linux-kernel - cpufreq userspace not listening

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>] [day] [month] [year] [list]

Date:	Mon, 4 Jul 2011 20:41:51 +0200
From:	Wim Heirman <wim@...rman.net>
To:	linux-kernel@...r.kernel.org
Subject: cpufreq userspace not listening

Hi,

I'm trying to set the frequencies on a quad core Nehalem processor
(Xeon X5550). I want all four cores to have a different, fixed
frequency (I'm doing research on load-balancing applications). For
this processor type,
/sys/devices/system/cpu/cpu*/cpufreq/affected_cpus lists only a single
core so I would assume that, when using the userspace governor and
writing different frequencies to scaling_setspeed, it should work.

Yet when I try this, cpuinfo_cur_freq often reports a different
frequency than the assigned one. Concretely, when loading the system,
cores seem to throttle up to the maximum frequency assigned to any of
the cores in the same package, rather than sticking to their own
assigned frequency.

Are there any settings I may have missed? Is it at all possible to use
different frequencies within a socket? scaling_driver is acpi-cpufreq,
kernel version 2.6.38 (Ubuntu 11.04).

Thanks,
Wim
--
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/