Re: [RESEND PATCH v3] iommu/iova: Optimize alloc_iova with rbtree_augmented

[Peng Zhang]

Hello everyone,

Compared with v2, v3 mainly updated the commit message,
and set the upper limit when scanning the linked list.

In the normal case, the performance of this algorithm
is almost the same as before. In extreme cases, this
algorithm is much better than before. And this algorithm
does not bring additional memory overhead.

Comments welcome.

Thanks.
Peng