public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Tejun Heo <tj@kernel.org>
Cc: Lianwei Wang <lianwei.wang@gmail.com>,
	linux-kernel@vger.kernel.org, "Rafael J. Wysocki" <rjw@sisk.pl>,
	Mandeep Singh Baines <msb@chromium.org>
Subject: Re: PATCH: freezer: add fake signal clearing back when thaw task
Date: Tue, 26 Feb 2013 15:28:04 +0100	[thread overview]
Message-ID: <20130226142804.GC28940@redhat.com> (raw)
In-Reply-To: <20130225235313.GE2679@htj.dyndns.org>

Ah, I just noticed that the changelog mentions wait_event_freezable() too,

On 02/25, Tejun Heo wrote:
>
> > Subject: [PATCH] freezer: add fake signal clearing back when thaw task
> >
> > The fake TIF_SIGPENDING is set during freeze userspace process, but it
> > is not cleared when thaw tasks after below commit:
> >   34b087e freezer: kill unused set_freezable_with_signal()
> >
> > This will cause the userspace task that wait_event_freezable

But there is no userspace users?

I guess wait_event_freezable() is only used to avoid the contribution
to load_avg. So we can simplify it, __wait_event_freezable() can simply
do freezable_schedule() or wait_event_freezable() can do
freezer_do_not_count + wait_event_interruptible. Or we are going to add
more users?

As for current users, I think they do not need this interface. Something
like below. I was going to try to convert khugepaged/ksm_scan_thread
yesterday but I was distracted, will try to do tomorrow.

Oleg.

--- x/mm/huge_memory.c
+++ x/mm/huge_memory.c
@@ -2616,7 +2616,7 @@ static int khugepaged_has_work(void)
 static int khugepaged_wait_event(void)
 {
 	return !list_empty(&khugepaged_scan.mm_head) ||
-		kthread_should_stop();
+		kthread_should_stop_or_freeze();
 }
 
 static void khugepaged_do_scan(void)
@@ -2655,20 +2655,18 @@ static void khugepaged_do_scan(void)
 
 static void khugepaged_wait_work(void)
 {
-	try_to_freeze();
-
 	if (khugepaged_has_work()) {
 		if (!khugepaged_scan_sleep_millisecs)
 			return;
 
-		wait_event_freezable_timeout(khugepaged_wait,
-					     kthread_should_stop(),
+		wait_event_interruptible_timeout(khugepaged_wait,
+					     kthread_should_stop_or_freeze(),
 			msecs_to_jiffies(khugepaged_scan_sleep_millisecs));
 		return;
 	}
 
 	if (khugepaged_enabled())
-		wait_event_freezable(khugepaged_wait, khugepaged_wait_event());
+		wait_event_interruptible(khugepaged_wait, khugepaged_wait_event());
 }
 
 static int khugepaged(void *none)
@@ -2678,7 +2676,7 @@ static int khugepaged(void *none)
 	set_freezable();
 	set_user_nice(current, 19);
 
-	while (!kthread_should_stop()) {
+	while (!kthread_freezable_should_stop(NULL)) {
 		khugepaged_do_scan();
 		khugepaged_wait_work();
 	}


  parent reply	other threads:[~2013-02-26 14:29 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-02-21  6:19 PATCH: freezer: add fake signal clearing back when thaw task Lianwei Wang
2013-02-25 23:53 ` Tejun Heo
2013-02-26 14:14   ` Oleg Nesterov
2013-02-26 14:54     ` Oleg Nesterov
2013-02-26 14:28   ` Oleg Nesterov [this message]
2013-03-04  3:24   ` Lianwei Wang
2013-03-04 17:01     ` Tejun Heo
2013-03-04 17:32     ` Oleg Nesterov

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=20130226142804.GC28940@redhat.com \
    --to=oleg@redhat.com \
    --cc=lianwei.wang@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=msb@chromium.org \
    --cc=rjw@sisk.pl \
    --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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox