Re: [Qemu-devel] [PATCH V2 2/3] virtio: destroy region cache during reset

[Jason Wang <jasowang@redhat.com>]

On 2017年03月13日 18:05, Cornelia Huck wrote:
> On Mon, 13 Mar 2017 14:29:42 +0800
> Jason Wang <jasowang@redhat.com> wrote:
>
>> We don't destroy region cache during reset which can make the maps
>> of previous driver leaked to a buggy or malicious driver that don't
>> set vring address before starting to use the device. Fix this by
>> destroy the region cache during reset and validate it before trying to
>> see them.
>>
>> Cc: Cornelia Huck <cornelia.huck@de.ibm.com>
>> Cc: Paolo Bonzini <pbonzini@redhat.com>
>> Signed-off-by: Jason Wang <jasowang@redhat.com>
>> ---
>> Changes from v1:
>> - switch to use rcu in virtio_virtqueue_region_cache()
>> - use unlikely() when needed
>> ---
>>   hw/virtio/virtio.c | 60 +++++++++++++++++++++++++++++++++++++++++++++++-------
>>   1 file changed, 53 insertions(+), 7 deletions(-)
>>
>> diff --git a/hw/virtio/virtio.c b/hw/virtio/virtio.c
>> index 76cc81b..f086452 100644
>> --- a/hw/virtio/virtio.c
>> +++ b/hw/virtio/virtio.c
>> @@ -190,6 +190,10 @@ static inline uint16_t vring_avail_flags(VirtQueue *vq)
>>   {
>>       VRingMemoryRegionCaches *caches = atomic_rcu_read(&vq->vring.caches);
>>       hwaddr pa = offsetof(VRingAvail, flags);
>> +    if (unlikely(!caches)) {
>> +        virtio_error(vq->vdev, "Cannot map avail flags");
>> +        return 0;
> I'm still not 100% convinced of those checks; but they don't do any
> harm.

Right, consider we've already had patch 1, I think it should be fine to 
use assert here.

>
>> +    }
>>       return virtio_lduw_phys_cached(vq->vdev, &caches->avail, pa);
>>   }
>>
> (...)
>
>> +static void virtio_virtqueue_reset_region_cache(struct VirtQueue *vq)
>> +{
>> +    VRingMemoryRegionCaches *caches;
>> +
>> +    caches = atomic_read(&vq->vring.caches);
>> +    atomic_set(&vq->vring.caches, NULL);
> Needs atomic_rcu_set(), I think.

Any atomic write should be fine here, but I agree atomic_rcu_set() is 
better. Will use it in next version.

Thanks

>
>> +    if (caches) {
>> +        call_rcu(caches, virtio_free_region_cache, rcu);
>> +    }
>> +}
>> +
>>   void virtio_reset(void *opaque)
>>   {
>>       VirtIODevice *vdev = opaque;
>