Re: [Xen-devel] [PATCH] vt-d: fix wrong addr in IOTLB invalidation descriptor

["Zhang, Yang Z" <yang.z.zhang@xxxxxxxxx>]

Ian Jackson wrote on 2012-07-24:
> Zhang, Yang Z writes ("[Xen-devel] [PATCH] vt-d: fix wrong addr in IOTLB
> invalidation descriptor"):
>> According to vt-d specs, the addr in IOTLB invalidation descriptor
>> should be 4K page aligned.
>> 
>> Signed-off-by: Yang Zhang <yang.z.zhang@xxxxxxxxx>
>> 
>> diff -r 0455d8317631 xen/drivers/passthrough/vtd/qinval.c ---
>> a/xen/drivers/passthrough/vtd/qinval.c      Thu Jun 28 18:43:28 2012
>> +0100 +++ b/xen/drivers/passthrough/vtd/qinval.c      Wed Jul 11
>> 08:43:24 2012 +0800 @@ -140,7 +140,7 @@ static int
>> gen_iotlb_inv_dsc(struct iomm
>>      qinval_entry->q.iotlb_inv_dsc.hi.am = am;
>>      qinval_entry->q.iotlb_inv_dsc.hi.ih = ih;
>>      qinval_entry->q.iotlb_inv_dsc.hi.res_1 = 0;
>> -    qinval_entry->q.iotlb_inv_dsc.hi.addr = addr;
>> +    qinval_entry->q.iotlb_inv_dsc.hi.addr = addr >> PAGE_SHIFT_4K;
>> 
>>      unmap_vtd_domain_page(qinval_entries);
>>      spin_unlock_irqrestore(&qi_ctrl->qinval_lock, flags);
> 
> Does this mean that previously we were invalidating the wrong place
> and therefore our invalidations were ineffective ?  Can this be
> exploited somehow by a malicious guest ?

Yes. All the invalidations were ineffective before this patch. The previous 
release which supporting the QI also need this patch.

No, hypervisor owns it and guest is unaware of it.

Best regards,
Yang



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