On Thu, 5 Nov 1998, Alan Cox wrote:
> > > Believe it or not, there are a few (ahem) folks running SMP systems in
> > > production, and the 2.0 kernels may not be the cat's meow, but they
> >
> > any 2.0 SMP box that is not strictly cpu-bound is basically crippled.
>
> Thats a definite overstatement with 2.0.3x
I would change the wording slightly. I would say "any SMP machine that
has more than *two* processors is not going to work reliably on 2.0.3x...
based on lots of tests here. 2.1's work perfectly on such platforms,
but with more than 2 cpus, I get nothing but IRQ deadlocks on any kind of
I/O activity. Linus has said many times that 2.0 is broken for more than
two cpus... experience shows that to be true.
I ran 2.0 on a dual for months with zero glitches.. but not on any
quads...
>
> > > Besides it's hard to brag about months of uptime if you're rebooting to
> > > install new kernels. :-)
> >
> > this doesn't imply that you must run each one. can anyone produce evidence
> > that no 2.1 is as stable as 2.0? 2.1 is the product of over 2 years of
> > concerted effort, most of which will never benefit anyone running 2.0.
> >
> > remember, 2.0 is basically bug-fixes since June 6, 1996.
>
> The SMP bug fixes included IRQ forwarding which in some cases tripled the
> scalability of non CPU bound machines.
>
> Alan
>
