KVM: arm64: Fix host stage-2 PGD refcount

The KVM page-table library refcounts the pages of concatenated stage-2
PGDs individually. However, when running KVM in protected mode, the
host's stage-2 PGD is currently managed by EL2 as a single high-order
compound page, which can cause the refcount of the tail pages to reach 0
when they shouldn't, hence corrupting the page-table.

Fix this by introducing a new hyp_split_page() helper in the EL2 page
allocator (matching the kernel's split_page() function), and make use of
it from host_s2_zalloc_pages_exact().

Fixes: 07193dbcc777 ("KVM: arm64: Wrap the host with a stage 2")
Acked-by: Will Deacon <will@kernel.org>
Suggested-by: Will Deacon <will@kernel.org>
Signed-off-by: Quentin Perret <qperret@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20211005090155.734578-5-qperret@google.com

diff --git a/arch/arm64/kvm/hyp/include/nvhe/gfp.h b/arch/arm64/kvm/hyp/include/nvhe/gfp.h
index fb0f523d14921c6c79b246e63111f4c1cbbf0767..0a048dc06a7d789ea93f669d08a8d426630771f0 100644 (file)
--- a/arch/arm64/kvm/hyp/include/nvhe/gfp.h
+++ b/arch/arm64/kvm/hyp/include/nvhe/gfp.h
@@ -24,6 +24,7 @@ struct hyp_pool {
 
 /* Allocation */
 void *hyp_alloc_pages(struct hyp_pool *pool, unsigned short order);
+void hyp_split_page(struct hyp_page *page);
 void hyp_get_page(struct hyp_pool *pool, void *addr);
 void hyp_put_page(struct hyp_pool *pool, void *addr);
 

diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
index bacd493a4eacdefc1af400ec1a97309f02ac3510..34eeb524b686075aca7d60876b6429d25df0cbf6 100644 (file)
--- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c
+++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
@@ -35,7 +35,18 @@ const u8 pkvm_hyp_id = 1;
 
 static void *host_s2_zalloc_pages_exact(size_t size)
 {
-       return hyp_alloc_pages(&host_s2_pool, get_order(size));
+       void *addr = hyp_alloc_pages(&host_s2_pool, get_order(size));
+
+       hyp_split_page(hyp_virt_to_page(addr));
+
+       /*
+        * The size of concatenated PGDs is always a power of two of PAGE_SIZE,
+        * so there should be no need to free any of the tail pages to make the
+        * allocation exact.
+        */
+       WARN_ON(size != (PAGE_SIZE << get_order(size)));
+
+       return addr;
 }
 
 static void *host_s2_zalloc_page(void *pool)

diff --git a/arch/arm64/kvm/hyp/nvhe/page_alloc.c b/arch/arm64/kvm/hyp/nvhe/page_alloc.c
index 41fc25bdfb34660f470ef9dda78fe0a8a7aee989..a6e874e61a40e7cc989394960d11fcd1577fbfa4 100644 (file)
--- a/arch/arm64/kvm/hyp/nvhe/page_alloc.c
+++ b/arch/arm64/kvm/hyp/nvhe/page_alloc.c
@@ -193,6 +193,20 @@ void hyp_get_page(struct hyp_pool *pool, void *addr)
        hyp_spin_unlock(&pool->lock);
 }
 
+void hyp_split_page(struct hyp_page *p)
+{
+       unsigned short order = p->order;
+       unsigned int i;
+
+       p->order = 0;
+       for (i = 1; i < (1 << order); i++) {
+               struct hyp_page *tail = p + i;
+
+               tail->order = 0;
+               hyp_set_page_refcounted(tail);
+       }
+}
+
 void *hyp_alloc_pages(struct hyp_pool *pool, unsigned short order)
 {
        unsigned short i = order;