On Mon, 27 Mar 2000, Richard Henderson wrote:
>No, it is true that the cpus that did not handle the interrupt did
>waste time. There's no denying that.[..]
In theory with level triggered irqs (that's the case) they may not ever
notice the interrupt at all and not to waste any time. The cpu may be
running with irq disabled and the level irq can be disabled by the other
cpu before the cpu raise the ipl.
>I didn't do anything in userspace. It's all in the kernel.
That's fine.
Andrea
This archive was generated by hypermail 2b29 : Fri Mar 31 2000 - 21:00:32 EST