>On Tue, 16 May 2006 08:15:48 +0200 [EMAIL PROTECTED] wrote: > >[snip] >> If the system dies with ipfilter in the stack trace, it's ipfilter >> which is to blame; blaming it on bad hardware is ludicrous. >[snip] > >Some hardware bugs may be provocated by software. For examle, Pentium's >F00F bug.
Bugs like that are not triggered by the kernel as such instruction sequences, when discovered, are purposely avoided by compilers and in hand crafted assembler. The FOOF bug could not be triggered accidentily: it required the use of a handcrafted invalid instruction. But I was actually thinking of "hardware issues" (broken hardware), not hardware bugs. FOOF, of course, did not cause a panic. It caused a hard hang. The Solaris kernel has implemented the FOOF workaround since Solaris 7 (with patches for earlier releases) These hardware bugs are completely predictable. Casper
