On Mon, Mar 21, 2016 at 4:30 AM, Peter Xu <peterx@redhat.com> wrote:

> On Sat, Mar 19, 2016 at 11:40:04AM +0200, Aviv B.D. wrote:
> [...]
> > As far as I understand the code, currently there is no way to turn off
> the
> > IOTLB.
> > Furthermore. the IOTLB is not implemented as LRU, and actually caches
> > (indefinitely)
> > any accessed address, without any size constrains. I use those
> assumptions
> > to know
> > whether the current invalidation is for unmap operation or map operation.
>
> Please have a look at VTD_IOTLB_MAX_SIZE. It seems to be the size of
> the hash.
>
> Btw, I guess it's a much bigger problem if IOTLB has unlimited cache
> size...
>
> Thanks.

-- peterx
>

You are correct, VTD_IOTLB_MAX_SIZE limits the cache size (and reset the
whole cache
if this threshold exceeds...) I will think on another mechanism to identify
the correct
action for each invalidation.

Thanks,
Aviv.