Pentium bug and Solaris 2.5.1 SMP

Hi,

some days before it was a discussion about Pentium bug
(lock cmpxchg8b qword ptrax -  F0 0F C7 C8) which hangs the machine.

I've tested it on several configurations and it is really very bad.

But what I'd like to know is why it happens on 2xPentium Solaris 2.5.1

I'm not a specialist in Solaris kernel and multitasking, but I have a
question which may be stupid.

If I execute the command I hang one of my two processors. Why the system
can't work on the rest one performing normal (or emergency) shutdown or
notifying sysadmin about the problem? Am I right that such a command is
executed only on one processor or not?

If yes then is it a bug in Solaris OS?

Is it a problem in Solaris or it is the problem of the SMP architecture on Intel platform (bus lock, etc.)?

Thanks in advance

Vitali

korostelAsked:
Who is Participating?
I wear a lot of hats...

"The solutions and answers provided on Experts Exchange have been extremely helpful to me over the last few years. I wear a lot of hats - Developer, Database Administrator, Help Desk, etc., so I know a lot of things but not a lot about one thing. Experts Exchange gives me answers from people who do know a lot about one thing, in a easy to use platform." -Todd S.

BelarCommented:
Well, to answer your first question. It happens on all pentium or pentium MMX. Pentium Pro and Pentium II are unaffected. For example it crashes my linux SMP so it is not OS related.

About your second answer, yes, indeed only one of the CPU is halted. It might be possible with the appropriate code for the second CPU to "reboot" the first. I actually saw a discussion about this exact issue in the linux-kernel mailing list. The only problem with this "fix" is that if an attacker just do : crash&crash&. It will hang both CPUs without giving time to them to reboot their peer. As for why, there is no notification that one of your cpu is halted. It is simple. I think Solaris or any other OS vendors didn't think it was possible to do such an event.

3) Solaris is not buggy. It is Intel that screwed up.

For more info on the bug go at www.x86.org (or is it com?) The exploit is pretty well documented there.
0

Experts Exchange Solution brought to you by

Your issues matter to us.

Facing a tech roadblock? Get the help and guidance you need from experienced professionals who care. Ask your question anytime, anywhere, with no hassle.

Start your 7-day free trial
It's more than this solution.Get answers and train to solve all your tech problems - anytime, anywhere.Try it for free Edge Out The Competitionfor your dream job with proven skills and certifications.Get started today Stand Outas the employee with proven skills.Start learning today for free Move Your Career Forwardwith certification training in the latest technologies.Start your trial today
Unix OS

From novice to tech pro — start learning today.