Re: [PATCH v3 2/2] xen/dt: Allow only IRQ translation that are mapped to main GIC

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

On 16/07/14 14:14, Ian Campbell wrote:
> On Wed, 2014-07-09 at 14:23 +0100, Julien Grall wrote:
>> Xen is only able to handle one GIC controller. Some platform may contain
>> other interrupt controller.
>>
>> Make sure to only translate IRQ mapped into the GIC handled by Xen.
>>
>> Signed-off-by: Julien Grall <julien.grall@linaro.org>
>>
>> ---
>>      Changes in v3:
>>          - Add an ASSERT to check that dt_interrupt_controller is not
>>          NULL.
>>
>>      Changes in v2:
>>          - Fix compilation...
>> ---
>>   xen/common/device_tree.c |    5 ++++-
>>   1 file changed, 4 insertions(+), 1 deletion(-)
>>
>> diff --git a/xen/common/device_tree.c b/xen/common/device_tree.c
>> index 310635e..cc45bd1 100644
>> --- a/xen/common/device_tree.c
>> +++ b/xen/common/device_tree.c
>> @@ -1442,9 +1442,12 @@ int dt_irq_translate(const struct dt_raw_irq *raw,
>>                        struct dt_irq *out_irq)
>>   {
>>       ASSERT(dt_irq_xlate != NULL);
>> +    ASSERT(dt_interrupt_controller != NULL);
>>
>> -    /* TODO: Retrieve the right irq_xlate. This is only work for the gic */
>> +    if ( raw->controller != dt_interrupt_controller )
>> +        return -EINVAL;
>>
>> +    /* TODO: Retrieve the right irq_xlate. This is only work for the gic */
>
> "This only works for ...".
>
> Do you mean it to say "primary gic"? In which case I think it was in the
> correct location before (i.e. before the check which enforced that).

Which location are you talking about? The one in map_device?

If so, it doesn't protect anything, but only avoid to assign an IRQ to 
DOM0 which is not routed to the primary interrupt controller.

Checking the controller in the location will prevent platform_get_irq to 
translate IRQ not handled by the gic controller. Futhermore, it will 
avoid the new hypercall to list interrupt for a specific device node 
returning an invalid IRQ number.

Regards,

-- 
Julien Grall