Chris Wilson wrote: > This is similar to remap_pfn_range(), and uses the recently refactor > code to do the page table walking. The key difference is that is back > propagates its error as this is required for use from within a pagefault > handler. The other difference, is that it combine the page protection > from io-mapping, which is known from when the io-mapping is created, > with the per-vma page protection flags. This avoids having to walk the > entire system description to rediscover the special page protection > established for the io-mapping. > > Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> Looks okay to me: Acked-by: Kirill A. Shutemov <kirill.shutemov@xxxxxxxxxxxxxxx> -- Kirill A. Shutemov _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/intel-gfx
