On Thu, 3 Jan 2002, Mikael Pettersson wrote:
> > Also note that's really an APIC and not a kernel bug -- writing a
> >previously read value to a register is defined as valid by Intel.
>
> Can you quote chapter and verse on that last statement?
> (Preferably from IA32 Vol3.) Writes to APIC registers obviously
> _do_ cause side-effects in some cases...
Obviously I confused defined bits with reserved ones, sorry. Still the
APIC looks insane for me -- it should really signal an error only once an
interrupt is received, like it does for interrupts from remote sources.
> Pentium erratum 3AP can be ignored, since the whole purpose of the
> two ESR accesses is to clear ESR. Unfortunately, Pentium erratum 11AP
> does apply, since there are back-to-back APIC writes (first to LVTERR,
> then to ESR).
I checked 3AP and you are right, since we don't need the ESR's value.
> + if (maxlvt > 3) /* Due to Pentium errata 3AP and 11AP. */
> + apic_write(APIC_ESR, 0);
Use apic_write_around() instead as the 11AP workaround -- it was
introduced specifically for this purpose. Using anything else doesn't
guarantee no back-to-back APIC writes due to interrupts (specifically
writes to the EOI register).
-- + Maciej W. Rozycki, Technical University of Gdansk, Poland + +--------------------------------------------------------------+ + e-mail: macro@ds2.pg.gda.pl, PGP key available +- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Mon Jan 07 2002 - 21:00:24 EST