Message-ID: <e2141f6a-c530-46d5-d5d9-26806b02d55b@redhat.com>
Date:   Tue, 29 Oct 2019 13:49:10 +0100
From:   Auger Eric <eric.auger@...hat.com>
To:     Zenghui Yu <yuzenghui@...wei.com>, Marc Zyngier <maz@...nel.org>
Cc:     suzuki.poulose@....com, linux-kernel@...r.kernel.org,
        james.morse@....com, julien.thierry.kdev@...il.com,
        wanghaibin.wang@...wei.com, kvmarm@...ts.cs.columbia.edu,
        linux-arm-kernel@...ts.infradead.org
Subject: Re: [PATCH 3/3] KVM: arm/arm64: vgic: Don't rely on the wrong pending
 table

Hi Zenghui,

On 10/29/19 1:27 PM, Zenghui Yu wrote:
> Hi Marc,
> 
> On 2019/10/29 17:23, Marc Zyngier wrote:
>> On Tue, 29 Oct 2019 07:19:19 +0000,
>> Zenghui Yu <yuzenghui@...wei.com> wrote:
>>>
>>> It's possible that two LPIs locate in the same "byte_offset" but target
>>> two different vcpus, where their pending status are indicated by two
>>> different pending tables.  In such a scenario, using last_byte_offset
>>> optimization will lead KVM relying on the wrong pending table entry.
>>> Let us use last_ptr instead, which can be treated as a byte index into
>>> a pending table and also, can be vcpu specific.
>>>
>>> Signed-off-by: Zenghui Yu <yuzenghui@...wei.com>
>>> ---
>>>
>>> If this patch has done the right thing, we can even add the:
>>>
>>> Fixes: 280771252c1b ("KVM: arm64: vgic-v3:
>>> KVM_DEV_ARM_VGIC_SAVE_PENDING_TABLES")
>>>
>>> But to be honest, I'm not clear about what has this patch actually
>>> fixed.
>>> Pending tables should contain all zeros before we flush vgic_irq's
>>> pending
>>> status into guest's RAM (thinking that guest should never write anything
>>> into it). So the pending table entry we've read from the guest memory
>>> seems always be zero. And we will always do the right thing even if we
>>> rely on the wrong pending table entry.
>>>
>>> I think I must have some misunderstanding here... Please fix me.
>>
>> I think you're spot on, and it is the code needs fixing, not you! The
>> problem is that we only read a byte once, irrespective of the vcpu the
>> interrupts is routed to. If we switch to another vcpu for the same
>> byte offset, we must reload it.
>>
>> This can be done by either checking the vcpu, or by tracking the guest
>> address that we read from (just like you do here).
> 
> okay, the remaining question is that in vgic_v3_save_pending_tables():
> 
>     stored = val & (1U << bit_nr);
>     if (stored == irq->pending_latch)
>         continue;
> 
>     if (irq->pending_latch)
>         val |= 1 << bit_nr;
>     else
>         val &= ~(1 << bit_nr);
> 
> Do we really have a scenario where irq->pending_latch==false and
> stored==true (corresponds to the above "else") and then we clear
> pending status of this LPI in guest memory?
> I can not think out one now.

if you save, restore and save again. On the 1st save the LPI may be
pending, it gets stored. On the second save the LPI may be not pending
anymore?

Thanks

Eric
> 
>>
>> A small comment below:
>>
>>>   virt/kvm/arm/vgic/vgic-v3.c | 6 +++---
>>>   1 file changed, 3 insertions(+), 3 deletions(-)
>>>
>>> diff --git a/virt/kvm/arm/vgic/vgic-v3.c b/virt/kvm/arm/vgic/vgic-v3.c
>>> index 5ef93e5041e1..7cd2e2f81513 100644
>>> --- a/virt/kvm/arm/vgic/vgic-v3.c
>>> +++ b/virt/kvm/arm/vgic/vgic-v3.c
>>> @@ -363,8 +363,8 @@ int vgic_v3_lpi_sync_pending_status(struct kvm
>>> *kvm, struct vgic_irq *irq)
>>>   int vgic_v3_save_pending_tables(struct kvm *kvm)
>>>   {
>>>       struct vgic_dist *dist = &kvm->arch.vgic;
>>> -    int last_byte_offset = -1;
>>>       struct vgic_irq *irq;
>>> +    gpa_t last_ptr = -1;
>>
>> This should be written as
>>
>>       gpa_t last_ptr = ~(gpa_t)0;
> 
> Thanks for pointing it out.
> 
>>
>>>       int ret;
>>>       u8 val;
>>>   @@ -384,11 +384,11 @@ int vgic_v3_save_pending_tables(struct kvm *kvm)
>>>           bit_nr = irq->intid % BITS_PER_BYTE;
>>>           ptr = pendbase + byte_offset;
>>>   -        if (byte_offset != last_byte_offset) {
>>> +        if (ptr != last_ptr) {
>>>               ret = kvm_read_guest_lock(kvm, ptr, &val, 1);
>>>               if (ret)
>>>                   return ret;
>>> -            last_byte_offset = byte_offset;
>>> +            last_ptr = ptr;
>>>           }
>>>             stored = val & (1U << bit_nr);
>>
>> Otherwise, this looks good to me (no need to respin for the above
>> nit).
> 
> Thanks,
> 
> Zenghui
> 
> 
> _______________________________________________
> linux-arm-kernel mailing list
> linux-arm-kernel@...ts.infradead.org
> http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

Powered by blists - more mailing lists