Re: [Xen-devel] [PATCH v3] xen/arm: introduce PLATFORM_QUIRK_GUEST_PIRQ_NEED_EOI

[Anup Patel <apatel@xxxxxxx>]

On Wed, Jul 9, 2014 at 8:41 PM, Stefano Stabellini
<stefano.stabellini@xxxxxxxxxxxxx> wrote:
> On Wed, 9 Jul 2014, Julien Grall wrote:
>> On 07/09/2014 03:40 PM, Ian Campbell wrote:
>> >> I think we should check GIC Dist address passed in DTS by
>> >> u-boot to Xen. If GIC Dist address is secured address then
>> >> enable the quirk otherwise disable the qurik.
>> >
>> > Isn't IGROUPR readable from NS? In which case we should be able to spot
>> > the difference, I think?
>
> Anup, can you confirm that we can use IGROUPR to spot the difference?
> Otherwise is there another register we can use?

Yes, IGROUPR of the secured GIC registers would be correct register
to spot the difference.

More precisely, if IGROUPR0 == 0xFFFFFFFF in GIC secured registers then
don't use the quirk for X-Gene.

>
> In any case given that only you can test it, I think that it should be
> one of you guys that come up with a way to detect the change and submit
> a patch to conditionally disable the workaround.

Sure, most likely Pranav will send a patch for Xen once APM releases
updated u-boot binary.

Thanks,
Anup

>
>
>> > Stefano, you might want to arrange in your patch to cache the value of
>> > the quirk in the gicv2 struct -- otherwise all the calls to
>> > platform_has_quirk are going to add up.
>>
>> Can't we cache the value in the platform code? I don't think it's too
>> expensive to call the platform_has_quirk in this case.
>
> Caching the value in gic-v2.c seems a bit overkill. But we could turn
> platform->quirks() into a bitfield, then we can significantly reduce
> memory accesses when checking for quirks, from 3 to just 1.

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel