Re: [Qemu-devel] [PATCH 5/7] memory: Allow replay of IOMMU mapping notifications

[Alex Williamson]

On Thu, 2015-09-24 at 14:33 +1000, David Gibson wrote:
> When we have guest visible IOMMUs, we allow notifiers to be registered
> which will be informed of all changes to IOMMU mappings.  This is used by
> vfio to keep the host IOMMU mappings in sync with guest IOMMU mappings.
> 
> However, unlike with a memory region listener, an iommu notifier won't be
> told about any mappings which already exist in the (guest) IOMMU at the
> time it is registered.  This can cause problems if hotplugging a VFIO
> device onto a guest bus which had existing guest IOMMU mappings, but didn't
> previously have an VFIO devices (and hence no host IOMMU mappings).
> 
> This adds a memory_region_register_iommu_notifier_replay() function to
> handle this case.  As well as registering the new notifier it replays
> existing mappings.  Because the IOMMU memory region doesn't internally
> remember the granularity of the guest IOMMU it has a small hack where the
> caller must specify a granularity at which to replay mappings.
> 
> If there are finer mappings in the guest IOMMU these will be reported in
> the iotlb structures passed to the notifier which it must handle (probably
> causing it to flag an error).  This isn't new - the VFIO iommu notifier
> must already handle notifications about guest IOMMU mappings too short
> for it to represent in the host IOMMU.
> 
> Signed-off-by: David Gibson <address@hidden>
> ---
>  include/exec/memory.h | 17 +++++++++++++++++
>  memory.c              | 18 ++++++++++++++++++
>  2 files changed, 35 insertions(+)
> 
> diff --git a/include/exec/memory.h b/include/exec/memory.h
> index 5baaf48..304f985 100644
> --- a/include/exec/memory.h
> +++ b/include/exec/memory.h
> @@ -583,6 +583,23 @@ void memory_region_notify_iommu(MemoryRegion *mr,
>  void memory_region_register_iommu_notifier(MemoryRegion *mr, Notifier *n);
>  
>  /**
> + * memory_region_register_iommu_notifier_replay: register a notifier
> + * for changes to IOMMU translation entries, and replay existing IOMMU
> + * translations to the new notifier.
> + *
> + * @mr: the memory region to observe
> + * @n: the notifier to be added; the notifier receives a pointer to an
> + *     #IOMMUTLBEntry as the opaque value; the pointer ceases to be
> + *     valid on exit from the notifier.
> + * @granularity: Minimum page granularity to replay notifications for
> + * @is_write: Whether to treat the replay as a translate "write"
> + *     through the iommu
> + */
> +void memory_region_register_iommu_notifier_replay(MemoryRegion *mr, Notifier 
> *n,
> +                                                  hwaddr granularity,
> +                                                  bool is_write);
> +
> +/**
>   * memory_region_unregister_iommu_notifier: unregister a notifier for
>   * changes to IOMMU translation entries.
>   *
> diff --git a/memory.c b/memory.c
> index ef87363..b4b6861 100644
> --- a/memory.c
> +++ b/memory.c
> @@ -1403,6 +1403,24 @@ void 
> memory_region_register_iommu_notifier(MemoryRegion *mr, Notifier *n)
>      notifier_list_add(&mr->iommu_notify, n);
>  }
>  
> +void memory_region_register_iommu_notifier_replay(MemoryRegion *mr, Notifier 
> *n,
> +                                                  hwaddr granularity,
> +                                                  bool is_write)
> +{
> +    hwaddr addr;
> +    IOMMUTLBEntry iotlb;
> +
> +    memory_region_register_iommu_notifier(mr, n);
> +
> +    for (addr = 0; addr < memory_region_size(mr); addr += granularity) {
> +
> +        iotlb = mr->iommu_ops->translate(mr, addr, is_write);
> +        if (iotlb.perm != IOMMU_NONE) {
> +            n->notify(n, &iotlb);
> +        }
> +    }
> +}
> +


When memory_listener_register() replays mappings, it does so on an rcu
copy of the flatview for each AddressSpace.  Here we don't seem to have
anything protecting against concurrency... do we need to worry about
that?

>  void memory_region_unregister_iommu_notifier(Notifier *n)
>  {
>      notifier_remove(n);