Re: Intel Pentium FDIV bug consequences

[RUMI Szabolcs <rumi_ml%rtfm.hu@localhost>]

On Thu, 7 Jan 2010 12:50:14 -0600
David Young <dyoung%pobox.com@localhost> wrote:

> On Thu, Jan 07, 2010 at 10:46:37AM +0100, RUMI Szabolcs wrote:
> > What I'm concerned about is that could this possibly affect
> > the operation of compute-intensive crypto code like OpenSSL
> > and similar in the form of some sporadic corruption in the
> > data transferred over SSL for example?
> 
> Are you seeing sporadic data corruption?

No, so far I'm not.

I just wanted to try to assess the probability of such issues
proactively on a theoretical level because I learned about the
presence of this bug and I wasn't sure about it's real-world
impact. Some years ago I had a router which corrupted packets
sporadically and it was no fun at all, it took months to notice
and days to diagnose where some files got corrupted...

Thanks,
Sab