Turned out that I was able to iterate over it, and remove both the ->cookie_to_index and the ->demap indirection from iommu_tbl_ops.
That leaves only the odd iommu_flushall() hook, I'm trying to find the history behind that (needed for sun4u platforms, afaik, and not sure if there are other ways to achieve this). Just sent out a new patch-set, which should reach all the people/lists on this thread. If it's possible, it would also help to share any preliminary code-samples with the iommu_table_ops you folks have in mind. Please take a look at patchv5, and share your thoughts. --Sowmini _______________________________________________ Linuxppc-dev mailing list Linuxppc-dev@lists.ozlabs.org https://lists.ozlabs.org/listinfo/linuxppc-dev
