On Mon 2015-08-24 13:30:43, Andrew Morton wrote:
> On Mon, 24 Aug 2015 17:13:23 +0200 Petr Mladek <pmladek@xxxxxxxx> wrote:
>
> > The commit 1dfb059b9438633b0546 ("thp: reduce khugepaged freezing
> > latency") fixed khugepaged to do not block a system suspend. But
> > the result is that it could not get interrupted before the given
> > timeout because the condition for the wait event is "false".
>
> What are the userspace-visible effects of this bug?
I believe that the change will not make any visible difference. It
is just a bit cleaner code.
If I get it correctly. This function is called when the daemon
is not able to allocate any new huge page. It is used to throttle the
attempts. Then the thread is waken in the following situations:
+ when user modifies "alloc_sleep" or "scan_sleep" from sysfs;
this is rare
+ in __khugepaged_enter() when there is a new page to scan and
the list was empty before. This is because the same waitqueue
is used to wait between scans. IMHO, it is kind of bug to mix
these two things. But I guess that this wake is rare as well.
Also I guess that it will be solved by Vlastimil's rework.
+ when the kthread is stopped; this is the only place when it could
make a visible difference if the sleep is longer; but this is
rare situation as well
Best Regards,
Petr
> > This patch puts back the original approach but it uses
> > freezable_schedule_timeout_interruptible() instead of
> > schedule_timeout_interruptible(). It does the right thing.
> > I am pretty sure that the freezable variant was not used in
> > the original fix only because it was not available at that time.
> >
> > The regression has been there for ages. It was not critical. It just
> > did the allocation throttling a little bit more aggressively.
> >
> > I found this problem when converting the kthread to kthread worker API
> > and trying to understand the code.
> >
> > ...
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>