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org,
	Jiri Slaby <jirislaby@gmail.com>,
	linux-mm@kvack.org, Andrew Morton <akpm@linux-foundation.org>
Subject: Re: [PATCH] thp: reduce khugepaged freezing latency
Date: Wed, 9 Nov 2011 20:40:48 +0100	[thread overview]
Message-ID: <20111109194047.GO5075@redhat.com> (raw)
In-Reply-To: <20111109183447.GG1260@google.com>

On Wed, Nov 09, 2011 at 10:34:47AM -0800, Tejun Heo wrote:
> know.  My instinct tells me to strongly recommend use of
> wait_event_freezable_timeout() and run away.  :)

Passing false wasn't so appealing to me but ok. Jiri can you test this
with some suspend? (beware builds but untested)

===
From: Andrea Arcangeli <aarcange@redhat.com>
Subject: thp: reduce khugepaged freezing latency

Use wait_event_freezable_timeout() instead of
schedule_timeout_interruptible() to avoid missing freezer wakeups. A
try_to_freeze() would have been needed in the
khugepaged_alloc_hugepage tight loop too in case of the allocation
failing repeatedly, and wait_event_freezable_timeout will provide it
too.

khugepaged would still freeze just fine by trying again the next
minute but it's better if it freezes immediately.

Reported-by: Jiri Slaby <jslaby@suse.cz>
Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
---
 mm/huge_memory.c |   14 ++++----------
 1 files changed, 4 insertions(+), 10 deletions(-)

diff --git a/mm/huge_memory.c b/mm/huge_memory.c
index 4298aba..fd925d0 100644
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@ -2259,12 +2259,9 @@ static void khugepaged_do_scan(struct page **hpage)
 
 static void khugepaged_alloc_sleep(void)
 {
-	DEFINE_WAIT(wait);
-	add_wait_queue(&khugepaged_wait, &wait);
-	schedule_timeout_interruptible(
-		msecs_to_jiffies(
-			khugepaged_alloc_sleep_millisecs));
-	remove_wait_queue(&khugepaged_wait, &wait);
+	wait_event_freezable_timeout(khugepaged_wait, false,
+				     msecs_to_jiffies(
+					     khugepaged_alloc_sleep_millisecs));
 }
 
 #ifndef CONFIG_NUMA
@@ -2313,14 +2310,11 @@ static void khugepaged_loop(void)
 		if (unlikely(kthread_should_stop()))
 			break;
 		if (khugepaged_has_work()) {
-			DEFINE_WAIT(wait);
 			if (!khugepaged_scan_sleep_millisecs)
 				continue;
-			add_wait_queue(&khugepaged_wait, &wait);
-			schedule_timeout_interruptible(
+			wait_event_freezable_timeout(khugepaged_wait, false,
 				msecs_to_jiffies(
 					khugepaged_scan_sleep_millisecs));
-			remove_wait_queue(&khugepaged_wait, &wait);
 		} else if (khugepaged_enabled())
 			wait_event_freezable(khugepaged_wait,
 					     khugepaged_wait_event());

--
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/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

  reply	other threads:[~2011-11-09 19:41 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-11-08  8:33 khugepaged doesn't want to freeze Jiri Slaby
2011-11-08  8:33 ` Jiri Slaby
2011-11-08 15:29 ` Andrea Arcangeli
2011-11-08 15:29   ` Andrea Arcangeli
2011-11-08 15:29   ` [PATCH] thp: reduce khugepaged freezing latency Andrea Arcangeli
2011-11-08 15:29     ` Andrea Arcangeli
2011-11-08 20:01     ` Srivatsa S. Bhat
2011-11-08 20:01       ` Srivatsa S. Bhat
2011-11-09  0:01       ` Andrea Arcangeli
2011-11-09  0:01         ` Andrea Arcangeli
2011-11-09  9:03         ` Srivatsa S. Bhat
2011-11-09  9:03           ` Srivatsa S. Bhat
2011-11-09 12:45       ` Srivatsa S. Bhat
2011-11-09 12:45         ` Srivatsa S. Bhat
2011-11-09 15:53         ` Tejun Heo
2011-11-09 15:53           ` Tejun Heo
2011-11-09 16:20           ` Srivatsa S. Bhat
2011-11-09 16:20             ` Srivatsa S. Bhat
2011-11-09 16:52           ` Andrea Arcangeli
2011-11-09 16:52             ` Andrea Arcangeli
2011-11-09 16:59             ` Tejun Heo
2011-11-09 16:59               ` Tejun Heo
2011-11-09 17:02               ` Tejun Heo
2011-11-09 17:02                 ` Tejun Heo
2011-11-09 17:29                 ` Andrea Arcangeli
2011-11-09 17:29                   ` Andrea Arcangeli
2011-11-09 18:09                   ` Tejun Heo
2011-11-09 18:09                     ` Tejun Heo
2011-11-09 18:19                     ` Andrea Arcangeli
2011-11-09 18:19                       ` Andrea Arcangeli
2011-11-09 18:34                       ` Tejun Heo
2011-11-09 18:34                         ` Tejun Heo
2011-11-09 19:40                         ` Andrea Arcangeli [this message]
2011-11-09 19:40                           ` Andrea Arcangeli
2011-11-11 12:20                           ` Jiri Slaby
2011-11-11 12:20                             ` Jiri Slaby
2011-11-09 17:06               ` Tejun Heo
2011-11-09 17:06                 ` Tejun Heo
2011-11-09 17:33                 ` Andrea Arcangeli
2011-11-09 17:33                   ` Andrea Arcangeli

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=20111109194047.GO5075@redhat.com \
    --to=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=jirislaby@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=oleg@redhat.com \
    --cc=rjw@suse.com \
    --cc=srivatsa.bhat@linux.vnet.ibm.com \
    --cc=tj@kernel.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.