Re: [PATCH for-4.15 2/4] xen/iommu: x86: Free the IOMMU page-tables with the pgtables.lock held

[Julien Grall <julien@xen.org>]

Hi Jan,

On 23/12/2020 13:48, Jan Beulich wrote:
> On 22.12.2020 16:43, Julien Grall wrote:
>> From: Julien Grall <jgrall@amazon.com>
>>
>> The pgtables.lock is protecting access to the page list pgtables.list.
>> However, iommu_free_pgtables() will not held it. I guess it was assumed
>> that page-tables cannot be allocated while the domain is dying.
>>
>> Unfortunately, there is no guarantee that iommu_map() will not be
>> called while a domain is dying (it looks like to be possible from
>> XEN_DOMCTL_memory_mapping).
> 
> I'd rather disallow any new allocations for a dying domain, not
> the least because ...

Patch #4 will disallow such allocation. However...

> 
>> So it would be possible to be concurrently
>> allocate memory and free the page-tables.
>>
>> Therefore, we need to held the lock when freeing the page tables.
> 
> ... we should try to avoid holding locks across allocation /
> freeing functions wherever possible. >
> As to where to place a respective check - I wonder if we wouldn't
> be better off disallowing a majority of domctl-s (and perhaps
> other operations) on dying domains. Thoughts?

... this is still pretty racy because you need to guarantee that 
d->is_dying is seen by the other processors to prevent allocation.

As to whether we can forbid most of the domctl-s, I would agree this is 
a good move. But this doesn't remove the underlying problem here.

We are hoping that a top-level function will protect us against the 
race. Given the IOMMU code is quite deep in the callstack, this is 
something pretty hard to guarantee with future change.

So I still think we need a way to mitigate the issue.

Cheers,

-- 
Julien Grall