4 Replies Latest reply on Dec 3, 2012 10:48 AM by enriqueb99

    divide-by-zero floating point exception

    enriqueb99

      Greetings...

       

      This is a question directly for Intel Support tech.

       

      Over the last 1.5 week, I've been tracking some weird divide-by-zero debug behavior I was getting on a new HP machine.  To make a very very very long story short... after numerous steps taken to troubleshoot the problem, I finally concluded yesterday that this was actually a hardware, BIOS or firmware generic issue for the specific machine series I had.  Finally... one thing led to another and I came across this intel document:

      http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/2nd-gen-core-desktop-specification-update.pdf

       

      errata BJ1

       

      My CPU is an Intel i5-2500.  As the documentation states, "Intel has not observed this erratum with any commercially available software or system."  well... we just did.  The software we use is a fairly common software in the Nuclear Simulator World and that's where the problem showed up.  So I suspect this problem could affect many nuclear simulators in the country (USA).  Are there any plans on fixing this, one way or another?  We were planning to purchase 3 more identical computers before the end of the year... but this bug is now putting a stop to this.  We need to know what can be done about this, if anything, before we proceed.

       

      I can provide more details... but since this problem is already identified within that linked documentation... I'm assuming Intel's engineers will know what I'm talking about.

       

      Thanks.