[parisc-linux] [PATCH] timer_interrupt and gettimeoffset.

[James Bottomley]

On Sun, 2006-09-03 at 13:38 -0600, Grant Grundler wrote:
> Doesn't this violate one of the basic tenants of linux interrupts?
> 
> I know the interrupt handler (same instance) can't be invoked on
> two different CPUs and it seems re-entrance would be a similar
> case.

re-entrant is the wrong word ... however all linux interrupt handlers
should be interruptible (unless they specifically disable interrupts).
Ours apparently still aren't, so I'll see about fixing them.

The timer interrupt is the exception, actually, because it's set
IRQF_DISABLED which means no interrupts in timer_interrupt() until
they're specifically enabled, which doesn't happen until well into the
softirq processing.

James