On Thu, Sep 22, 2005 at 09:50:28PM +0200, Ernest jw ter Kuile wrote: > On Thursday 22 September 2005 14:49, Jo Shields wrote: > > dmesg | grep powernow-k8 > > anybody knows what this means ? > > terkuile:~$ dmesg | grep powernow-k8 > powernow-k8: Found 1 AMD Athlon 64 / Opteron processors (version 1.50.3) > powernow-k8: BIOS error: maxvid exceeded with pstate 2 > > I can find the error string in the kernel source code, however there is this > amazing amount of comments associated with it. Google is also quite silent > about this specific error.
There is a MSR onto the processor which will give what is the max voltage (and the max multiplier as well) that can be applied to the processor. When the driver parse a BIOS table in order to get the supported pair of frequency/voltage, it detect that such an entry is out of range. Therefore the driver bail out. > > Anybody else ever seen error above ? There were some BIOS that give configurations for an XP-M instead of an AMD64 and give the same error (though it was for a laptop). By warning the OEM it was possible to make this work after a BIOS upgrade. An other option is to hack the driver by hardcoding a good table instead of the one supplied by the BIOS. Also if you have ACPI enabled, you can correct that by hacking a little the DSDT (or one of the SSDT). See http://acpi.sf.net/ (for instruction on how to do so), and the documentations I pointed to Mike in order to know what to put for a corrected table. -- Bruno Ducrot -- Which is worse: ignorance or apathy? -- Don't know. Don't care. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
