cpu overheating

Les hlhowell at pacbell.net
Tue Dec 12 04:03:29 UTC 2006


On Mon, 2006-12-11 at 20:15 -0600, Mike McCarty wrote:
> Ed Greshko wrote:
> > Mike Chalmers wrote:
> > 
> > 
> >>I like Linux, alot. I mean alot. I like what it stands for (besides
> >>the corporations). But my CPU has never overheated. I am pretty sure
> >>that it is not my hardware. It could be a bug in Linux. The kernel
> >>could be be sending incorrect frequencies to the hardware or something
> >>like that.
> > 
> > 
> > Hummmm....  Something like that happened to me years ago...  Let me think.
> > Oh, right.....
> 
> [snip]
> 
> It is a well-known fact that faulty software can overheat CPUs.
[snip]

I do believe that I can cause some processors to overheat, however, that
software would not do anything useful.  Once you start accessing memory,
dealing with page faults and taking interrupts in the "real" environment
that type of cycle intensive processing sort of goes out the window.
Yes, processors are heat sensitive, and can fail due to heat, but the
new coolers are very good, and the testing process ensures that the weak
ones probably don't make it into the field.  Believe me when I say the
tests, which bypass many of the protections built in, stress the
processors quite heavily, I do know what I am talking about.  A modern
processor can sustain a hundred amps or so of instantaneous current,
which is made up by the multiple bypassing, board construction, path
design and modern power supply design.  In the test environment, they
are not as fully cooled and protected as they are in your PC, so if they
fail, that is where and when.  You can check out some of the board
designs on Teradyne's public materials (you don't get details, but you
can see the complexity of connecting to over 1000 pins via a zero
insertion force socket that must operate thousands of times reliably in
a factory floor.  

	The GIMP testing program I would suspect verifies such things as look
ahead, que length call/return overhead and so forth to give a good
"frames per second presentation" vs a temperature test of the processor.
Flops (floating point operations per second) is the processing number
important to graphics processing, and the hardware design will determine
such things as how to best optimize the algorithms for best performance.
That is one aspect of real time benchmarking.  I can write the same 3
level nested loop at least 9 different ways.  Some ways are most
efficient on AMD systems, some are better on intel, and some fit power
pc better.  It depends on lots of things outside the processor as well.
Memory speed, cache size and clock frequency to name some of the better
known ones.

	As to the statement that faulty software could cause it, perhaps, but
it would be a real fluke, because as I said lots of things get in the
way in a real normal system.  Such a software bug would have to disable
lots of hardware besides the processor to create that kind of havoc.

Maybe I should write a book....

Regards,
Les H




More information about the users mailing list