Re: [RFC 0/3] iommu/iova: Unsafe locking in find_iova()

[Dmitry Safonov]

Hi,

Any opinion on that?
It looks to remove a source of possible issues and
has a nice diffstat.

2018-06-21 19:08 GMT+01:00 Dmitry Safonov <dima@xxxxxxxxxx>:
> find_iova() looks to be using a bad locking practice: it locks the
> returned iova only for the search time.
> And looking in code, the element can be removed from the tree and freed
> under rbtree lock. That happens during memory hot-unplug and cleanup on
> module removal.
> Here I cleanup users of the function and delete it.
>
> Dmitry Safonov (3):
>   iommu/iova: Find and split iova under rbtree's lock
>   iommu/iova: Make free_iova() atomic
>   iommu/iova: Remove find_iova()
>
>  drivers/iommu/intel-iommu.c | 14 +++----------
>  drivers/iommu/iova.c        | 48 +++++++++++++++++----------------------------
>  include/linux/iova.h        | 17 ++++------------
>  3 files changed, 25 insertions(+), 54 deletions(-)
>

Thanks,
             Dmitry