[PATCH 4.19 048/131] KVM: Properly check if "page" is valid in kvm_vcpu_unmap

[Greg Kroah-Hartman]

From: KarimAllah Ahmed <karahmed@xxxxxxxxx>

commit b614c6027896ff9ad6757122e84760d938cab15e upstream.

The field "page" is initialized to KVM_UNMAPPED_PAGE when it is not used
(i.e. when the memory lives outside kernel control). So this check will
always end up using kunmap even for memremap regions.

Fixes: e45adf665a53 ("KVM: Introduce a new guest mapping API")
Signed-off-by: KarimAllah Ahmed <karahmed@xxxxxxxxx>
Signed-off-by: Paolo Bonzini <pbonzini@xxxxxxxxxx>
Signed-off-by: Ben Hutchings <ben.hutchings@xxxxxxxxxxxxxxx>
Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx>
---
 virt/kvm/kvm_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 5b949aa273de5..33b288469c70c 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -1754,7 +1754,7 @@ void kvm_vcpu_unmap(struct kvm_vcpu *vcpu, struct kvm_host_map *map,
 	if (!map->hva)
 		return;
 
-	if (map->page)
+	if (map->page != KVM_UNMAPPED_PAGE)
 		kunmap(map->page);
 #ifdef CONFIG_HAS_IOMEM
 	else
-- 
2.20.1