Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 5 Feb 2010 14:34:48 +0100 (CET)
From:      Oliver Fromme <olli@lurza.secnetix.de>
To:        freebsd-stable@FreeBSD.ORG
Subject:   amdtemp(4) oddities, Athlon 64 X2 (8-stable)
Message-ID:  <201002051334.o15DYm0a070827@lurza.secnetix.de>

next in thread | raw e-mail | index | archive | help
This is a RELENG_8 box, csupped on 2010-01-22, kernel built
with amdtemp(4).  It's a dual-core Athlon 64, running i386
(32bit) SMP.  Excerpt from dmesg:

CPU: AMD Athlon(tm) 64 X2 Dual Core Processor 3800+ (2009.28-MHz 686-class CPU)
  Origin = "AuthenticAMD"  Id = 0x40fb2  Stepping = 2
  Features=0x178bfbff<FPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,CLFLUSH,MMX,FXSR,SSE,SSE2,HTT>
  Features2=0x2001<SSE3,CX16>
  AMD Features=0xea500800<SYSCALL,NX,MMX+,FFXSR,RDTSCP,LM,3DNow!+,3DNow!>
  AMD Features2=0x1f<LAHF,CMP,SVM,ExtAPIC,CR8>
[...]
FreeBSD/SMP: Multiprocessor System Detected: 2 CPUs
FreeBSD/SMP: 1 package(s) x 2 core(s)
[...]
amdtemp0: <AMD K8 Thermal Sensors> on hostb3

sysctl displays the following values (idle system):

dev.cpu.0.temperature: 14.0C
dev.cpu.1.temperature: 14.0C
dev.amdtemp.0.sensor0.core0: 14.0C
dev.amdtemp.0.sensor0.core1: 22.0C
dev.amdtemp.0.sensor1.core0: -49.0C
dev.amdtemp.0.sensor1.core1: -49.0C
hw.acpi.thermal.tz0.temperature: 40.0C

Upon consecutive calls, the first three vary from 13.0C to
16.0C, which is below room temperature, so it can't be true.

The fourth one varies from 22.0C to 23.0C, which might be
correct, although it still seems a bit low, even on an idle
system.

Under load (buildworld -j4), it looks like this:

dev.cpu.0.temperature: 38.0C
dev.cpu.1.temperature: 39.0C
dev.amdtemp.0.sensor0.core0: 38.0C
dev.amdtemp.0.sensor0.core1: 44.0C
dev.amdtemp.0.sensor1.core0: -49.0C
dev.amdtemp.0.sensor1.core1: -49.0C
hw.acpi.thermal.tz0.temperature: 40.0C

That's the maximum readings I've seen:  The first three go
up to 39.0C, the fourth one reaches 44.0C (I've never seen
values higher than this).  The first three are always about
the same (+/-1), and the fourth is always 4 to 6 more.

When buildworld -j4 is finished, the values quickly go down
and reach the ones given for the idle system above after a
few minutes.  BTW, the machine is running powerd(8) which
reduces the CPU clock from 2000 to 1000 when it is idle.

The sensor1.* values are always fixed at -49.0C, and the
ACPI temperature stays at exactly 40.0C.  These three don't
seem to be connected to any actual sensors.  I'm not too
worried about those.

I've got two questions:

First, I'm wondering why sensor0.core1 is different from all
the other values.  I think it should be the same as the
dev.cpu.1.temperature value.  This smells like a bug.
According to the description in the manual page, I would
expect it to be the same.

Second, the temperature readings on the idle system are too
low.  I don't think they can realistically be below room
temperature.

Best regards
   Oliver

-- 
Oliver Fromme, secnetix GmbH & Co. KG, Marktplatz 29, 85567 Grafing b. M.
Handelsregister: Registergericht Muenchen, HRA 74606,  Geschäftsfuehrung:
secnetix Verwaltungsgesellsch. mbH, Handelsregister: Registergericht Mün-
chen, HRB 125758,  Geschäftsführer: Maik Bachmann, Olaf Erb, Ralf Gebhart

FreeBSD-Dienstleistungen, -Produkte und mehr:  http://www.secnetix.de/bsd

With Perl you can manipulate text, interact with programs, talk over
networks, drive Web pages, perform arbitrary precision arithmetic,
and write programs that look like Snoopy swearing.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?201002051334.o15DYm0a070827>