Merge tag 'vfio-v4.4-rc1' of git://github.com/awilliam/linux-vfio

Pull VFIO updates from Alex Williamson:
 - Use kernel interfaces for VPD emulation (Alex Williamson)
 - Platform fix for releasing IRQs (Eric Auger)
 - Type1 IOMMU always advertises PAGE_SIZE support when smaller mapping
   sizes are available (Eric Auger)
 - Platform fixes for incorrectly using copies of structures rather than
   pointers to structures (James Morse)
 - Rework platform reset modules, fix leak, and add AMD xgbe reset
   module (Eric Auger)
 - Fix vfio_device_get_from_name() return value (Joerg Roedel)
 - No-IOMMU interface (Alex Williamson)
 - Fix potential out of bounds array access in PCI config handling (Dan
   Carpenter)

* tag 'vfio-v4.4-rc1' of git://github.com/awilliam/linux-vfio:
  vfio/pci: make an array larger
  vfio: Include No-IOMMU mode
  vfio: Fix bug in vfio_device_get_from_name()
  VFIO: platform: reset: AMD xgbe reset module
  vfio: platform: reset: calxedaxgmac: fix ioaddr leak
  vfio: platform: add dev_info on device reset
  vfio: platform: use list of registered reset function
  vfio: platform: add compat in vfio_platform_device
  vfio: platform: reset: calxedaxgmac: add reset function registration
  vfio: platform: introduce module_vfio_reset_handler macro
  vfio: platform: add capability to register a reset function
  vfio: platform: introduce vfio-platform-base module
  vfio/platform: store mapped memory in region, instead of an on-stack copy
  vfio/type1: handle case where IOMMU does not support PAGE_SIZE size
  VFIO: platform: clear IRQ_NOAUTOEN when de-assigning the IRQ
  vfio/pci: Use kernel VPD access functions
  vfio: Whitelist PCI bridges

diff --cc drivers/vfio/Kconfig
index 850d86ca685b273344271eda7c95ad4f82b28244,b6d3cdc2791b38de76c3c5e3644128700b3eb7aa..da6e2ce77495b21ec127209664567b7e2f8f467e
--- 1/drivers/vfio/Kconfig
--- 2/drivers/vfio/Kconfig
+++ b/drivers/vfio/Kconfig
@@@ -31,6 -31,20 +31,21 @@@ menuconfig VFI
  
          If you don't know what to do here, say N.
  
+ menuconfig VFIO_NOIOMMU
+       bool "VFIO No-IOMMU support"
+       depends on VFIO
+       help
+         VFIO is built on the ability to isolate devices using the IOMMU.
+         Only with an IOMMU can userspace access to DMA capable devices be
+         considered secure.  VFIO No-IOMMU mode enables IOMMU groups for
+         devices without IOMMU backing for the purpose of re-using the VFIO
+         infrastructure in a non-secure mode.  Use of this mode will result
+         in an unsupportable kernel and will therefore taint the kernel.
+         Device assignment to virtual machines is also not possible with
+         this mode since there is no IOMMU to provide DMA translation.
+ 
+         If you don't know what to do here, say N.
+ 
  source "drivers/vfio/pci/Kconfig"
  source "drivers/vfio/platform/Kconfig"
 +source "virt/lib/Kconfig"