Re: [PATCH v4 4/6] vfio/type1: check dma map request is within a valid iova range

From: Auger Eric
Date: Wed Feb 28 2018 - 06:53:38 EST


Hi Shameer,

On 28/02/18 10:25, Shameerali Kolothum Thodi wrote:
>
>
>> -----Original Message-----
>> From: Auger Eric [mailto:eric.auger@xxxxxxxxxx]
>> Sent: Wednesday, February 28, 2018 9:02 AM
>> To: Shameerali Kolothum Thodi <shameerali.kolothum.thodi@xxxxxxxxxx>;
>> Alex Williamson <alex.williamson@xxxxxxxxxx>
>> Cc: pmorel@xxxxxxxxxxxxxxxxxx; kvm@xxxxxxxxxxxxxxx; linux-
>> kernel@xxxxxxxxxxxxxxx; Linuxarm <linuxarm@xxxxxxxxxx>; John Garry
>> <john.garry@xxxxxxxxxx>; xuwei (O) <xuwei5@xxxxxxxxxx>; Robin Murphy
>> <robin.murphy@xxxxxxx>
>> Subject: Re: [PATCH v4 4/6] vfio/type1: check dma map request is within a valid
>> iova range
>>
>> Hi Shameer,
>>
>> On 27/02/18 10:57, Shameerali Kolothum Thodi wrote:
>>>
>>>
>>>> -----Original Message-----
>>>> From: Auger Eric [mailto:eric.auger@xxxxxxxxxx]
>>>> Sent: Tuesday, February 27, 2018 8:27 AM
>>>> To: Alex Williamson <alex.williamson@xxxxxxxxxx>
>>>> Cc: Shameerali Kolothum Thodi <shameerali.kolothum.thodi@xxxxxxxxxx>;
>>>> pmorel@xxxxxxxxxxxxxxxxxx; kvm@xxxxxxxxxxxxxxx; linux-
>>>> kernel@xxxxxxxxxxxxxxx; Linuxarm <linuxarm@xxxxxxxxxx>; John Garry
>>>> <john.garry@xxxxxxxxxx>; xuwei (O) <xuwei5@xxxxxxxxxx>; Robin
>> Murphy
>>>> <robin.murphy@xxxxxxx>
>>>> Subject: Re: [PATCH v4 4/6] vfio/type1: check dma map request is within a
>> valid
>>>> iova range
>>>>
>>>> Hi,
>>>> On 27/02/18 00:13, Alex Williamson wrote:
>>>>> On Mon, 26 Feb 2018 23:05:43 +0100
>>>>> Auger Eric <eric.auger@xxxxxxxxxx> wrote:
>>>>>
>>>>>> Hi Shameer,
>>>>>>
>>>>>> [Adding Robin in CC]
>>>>>> On 21/02/18 13:22, Shameer Kolothum wrote:
>>>>>>> This checks and rejects any dma map request outside valid iova
>>>>>>> range.
>>>>>>>
>>>>>>> Signed-off-by: Shameer Kolothum
>>>> <shameerali.kolothum.thodi@xxxxxxxxxx>
>>>>>>> ---
>>>>>>> drivers/vfio/vfio_iommu_type1.c | 22 ++++++++++++++++++++++
>>>>>>> 1 file changed, 22 insertions(+)
>>>>>>>
>>>>>>> diff --git a/drivers/vfio/vfio_iommu_type1.c
>>>> b/drivers/vfio/vfio_iommu_type1.c
>>>>>>> index a80884e..3049393 100644
>>>>>>> --- a/drivers/vfio/vfio_iommu_type1.c
>>>>>>> +++ b/drivers/vfio/vfio_iommu_type1.c
>>>>>>> @@ -970,6 +970,23 @@ static int vfio_pin_map_dma(struct vfio_iommu
>>>> *iommu, struct vfio_dma *dma,
>>>>>>> return ret;
>>>>>>> }
>>>>>>>
>>>>>>> +/*
>>>>>>> + * Check dma map request is within a valid iova range
>>>>>>> + */
>>>>>>> +static bool vfio_iommu_iova_dma_valid(struct vfio_iommu *iommu,
>>>>>>> + dma_addr_t start, dma_addr_t end)
>>>>>>> +{
>>>>>>> + struct list_head *iova = &iommu->iova_list;
>>>>>>> + struct vfio_iova *node;
>>>>>>> +
>>>>>>> + list_for_each_entry(node, iova, list) {
>>>>>>> + if ((start >= node->start) && (end <= node->end))
>>>>>>> + return true;
>>>>>> I am now confused by the fact this change will prevent existing QEMU
>>>>>> from working with this series on some platforms. For instance QEMU virt
>>>>>> machine GPA space collides with Seattle PCI host bridge windows. On
>> ARM
>>>>>> the smmu and smmuv3 drivers report the PCI host bridge windows as
>>>>>> reserved regions which does not seem to be the case on other platforms.
>>>>>> The change happened in commit
>>>> 273df9635385b2156851c7ee49f40658d7bcb29d
>>>>>> (iommu/dma: Make PCI window reservation generic).
>>>>>>
>>>>>> For background, we already discussed the topic after LPC 2016. See
>>>>>> https://www.spinics.net/lists/kernel/msg2379607.html.
>>>>>>
>>>>>> So is it the right choice to expose PCI host bridge windows as reserved
>>>>>> regions? If yes shouldn't we make a difference between those and MSI
>>>>>> windows in this series and do not reject any user space DMA_MAP
>> attempt
>>>>>> within PCI host bridge windows.
>>>>>
>>>>> If the QEMU machine GPA collides with a reserved region today, then
>>>>> either:
>>>>>
>>>>> a) The mapping through the IOMMU works and the reserved region is
>> wrong
>>>>>
>>>>> or
>>>>>
>>>>> b) The mapping doesn't actually work, QEMU is at risk of data loss by
>>>>> being told that it worked, and we're justified in changing that
>>>>> behavior.
>>>>>
>>>>> Without knowing the specifics of SMMU, it doesn't particularly make
>>>>> sense to me to mark the entire PCI hierarchy MMIO range as reserved,
>>>>> unless perhaps the IOMMU is incapable of translating those IOVAs.
>>>> to me the limitation does not come from the smmu itself, which is a
>>>> separate HW block sitting between the root complex and the interconnect.
>>>> If ACS is not enforced by the PCIe subsystem, the transaction will never
>>>> reach the IOMMU.
>>>
>>> True. And we do have one such platform where ACS is not enforced but
>>> reserving the regions and possibly creating holes while launching VM will
>>> make it secure. But I do wonder how we will solve the device grouping
>>> in such cases.
>>>
>>> The Seattle PCI host bridge windows case you mentioned has any pci quirk
>>> to claim that they support ACS?
>> No there is none to my knowledge. I am applying Alex' not upstream ACS
>> overwrite patch.
>
> Ok. But isn't that patch actually applicable to cases where ACS is really supported
> by hardware but the capability is not available?

