KA620

From: Paul Koning <pkoning_at_equallogic.com>
Date: Tue Mar 23 08:18:12 2004

>>>>> "ben" == ben franchuk <bfranchuk_at_jetnet.ab.ca> writes:

 ben> Antonio Carlini wrote:
>> I never had access to AXE, but as I understand it, it would
>> generate semi-random sequences of instructions and then execute
>> them. Then it would compare the actual results with known good
>> results. I don't know the details of how it would determine "known
>> good results" given that the sequence was semi-random.
>>
>> So you would kick it off and if the machine under test did not
>> fall over in a heap after a few weeks or months of running, then
>> it was probably good enough.

 ben> But it still might never catch something like the Pentium divide
 ben> bug. Anyhow are not most problems with hardware and software
 ben> now dynamic in nature? My latest game freezes when I click on
 ben> blah blah blah after switching video modes during the internet
 ben> software driver download.

Sure, but those are software bugs, not processor instruction
implementation bugs.

Meanwhile, something like AXE is your best bet to catch unexpected
instruction interactions. It was used because it provides FAR better
coverage than conventional CPU diagnostics. Not alone, of course --
you still want to pass regular diagnostics, and run all the
applications you can think of. And the microcode still has to be
carefully designed and coded. But AXE adds yet another level of
assurance to the implementation.

          paul
Received on Tue Mar 23 2004 - 08:18:12 GMT