Messages in this thread | | | Date | Thu, 16 Dec 2004 15:42:44 +0000 (GMT) | From | "Maciej W. Rozycki" <> | Subject | Re: [patch, 2.6.10-rc3] safe_hlt() & NMIs |
| |
On Thu, 16 Dec 2004, Ingo Molnar wrote:
> The 'sti' "shadows" the cli, i.e. we'll never get an interrupt that gets > inbetween 'sti;cli'. I.e. sti is the black-hole generator, and 'cli' is > in the black hole. In that sense the 'cli' is in a black hole to the > NMI: the NMI will never see cli as the 'next to be executed' > instruction.
That's what I meant indeed, but I'd like to emphasise, for readers to be aware, the black hole is not tied to the 'cli' instruction itself in any way. The black-holed instruction needs not be a 'cli' -- it can be an arbitrary one except from ones creating black holes as you've observed with your test.
Maciej - 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/
| |