x86/mm/p2m: stop checking for IOMMU shared page tables in mmio_order()
authorPaul Durrant <paul.durrant@citrix.com>
Mon, 17 Dec 2018 09:22:59 +0000 (09:22 +0000)
committerAndrew Cooper <andrew.cooper3@citrix.com>
Thu, 3 Jan 2019 10:36:06 +0000 (10:36 +0000)
commita5b0eb363694e7e15405f0b3fc5fb6fab79df1db
tree3cf0142a800af671790d330b315a69a4564312f4
parente8afe1124cc11259fcd9942ff5af9c1b2bd329ca
x86/mm/p2m: stop checking for IOMMU shared page tables in mmio_order()

Now that the iommu_map() and iommu_unmap() operations take an order
parameter and elide flushing there's no strong reason why modifying MMIO
ranges in the p2m should be restricted to a 4k granularity simply because
the IOMMU is enabled but shared page tables are not in operation.

Signed-off-by: Paul Durrant <paul.durrant@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
xen/arch/x86/mm/p2m.c