linux-kernel - Re: AMD Quad Core clock problem?

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: <480403DE.4070901@zytor.com>
Date:	Mon, 14 Apr 2008 18:24:46 -0700
From:	"H. Peter Anvin" <hpa@...or.com>
To:	Krzysztof Halasa <khc@...waw.pl>
CC:	Lennart Sorensen <lsorense@...lub.uwaterloo.ca>,
	Marc Perkel <mperkel@...oo.com>,
	Chris Snook <csnook@...hat.com>, linux-kernel@...r.kernel.org
Subject: Re: AMD Quad Core clock problem?

Krzysztof Halasa wrote:
> lsorense@...lub.uwaterloo.ca (Lennart Sorensen) writes:
> 
>> True, but usually they give you an option to turn it off in the BIOS to
>> make the system actually work with things like NTP.
>>
>> Maybe I should have said "Any system with spread spectrum enabled".
> 
> I haven't investigated those spread spectrums at all but sometimes
> there are options: "center" (the frequency is centered around the
> nominal value, I guess) and "below" (or something like that). If done
> correctly, "center" shouldn't have visible effect on NTP, should it?

It shouldn't really matter, since "below" produces a similar spectrum at 
a slightly lower frequency.

The main issue is how slow the frequency slew really is; typically the 
slew should well average out to zero over the timescale that NTP cares 
about.

> The other important question is "what frequency has spread spectrum
> enabled"? CPU (= TSC), HPET, RAM, something else?

In PCs, the TSC and RAM will be affected; the HPET, PMTMR and PIT are 
fed from the 14.31818 MHz timekeeping crystal which is generally *not* 
spread.

	-hpa
--
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/