Date: Tue, 15 May 2007 13:14:53 +0100 From: Rui Paulo <rpaulo@fnop.net> To: Harald Schmalzbauer <h.schmalzbauer@omnisec.de> Cc: freebsd-current@freebsd.org, Colin Percival <cperciva@freebsd.org> Subject: Re: cpufreq & enhanced speedstep Message-ID: <864pmel1vm.wl%rpaulo@fnop.net> In-Reply-To: <200705151351.53363.h.schmalzbauer@omnisec.de> References: <42BC519E.8010309@satro.sk> <200705151247.19122.h.schmalzbauer@omnisec.de> <86646ul54p.wl%rpaulo@fnop.net> <200705151351.53363.h.schmalzbauer@omnisec.de>
next in thread | previous in thread | raw e-mail | index | archive | help
At Tue, 15 May 2007 13:51:52 +0200, Harald Schmalzbauer wrote: > I did (I had to copy src/backlight and src/msrtemp to src/sys/dev/ and=20 > src/sys/modules/). The patch should be applied with CWD =3D=3D src/sys. > Here is what I get when I load the module: > msrtemp0: <CPU On-Die Thermal Sensors> on cpu0 > est1: <Enhanced SpeedStep Frequency Control> on cpu1 > est: CPU supports Enhanced Speedstep, but is not recognized. > est: cpu_vendor GenuineIntel, msr 928092806000928 > device_attach: est1 attach returned 6 > msrtemp1: <CPU On-Die Thermal Sensors> on cpu1 >=20 > But these values are too high, my BIOS tells me ~50=B0C: > dev.cpu.0.temperature: 68 > dev.cpu.1.temperature: 69 Well, the digital readout is relative to maximum Thermal Junction (Tj(max)). So, the calculation is: Tj(max) - digital readout. On some systems Tj(max) is 100, on others is 85, but there's no know way to detect this realiably. Linux does a trick and I'm still investigating whether or not to apply it. It seems that your CPU's Tj(max) is 85 and not 100. 100-32 =3D 68 85-32 =3D 53 > Is there anything I can do to correct the ACPI table for the right power = > values of my CPU? Not likely. As Colin said, it would probably require an NDA to give you all the values. -- Rui Paulo
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?864pmel1vm.wl%rpaulo>