Date: Sun, 24 Apr 2011 17:48:18 -0700 From: Jeremy Chadwick <freebsd@jdc.parodius.com> To: Bartosz Fabianowski <freebsd@chillt.de> Cc: freebsd-stable@freebsd.org, John <john@theusgroup.com> Subject: Re: System extremely slow under light load Message-ID: <20110425004818.GA22579@icarus.home.lan> In-Reply-To: <4DB46ED4.2010500@chillt.de> References: <4DA596D3.1090803@chillt.de> <op.vt1efdn68527sy@pinky> <BANLkTik5Jq1QP776xQ0zQvQ5MKYe4LQZUA@mail.gmail.com> <4DB44DA3.5060509@chillt.de> <4DB4589B.2020909@ksu.ru> <4DB45D6C.20203@chillt.de> <20110424182456.9DD03589@server.theusgroup.com> <4DB46ED4.2010500@chillt.de>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Apr 24, 2011 at 08:41:24PM +0200, Bartosz Fabianowski wrote: > >I don't know which i7 you have, but the intel datasheet for the i7-870 states > >that the maximum case temperature is 72.7C. > > I have a Core i7 Q740 with a native speed of 1.73GHz. My previous > Dell had a Q730. Both were exhibiting the same problems. > > Since this is a laptop, I would expect temperatures to be higher > than in a desktop box. So up to 50-60??C CPU temperature while > idling and 90??C under load may be acceptable. But the behavior the > computer is exhibiting definitely is not acceptable. The temperatures you reported in your earlier post: http://lists.freebsd.org/pipermail/freebsd-stable/2011-April/062377.html Are not normal, nor are they acceptable, even for a laptop. Desktop i7 boxes tend to run around 45C idle, 60-65C load with stock cooling. Laptops should be higher, but not 60-65C idle with 90C under load. Others have already stated what the thermal cut-off point is. As the processor gets hotter, internal clocks and so on are throttled within the hardware to try and stabilise the temperature (to keep the thermal trip point being reached, re: "emergency shutdown"), which greatly decreases performance. I'm not sure if there's a way to detect this, but I would hope (?) that it would be visible via the CPU clock frequency (on FreeBSD this would be sysctl dev.cpu.X.freq). If you were running Windows there would be a multitude of tools you could check to confirm this behaviour (Core Temp, CPU-Z, RMClock, etc.). If you boot into another operating system such as Linux or Windows, do you see the same overall behaviour? Linux might be easier and might have some built-in way to get at CPU temperatures (via /proc?). Trying to figure out if this is a FreeBSD issue or not is difficult. Can you please provide: - Contents of rc.conf - sysctl -a hw.acpi - sysctl -a dev.cpu - sysctl -a dev.est - sysctl -a dev.cpufreq - sysctl -a kern.timecounter Finally, just as a data point -- and this should be honoured no matter what -- there have been cases where manufacturers have incorrectly been applying thermal grease (if used rather than a TIM pad): http://gizmodo.com/#!171394/thermal-greasy-apple-sics-lawyers-on-something-awful So don't think necessarily that just because Dell replaced the entire laptop that the next one wouldn't behave the same way. This is why I recommend trying out another OS to see if it exhibits the problem. -- | Jeremy Chadwick jdc@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?20110425004818.GA22579>