linux-kernel - How does group scheduling interact with sched_batch, sched

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 for Android: free password hash cracker in your pocket

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

Date:	Wed, 20 Jul 2011 18:32:21 -0500
From:	Steve Bergman <sbergman27@...il.com>
To:	linux-kernel@...r.kernel.org
Subject: How does group scheduling interact with sched_batch, sched_idle, etc?

I normally run einstein@...e under BOINC at a nice level of 19 with
sched_batch. I like the longer timeslices of sched_batch from the
standpoint of L2 cache efficiency. I've now turned
sched_autogroup_enable on, and set the autogroup to nice 19. The BOINC
startup script sets the individual boinc application processes to nice
19 and sched_batch. I'm trying to understand what things I've actually
changed. Do I still have the longer timeslices of sched_batch, along
with the guarantee that they will never be treated as interactive? How
do the autogroup nice level, the individual process nice level, and
the sched_normal, sched_idleprio, and sched_normal parameters all play
together?

Also, I'm wondering what the impact of the group scheduling overhead
might be in the most common case, where I'm not using the machine, and
it's mainly just the 4 boinc processes, all in the same cgroup, using
the processor.

I'm running 2.6.38 x86_64.

Thanks for any information,
Steve Bergman
--
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/