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]
Message-ID: <20060913193349.GA21182@dbz.icequake.net>
Date:	Wed, 13 Sep 2006 14:33:49 -0500
From:	Ryan Underwood <nemesis-lists@...quake.net>
To:	linux-kernel@...r.kernel.org
Subject: odd cpu_khz behavior (doubling)


OK, I realize cpu_khz should not be constant.  I wrote a cpufreq driver
for Toshiba laptops that has two CPU speeds (1/2 and full) and that is
managed via SMM.  Can anyone tell me why cpu_khz occasionally doubles
itself while the driver is in use?  It starts at ~120000, then doubles
to ~240000, then to ~480000, etc.  However, this is over ~15 loads and
unloads of the cpufreq target driver.  Small cpu_khz variations would
not bother anything wrt the cpufreq driver, but this is just plain
incorrect for things like /proc/cpuinfo which also use cpu_khz.

Is this a TSC issue (maybe related to SMM) or am I just doing something
in the wrong place in the driver?

Attached is the cpufreq part of the code, just a testing version.

-- 
Ryan Underwood, <nemesis@...quake.net>

View attachment "toshiba_freq.c" of type "text/x-csrc" (7293 bytes)

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