Re: [Xen-devel] [PATCH v4 23/27] ARM: vITS: handle INV command

[André Przywara <andre.przywara@xxxxxxx>]

On 04/04/17 17:51, Julien Grall wrote:
> Hi Andre,
> 
> On 03/04/17 21:28, Andre Przywara wrote:
>> The INV command instructs the ITS to update the configuration data for
>> a given LPI by re-reading its entry from the property table.
>> We don't need to care so much about the priority value, but enabling
>> or disabling an LPI has some effect: We remove or push virtual LPIs
>> to their VCPUs, also check the virtual pending bit if an LPI gets
>> enabled.
>>
>> Signed-off-by: Andre Przywara <andre.przywara@xxxxxxx>
>> ---
>>  xen/arch/arm/vgic-v3-its.c | 62
>> ++++++++++++++++++++++++++++++++++++++++++++++
>>  1 file changed, 62 insertions(+)
>>
>> diff --git a/xen/arch/arm/vgic-v3-its.c b/xen/arch/arm/vgic-v3-its.c
>> index 2f024b1..920c437 100644
>> --- a/xen/arch/arm/vgic-v3-its.c
>> +++ b/xen/arch/arm/vgic-v3-its.c

[ ... ]

>> +
>> +    if ( *property & LPI_PROP_ENABLED )
>> +    {
>> +        unsigned long flags;
>> +
>> +        set_bit(GIC_IRQ_GUEST_ENABLED, &p->status);
>> +        spin_lock_irqsave(&vcpu->arch.vgic.lock, flags);
>> +        if ( !list_empty(&p->inflight) &&
>> +             !test_bit(GIC_IRQ_GUEST_VISIBLE, &p->status) )
>> +            gic_raise_guest_irq(vcpu, vlpi, p->lpi_priority);
>> +        spin_unlock_irqrestore(&vcpu->arch.vgic.lock, flags);
>> +
>> +        /* Check whether the LPI has fired while the guest had it
>> disabled. */
>> +        if ( test_and_clear_bit(GIC_IRQ_GUEST_LPI_PENDING, &p->status) )
>> +            vgic_vcpu_inject_irq(vcpu, vlpi);
>> +    }
>> +    else
>> +    {
>> +        clear_bit(GIC_IRQ_GUEST_ENABLED, &p->status);
>> +        gic_remove_from_queues(vcpu, vlpi);
>> +    }
> 
> This could looks quite similar to vgic_{enable,disable}_irqs. This is a
> call to refactor the code there.

I tried it and it doesn't work easily, because the part that handles a
single (non-virtual) interrupt calls vgic_get_virq_type(), whose
parameters break the abstraction of a single interrupt.
Given that I am bit wary of bending a core GIC function just for the
sake of refactoring just a second user.

Cheers,
Andre.



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