I sent Tyan a fairly comprehensive explanation of the problems with the Tomcat
boards, along with a request for support. However, I have yet to receive even
an acknowledgement of my mail.
This was, remember, the sixth time I'd contacted them on the subject. They
don't seem to be particularly worried about the fact that their boards are
broken.
These boards do not comply with the Intel MP spec. As far as I'm concerned,
they're not fit for the purpose for which they were sold, and as such, under
British Law at least, I'm entitled to a replacement or a refund. I suspect
that similar legislation is in force just about everywhere.
Penalties for false advertisement are also a possibility. Speak to your dealer
and/or local Small Claims Court or equivalent.
If I hear from them, I'll be sure to keep you posted.
>
> On Sat, 31 Jan 1998, MOLNAR Ingo wrote:
>
> > this patch brings IO-APIC IRQ handling in sync with the changed IRQ model,
> > and enables IO-APIC IRQs in some very basic & safe situations:
> >
> > - when the IO-APIC has no PCI pins connected
> > - when the user has explicitly specified a pirq= line
> >
> > i've tested these changes with very heavy parallel networking and disk IO
> > load, and saw not a single problem. The UP and 486 case is not tested yet
> > (it's untouched).
> >
> > there is no black/whitelist yet, will add them ASAP.
> >
> > the patch also includes 8390.c changes, with them i got rid of those
> > annoying 'reentering ...' messages.
> >
> > the pt_regs solution in disable_irq() is quite ugly, is there any better
> > solution to 'simulate' interrupts from syscall level?
>
>
>
--
---- ---- ----
David Woodhouse, Robinson College, CB3 9AN, England. (+44) 0976 658355
Dave@imladris.demon.co.uk http://dwmw2.robinson.cam.ac.uk
finger pgp@dwmw2.robinson.cam.ac.uk for PGP key.