Ed,

For me, the USB booting was the win-factor to get a 6501.
Ironically I've only used it once, as network booting seems more convenient.

Now to the point - you seem to be right:
From 
'http://www.openbsd.org/cgi-bin/cvsweb/~checkout~/src/sys/arch/i386/i386/machdep.c?rev=1.533;content-type=text%2Fplain':

#if !defined(SMALL_KERNEL)
/*
* Temperature read on the CPU is relative to the maximum
* temperature supported by the CPU, Tj(Max).
* Poorly documented, refer to:
* http://softwarecommunity.intel.com/isn/Community/
* en-US/forums/thread/30228638.aspx
* Basically, depending on a bit in one msr, the max is either 85 or 100.
* Then we subtract the temperature portion of thermal status from
* max to get current temperature.
*/
void
intelcore_update_sensor(void *args)
{
struct cpu_info *ci = (struct cpu_info *) args;
u_int64_t msr;
int max = 100;

/* Only some Core family chips have MSR_TEMPERATURE_TARGET. */
if (ci->ci_model == 0xe &&
   (rdmsr(MSR_TEMPERATURE_TARGET) & MSR_TEMPERATURE_TARGET_LOW_BIT))
max = 85;

The link to the Intel forums is no longer valid, though it's obvious
we're getting max set to 100 here (as we're not running *core-series).
So we're actually suffering from the same problem too.

I'm not sure I can suggest a quality diff to the OpenBSD folks to
correctly detect E640s (hence set max to 90), but I am going to fix it
for my system.

Thanks for bringing that to my attention.

Sent from my Samsung Galaxy Note 3 LTE

On Mar 17, 2014 6:35 PM, "ED Fochler" <[email protected]> wrote:
>
> Nikola,
>
>         No offense taken, ever.  You seem a decent fellow.  As for 
> temperatures, I still think you’re +10C from reality.  Below is output from 
> my two identical 6501’s at 1GHz.  The FreeBSD sysctl output is a lot more 
> verbose, and allows you to see the assumed tjmax.  Temperatures are read from 
> the CPU as an offset from tjmax, and everybody seems to be assuming 
> tjmax=100, even the BSD’s.  For the 6501 and the E640 cpu with tjmax=90, you 
> just have a 10 off problem.
>
> Yay USB booting!  That was an easy comparison.
>
> OpenBSD 5.4:
> # sysctl hw
> hw.machine=i386
> hw.model=Genuine Intel(R) CPU @ 1.00GHz ("GenuineIntel" 686-class)
> hw.ncpu=1
> hw.byteorder=1234
> hw.pagesize=4096
> hw.disknames=sd0:,sd1:8da019107644f3a2
> hw.diskcount=2
> hw.sensors.cpu0.temp0=72.00 degC
> hw.cpuspeed=1001
> hw.setperf=100
>
> PFSense 2.1: (FreeBSD 8.3)
> > sysctl dev.cpu
> dev.cpu.0.%driver: cpu
> dev.cpu.0.%parent: legacy0
> dev.cpu.0.coretemp.delta: 28
> dev.cpu.0.coretemp.resolution: 1
> dev.cpu.0.coretemp.tjmax: 100.0C
> dev.cpu.0.coretemp.throttle_log: 0
> dev.cpu.0.temperature: 72.0C
> dev.cpu.1.%driver: cpu
> dev.cpu.1.%parent: legacy0
> dev.cpu.1.coretemp.delta: 28
> dev.cpu.1.coretemp.resolution: 1
> dev.cpu.1.coretemp.tjmax: 100.0C
> dev.cpu.1.coretemp.throttle_log: 0
> dev.cpu.1.temperature: 72.0C
>
>
> On 2014, Mar 16, at 8:31 PM, Nikola Gyurov <[email protected]> wrote:
>
> > I meant no disrespect, apologies if you felt offended.
> >
> > I'm also using OpenBSD, not Linux.
> > Best regards,
> > Nikola Gyurov
> >
> >
> > On Mon, Mar 17, 2014 at 12:18 AM, ED Fochler <[email protected]> 
> > wrote:
> >> I apologize for being unclear.  I agree that your TJ-Max is 90.  I think 
> >> your sysctl measurement sounds high, as if sysctl is assuming a tjmax off 
> >> 100 when calculating the offset.
> >>
> >> http://lists.soekris.com/pipermail/soekris-tech/2012-January/018121.html
> >>
> >> Or perhaps you are smarter than me, and I am being foolish.  :-)
> >>
> >>        ED.
> >>
> >>
> >>
> >> On 2014, Mar 16, at 8:10 PM, Nikola Gyurov <[email protected]> wrote:
> >>
> >>> ED, according to Intel, the tjmax for the 1 GHz Atom is 90 C (speaking
> >>> about the E640 here)?
> >>> There's also the 640T with tjmax of 110 C.
> >>>
> >>> http://www.intel.com/content/www/us/en/intelligent-systems/queens-bay/embedded-intel-atom-e6xx-series-with-intel-platform-controller-hub-eg20t.html
> >>>
> >>> But how does that affect my sysctl results?
> >>>
> >>> P.S.: I'm using the box, not the rackmount unit.
> >>> _______________________________________________
> >>> Soekris-tech mailing list
> >>> [email protected]
> >>> http://lists.soekris.com/mailman/listinfo/soekris-tech
> >>
> > _______________________________________________
> > Soekris-tech mailing list
> > [email protected]
> > http://lists.soekris.com/mailman/listinfo/soekris-tech
>
_______________________________________________
Soekris-tech mailing list
[email protected]
http://lists.soekris.com/mailman/listinfo/soekris-tech

Reply via email to