Date: Fri, 7 Nov 1997 13:55:31 -0800 From: Jonathan Ng <jng@creativedesign.com> To: hackers@FreeBSD.ORG, "'Julian Elischer'" <julian@whistle.com> Subject: RE: Newest Pentium bug (fatal) -- exception handling problem? Message-ID: <81EA05D65B95CF11822500A0243D54040103CA@host66.creativedesign.com>
next in thread | raw e-mail | index | archive | help
I've used the same 133MHz Pentium chip to run the experiment.
Under both FreeBSD and Linux, the system halted immediately.
However, under VxWorks, the system stopped the program due to an
exception fault indicating invalid opcode.
We probably need to look at the exception handling before pointing it to
the chip.
- Jonathan Ng
> ----------
> From: Julian Elischer[SMTP:julian@whistle.com]
> Sent: Friday, November 07, 1997 10:39 AM
> To: hackers@FreeBSD.ORG
> Subject: Newest Pentium bug (fatal)
>
> On a "genuine Intel" pentium (not pentium pro)
> execution of the following sequence, 0xf0 0x0f 0xc7 0xc8
>
> will stop the processor. This is doable from user mode and in
> 16bitmode, or in fact any mode.
>
> try the following c program.
>
>
>
>
> unsigned char x[] = { 0xfo, 0x0f, 0xc7, 0xc8 };
>
>
> main ()
> {
> void (*f)(void) = x;
> f();
> }
>
>
>
>
> We've checked:
> K5... OK
> P6... OK
> P5... *SPLAT*
>
> no idea about the pentium II or other pentium copies.
> K6?
>
> other pentium variants?
> versions?
>
> this one DEFINITLY dies:
> CPU: Pentium (99.38-MHz 586-class CPU)
> Origin = "GenuineIntel" Id = 0x525 Stepping=5
> Features=0x1bf<FPU,VME,DE,PSE,TSC,MSR,MCE,CX8>
>
>
> share and enjoy..
>
> julian
>
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?81EA05D65B95CF11822500A0243D54040103CA>