My understanding is normally yes. If you apply the patch whereas the HW
practically does not support ACS, then you fool the kernel pretending
there is isolation whereas there is not. I don't know the exact
capability of the HW on AMD Seattle and effectively I should have cared
about it much earlier and if the HW capability were supported and not
properly exposed we should have implemented a clean quirk for this platform.


I am just trying to see whether
> the argument that we should allow DMA MAP requests for this(non-ACS case)
> even if the Guest GPA conflict with reserved region holds good. The fact that may
> be it was working before is that the Guest never actually allocated any GPA from
> the reserved region or maybe I am missing something here.

If my understanding is correct, in ideal world we would report the PCI
host bridge window as reserved only in case ACS is not supported. If you
apply the patch and overrides the ACS, then the DMA_MAP would be
allowed. In case the HW does not support ACS, then you could face
situations where one EP tries to access GPA that never reaches the IOMMU
(because it corresponds to the BAR of another downstream EP). Same can
happen at the moment.

Thanks

Eric
>
> Thanks,
> Shameer
>
>> Thanks
>>
>> Eric
>>>
>>>> In the case of such overlap, shouldn't we just warn the end-user that
>>>> this situation is dangerous instead of forbidding the use case which
>>>> worked "in most cases" until now.
>>>
>>> Yes, may be something similar to the allow_unsafe_interrupts case, if
>>> that is acceptable.
>>>
>>> Thanks,
>>> Shameer
>>>
>>>>> Are we trying to prevent untranslated p2p with this reserved range?
>>>>> That's not necessarily a terrible idea, but it seems that doing it for
>>>>> that purpose would need to be a lot smarter, taking into account ACS
>>>>> and precisely selecting ranges within the peer address space that would
>>>>> be untranslated. Perhaps only populated MMIO within non-ACS
>>>>> hierarchies. Thanks,
>>>>
>>>> Indeed taking into account the ACS capability would refine the
>>>> situations where a risk exists.
>>>>
>>>> Thanks
>>>>
>>>> Eric
>>>>>
>>>>> Alex
>>>>>