Re: [PATCH for-4.5 6/8] xen/arm: IRQ: Add lock contrainst for gic_irq_{startup, shutdown}

[Julien Grall <julien.grall@linaro.org>]

On 02/21/2014 08:55 AM, Jan Beulich wrote:
>>>> On 20.02.14 at 21:48, Julien Grall <julien.grall@linaro.org> wrote:
>> Before the clean up there was 8 distinct startup routines for x86. No
>> there is only 2:
>>   - drivers/passthrough/amd/iommu_init.c: iommu_maskable_msi_startup
>>   - arch/x86/ioapic.c: startup_edge_ioapic_irq
>>
>> For a latter one, I'm a bit surprised that the function can return 1,
>> but the result is never used.
> 
> Which means consumption of the return value was intended, but
> never implemented (or lost _very_ long ago). Looking at the Linux
> code, the intention apparently would be for the non-zero return
> value to propagate into IRQ_PENDING in one very special case we
> didn't ever support (auto-probing). Re-sending of an already
> pending interrupt is being handled differently there anyway. So if
> needed something like the setting of IRQ_PENDING at some point,
> I guess we could as well have the startup routine do this itself. I.e.
> I think converting the return value to void is still fine, as long as
> you leave some commentary in
> arch/x86/ioapic.c:startup_edge_ioapic_irq().

I will send the patch to change startup prototype separately later.

-- 
Julien Grall