Date: Sun, 3 Aug 2008 06:43:28 -0700 From: Jeremy Chadwick <koitsu@FreeBSD.org> To: Torfinn Ingolfsen <torfinn.ingolfsen@broadpark.no> Cc: freebsd-stable@freebsd.org Subject: Re: Temperature monitoring on old desktop - Dell OptiPlex SX270? Message-ID: <20080803134328.GA65256@eos.sc1.parodius.com> In-Reply-To: <20080803135251.60d4bb7d.torfinn.ingolfsen@broadpark.no> References: <20080803015053.e67a39ee.torfinn.ingolfsen@broadpark.no> <20080803031912.GA38781@eos.sc1.parodius.com> <20080803135251.60d4bb7d.torfinn.ingolfsen@broadpark.no>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Aug 03, 2008 at 01:52:51PM +0200, Torfinn Ingolfsen wrote: > On Sat, 02 Aug 2008 20:19:12 -0700 > Jeremy Chadwick <koitsu@FreeBSD.org> wrote: > > > On Sun, Aug 03, 2008 at 01:50:53AM +0200, Torfinn Ingolfsen wrote: > > The first questions to ask are: 1) does this machine even have a H/W > > monitoring IC on it, and 2) is it enabled/wired to thermistors and > > fans? > > Yes, but so far I haven't found out anything by searching. Then the only possibility is to take a very high-resolution photo (read: 2048x1536 or higher) and send it to someone who can identify ICs (I'm good at recognising H/W monitoring ICs :-) ). But even that won't guarantee anything; an IC that supports H/W monitoring may be found/present but it may not be wired to support such (or the board lacks thermistors), or possibly the silkscreening is false (which I myself have personally seen on some server boards). > Ok, so what is the 'TM' feature of this cpu then? > cpuid thinks it is a thermal monitor: > Intel-specific functions: > Version 00000f29: > Type 0 - Original OEM > Family 15 - Pentium 4 > Extended family 0 > Model 2 - Intel Pentium 4 processor (generic) or newer > Stepping 9 > Reserved 0 This gets into semantics: "what exactly does 'monitor' mean?" The P4 TM feature is more of a thermal manager and not so much a "monitor" in the sense of what you think it might be (re: ability to provide thermal statistics to a program). It *is* a "monitor" in the sense that it reads temperature, but there's no way to access that internal data. I believe the P4 TM is used to decrease power usage (disabling some features, enabling some power-saving modes, etc.) based on temperature. It probably induces some form of clock throttling too, but it's probably done very differently compared to present-day Core2Duo processors, for example (and those processors DO have on-die per-core temperature monitors which you can monitor, re: coretemp(4)). Here's some reference material confirming my claim: http://www.intel.com/support/processors/pentium4/sb/cs-007999.htm http://www.digit-life.com/articles2/intel-thermal-features/index.html http://www.overclockers.com/articles517/ The point is, none of the internal data is accessible. I have never seen a program provide any sort of thermal data on a P4, unless there's an external H/W monitoring IC (with a thermistor that sits physically under the processor mounting socket) installed. You'll find such features on P4 server boards -- commonly Winbond ICs, with a thermistor that sits directly under the CPU. > > I just checked http://tingox.googlepages.com/sx270 and sure enough, an > > older P4. coretemp(4) won't work with this. > > I know, I just thought that ther might be something similar for the TM feature of Pentium 4's. Nope, because it's not something you can get data from. > > I would start by booting the machine into Windows and install > > SpeedFan. If that thing is able to detect and provide thermal data, > > Ouch. I was hoping that I wouldn't have to do that. The machine have no internal CD-drive, > and for some reason doesn't want to boot from a (usb) external cd-drive either (kind of funny - it boots from flash drives and external hard drives. But cd-rom -no). > > I was hoping to solve this without windows in the picture. You could try Linux. Their lm-sensors project is incredibly thorough, but based on what I've looked at in the code, it's hit-or-miss. It does do a form of auto-probing (which is very risky IMHO). Speedfan's auto-detection method is better, and safer (especially if there's an SMBus interface with H/W monitoring IC tie-ins made available). Again, this would only allow you to detect whether or not there's an actual H/W monitoring IC on the board somewhere. I'm strongly doubting there is. You could contact Dell and ask, but their Support folks probably have no idea what a hardware monitoring IC is (or will lie to you and say "Yes it has it" even though it may not). -- | Jeremy Chadwick jdc at parodius.com | | Parodius Networking http://www.parodius.com/ | | UNIX Systems Administrator Mountain View, CA, USA | | Making life hard for others since 1977. PGP: 4BD6C0CB |
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080803134328.GA65256>