[PATCH 5.4 004/151] KVM: arm64: Force PTE mapping on fault resulting in a device mapping

[Greg Kroah-Hartman]

From: Santosh Shukla <sashukla@xxxxxxxxxx>

[ Upstream commit 91a2c34b7d6fadc9c5d9433c620ea4c32ee7cae8 ]

VFIO allows a device driver to resolve a fault by mapping a MMIO
range. This can be subsequently result in user_mem_abort() to
try and compute a huge mapping based on the MMIO pfn, which is
a sure recipe for things to go wrong.

Instead, force a PTE mapping when the pfn faulted in has a device
mapping.

Fixes: 6d674e28f642 ("KVM: arm/arm64: Properly handle faulting of device mappings")
Suggested-by: Marc Zyngier <maz@xxxxxxxxxx>
Signed-off-by: Santosh Shukla <sashukla@xxxxxxxxxx>
[maz: rewritten commit message]
Signed-off-by: Marc Zyngier <maz@xxxxxxxxxx>
Reviewed-by: Gavin Shan <gshan@xxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx
Link: https://lore.kernel.org/r/1603711447-11998-2-git-send-email-sashukla@xxxxxxxxxx
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
 virt/kvm/arm/mmu.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/virt/kvm/arm/mmu.c b/virt/kvm/arm/mmu.c
index 8700402f3000d..03a586ab6d27b 100644
--- a/virt/kvm/arm/mmu.c
+++ b/virt/kvm/arm/mmu.c
@@ -1756,6 +1756,7 @@ static int user_mem_abort(struct kvm_vcpu *vcpu, phys_addr_t fault_ipa,
 	if (kvm_is_device_pfn(pfn)) {
 		mem_type = PAGE_S2_DEVICE;
 		flags |= KVM_S2PTE_FLAG_IS_IOMAP;
+		force_pte = true;
 	} else if (logging_active) {
 		/*
 		 * Faults on pages in a memslot with logging enabled
-- 
2.27.0