On Sun, 13 Mar 2016, Jianyu Zhan wrote:
> On Sun, Mar 13, 2016 at 5:11 PM, Thomas Gleixner <[email protected]> wrote:
> > On Sun, 13 Mar 2016, Jianyu Zhan wrote:
> >> On Sun, Mar 13, 2016 at 3:55 PM, Thomas Gleixner <[email protected]>
> >> wrote:
> >> > if LOCAL_APIC is disabled it does not use the interrupt, simply because
> >> > there
> >> > is no way to trigger it. That setup is inside #ifdef
> >> > CONFIG_X86_LOCAL_APIC for
> >> > exactly that reason.
> >> >
> >> > Just because IRQ_WORK has no config dependency on LOCAL APIC that does
> >> > not
> >> > mean it uses the interrupt gate unconditionally.
> >> >
> >>
> >> Thanks for clarification.
> >>
> >> I think IRQ_WORK works as generic hardirq context callbacks, it should
> >> reply on
> >> self IPI, which is a functionality provided by LOCAL_APIC, while
> >> legacy PIC doesn't
> >> provide this(correct?).
> >
> > As I said before IRQ_WORK can work w/o APIC. And therefor IRQ_WORK does NOT
> > depend on APIC.
> >
> > End of story. Nothing to change here at all.
>
> If so, then it is weird, because in current code, IRQ_WORK vector init
> is surrounded by
> CONFIG_X86_LOCAL_APIC. And actually my patch did move it out.
Sigh. Do you actaully read what I write?
IRQ_WORK can work w/o APIC
Emphasis on CAN. If the APIC is available it's used, if not then there is no
point in setting up the gate for nothing.
So why would your patch do any good?
Thanks,
tglx
