Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 28 Aug 2012 03:15:23 +1000 (EST)
From:      Ian Smith <smithi@nimnet.asn.au>
To:        jb <jb.1234abcd@gmail.com>
Cc:        Mike Manilone <crtmike@gmx.us>, freebsd-stable@freebsd.org, Christian Mangin <christian.mangin@gmail.com>
Subject:   Re: Temperature too high when high overload
Message-ID:  <20120828024834.K33776@sola.nimnet.asn.au>
In-Reply-To: <loom.20120827T181757-794@post.gmane.org>
References:  <503B6BCB.10407@gmx.us> <503B87C2.5080308@gmail.com> <loom.20120827T181757-794@post.gmane.org>

next in thread | previous in thread | raw e-mail | index | archive | help
  This message is in MIME format.  The first part should be readable text,
  while the remaining parts are likely unreadable without MIME-aware tools.

--0-204813080-1346087723=:33776
Content-Type: TEXT/PLAIN; charset=ISO-8859-1
Content-Transfer-Encoding: 8BIT

On Mon, 27 Aug 2012 16:37:41 +0000 (UTC), jb wrote:
 > Christian Mangin <christian.mangin <at> gmail.com> writes:
 > > Le 27.08.2012 08:44, Mike Manilone a écrit :
 > > > Hi all,
 > > >
 > > > I just switched from Fedora Linux to FreeBSD. But I noticed a problem,
 > > > the CPU temperature will be very high when the load is high.
 > > > Especially while I am building C++ programs. It shut down for even 3
 > > > times while I was building Firefox/Thunderbird, just because of high
 > > > temperature (86.5C).

Mike, 86.5C isn't really all that hot for a modern 4-core laptop under 
load; like jb below, show us `sysctl -a | grep thermal` so we can see 
its passive cooling and critical temperatures.

 > > I used to have the same problem with my laptop (i5) and this can be
 > > fixed by lowering the temperature threshold for passive cooling. (_PSV)
 > > 
 > > hw.acpi.thermal.user_override=1
 > > hw.acpi.thermal.tz0._PSV=80C
 > > 
 > > You should try to adjust _PSV to be significantly lower (> 15-20C) than
 > > the _CRT (critical shutdown temp) so that _CRT is never reached.
 > > 
 > > Christian

Modulo adjusting the right thermal zone, this is safe advice; you can 
always edge it up later, assuming it helps stay at say 10C below _CRT.

 > I too have the same problem (Lenovo dual core r61i).
 > You should see the relevant data before making any changes - below it is
 > explained why.
 > 
 > This is my data:
 > $ sysctl -a | grep -i thermal
 > hw.acpi.thermal.min_runtime: 0
 > hw.acpi.thermal.polling_rate: 10
 > hw.acpi.thermal.user_override: 0
 > hw.acpi.thermal.tz0.temperature: 42.0C
 > hw.acpi.thermal.tz0.active: -1
 > hw.acpi.thermal.tz0.passive_cooling: 0
 > hw.acpi.thermal.tz0.thermal_flags: 0
 > hw.acpi.thermal.tz0._PSV: -1
 > hw.acpi.thermal.tz0._HOT: -1
 > hw.acpi.thermal.tz0._CRT: 127.0C
 > hw.acpi.thermal.tz0._ACx: -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 > hw.acpi.thermal.tz0._TC1: -1
 > hw.acpi.thermal.tz0._TC2: -1
 > hw.acpi.thermal.tz0._TSP: -1
 > hw.acpi.thermal.tz1.temperature: 42.0C
 > hw.acpi.thermal.tz1.active: -1
 > hw.acpi.thermal.tz1.passive_cooling: 1
 > hw.acpi.thermal.tz1.thermal_flags: 0
 > hw.acpi.thermal.tz1._PSV: 95.5C
 > hw.acpi.thermal.tz1._HOT: -1
 > hw.acpi.thermal.tz1._CRT: 100.0C
 > hw.acpi.thermal.tz1._ACx: -1 -1 -1 -1 -1 -1 -1 -1 -1 -1
 > hw.acpi.thermal.tz1._TC1: 5
 > hw.acpi.thermal.tz1._TC2: 4
 > hw.acpi.thermal.tz1._TSP: 600
 > dev.acpi_tz.0.%desc: Thermal Zone
 > dev.acpi_tz.1.%desc: Thermal Zone
 > dev.p4tcc.0.%desc: CPU Frequency Thermal Control
 > dev.p4tcc.1.%desc: CPU Frequency Thermal Control
 > $
 > 
 > As you can see in my case:
 > hw.acpi.thermal.tz0.passive_cooling: 0
 > which is NOT available (so obviously any settings in tz0 zone are irrelevant).

That tz0 seems not to be a CPU, nor a fan.  Maybe just informational?

 > This is explained here:
 > ACPI_THERMAL(4):
 > ...
 >      hw.acpi.thermal.tz%d.passive_cooling
 >              If set to 1, passive cooling is enabled.  It does cooling without
 >              fans using cpufreq(4) as the mechanism for controlling CPU speed.
 >              Default is enabled for tz0 where it is available.
 > ...
 > 
 > In my case tz1 zone is available and active.

And your _PSV 95.5C and _CRT 100.0C aren't uncommon sort of values these 
days, hence my surprise at Mike's (apparent) CRT shutdown showing 86.5C.

On the other hand, even my 1133MHz P3-M can go from <50C to >60C inside 
one 10-second polling interval under applied high load, so a shorter 
hw.acpi.thermal.polling_rate may help trigger _PSV well before _CRT.

cheers, Ian
--0-204813080-1346087723=:33776--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20120828024834.K33776>