linux-kernel - Re: [PATCH 1/2] sched: normalize sleeper's vruntime during group change

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <AANLkTikVQist_VnCcpZzU78t0TVuocwOcDPG4POVzSdv@mail.gmail.com>
Date:	Thu, 30 Sep 2010 12:14:42 -0700
From:	Dima Zavin <dima@...roid.com>
To:	Peter Zijlstra <peterz@...radead.org>
Cc:	linux-kernel@...r.kernel.org, Ingo Molnar <mingo@...e.hu>,
	Mike Galbraith <efault@....de>,
	Arve Hjønnevåg <arve@...roid.com>
Subject: Re: [PATCH 1/2] sched: normalize sleeper's vruntime during group change

Peter,

>> This adds a new callback, prep_move_task, to struct sched_class
>> to give sched_fair the opportunity to adjust the task's vruntime
>> just before setting its new group.
>>
>> This allows us to properly normalize a sleeping task's vruntime
>> when moving it between different cgroups.
>
> Don't much like these patches, and changelogs need full descriptions not
> fuzzy links out to the interwebs that might or might not stay valid.

Sorry, I'll resend with a better changelog and no tubes.

> If you change a task's cgroup, the task is new in that cgroup and should
> be placed like a new task would, carrying over any history it has in the
> old cgroup is dubious at best.

That is certainly not the behavior today for running tasks as I
understand it. They get their vruntime normalized before being taken
off the run_queue of old group, and then get the new min_vruntime
added when they get re-enqueued. For sleeping tasks it's just plain
broken (see below).

Also, I am skeptical that the behavior you describe is desired. If you
were to just place the task in the new cgroup like it was a brand new
task, it could allow threads to game the system and potentially let
them reset their vruntime back. If I do a bunch of work and then move
myself out of the group and then back onto it, I may get lower
vruntime than by just staying on the group.

> Please explain this stuff..

The situation today is quite bad for sleeping tasks. Currently, when
you move a sleeping thread between cgroups, the thread can retain its
old vruntime value if the old group was far ahead of the new group
since it essentially does a max(se->vruntime, new_vruntime) in
place_entity. This can prevent the task from running for a very long
time. That is what this patch was trying to address. It normalizes the
sleeper thread's vruntime before moving it to the new group.

Thanks in advance.

--Dima
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/