KVM: arm64: Optimize host memory aborts

The kvm_pgtable_stage2_find_range() function is used in the host memory
abort path to try and look for the largest block mapping that can be
used to map the faulting address. In order to do so, the function
currently walks the stage-2 page-table and looks for existing
incompatible mappings within the range of the largest possible block.
If incompatible mappings are found, it tries the same procedure again,
but using a smaller block range, and repeats until a matching range is
found (potentially up to page granularity). While this approach has
benefits (mostly in the fact that it proactively coalesces host stage-2
mappings), it can be slow if the ranges are fragmented, and it isn't
optimized to deal with CPUs faulting on the same IPA as all of them will
do all the work every time.

To avoid these issues, remove kvm_pgtable_stage2_find_range(), and walk
the page-table only once in the host_mem_abort() path to find the
closest leaf to the input address. With this, use the corresponding
range if it is invalid and not owned by another entity. If a valid leaf
is found, return -EAGAIN similar to what is done in the
kvm_pgtable_stage2_map() path to optimize concurrent faults.

Reviewed-by: Fuad Tabba <tabba@google.com>
Signed-off-by: Quentin Perret <qperret@google.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
Link: https://lore.kernel.org/r/20210809152448.1810400-7-qperret@google.com

2 files changed, 44 insertions, 75 deletions

diff --git a/arch/arm64/kvm/hyp/nvhe/mem_protect.c b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
index 74280a753efb..2148d3968aa5 100644
--- a/arch/arm64/kvm/hyp/nvhe/mem_protect.c
+++ b/arch/arm64/kvm/hyp/nvhe/mem_protect.c
@@ -159,6 +159,11 @@ static int host_stage2_unmap_dev_all(void)
 	return kvm_pgtable_stage2_unmap(pgt, addr, BIT(pgt->ia_bits) - addr);
 }
 
+struct kvm_mem_range {
+	u64 start;
+	u64 end;
+};
+
 static bool find_mem_range(phys_addr_t addr, struct kvm_mem_range *range)
 {
 	int cur, left = 0, right = hyp_memblock_nr;
@@ -227,6 +232,44 @@ static inline int __host_stage2_idmap(u64 start, u64 end,
 		__ret;							\
 	 })
 
+static inline bool range_included(struct kvm_mem_range *child,
+				  struct kvm_mem_range *parent)
+{
+	return parent->start <= child->start && child->end <= parent->end;
+}
+
+static int host_stage2_adjust_range(u64 addr, struct kvm_mem_range *range)
+{
+	struct kvm_mem_range cur;
+	kvm_pte_t pte;
+	u32 level;
+	int ret;
+
+	hyp_assert_lock_held(&host_kvm.lock);
+	ret = kvm_pgtable_get_leaf(&host_kvm.pgt, addr, &pte, &level);
+	if (ret)
+		return ret;
+
+	if (kvm_pte_valid(pte))
+		return -EAGAIN;
+
+	if (pte)
+		return -EPERM;
+
+	do {
+		u64 granule = kvm_granule_size(level);
+		cur.start = ALIGN_DOWN(addr, granule);
+		cur.end = cur.start + granule;
+		level++;
+	} while ((level < KVM_PGTABLE_MAX_LEVELS) &&
+			!(kvm_level_supports_block_mapping(level) &&
+			  range_included(&cur, range)));
+
+	*range = cur;
+
+	return 0;
+}
+
 static int host_stage2_idmap(u64 addr)
 {
 	enum kvm_pgtable_prot prot = KVM_PGTABLE_PROT_R | KVM_PGTABLE_PROT_W;
@@ -238,7 +281,7 @@ static int host_stage2_idmap(u64 addr)
 		prot |= KVM_PGTABLE_PROT_X;
 
 	hyp_spin_lock(&host_kvm.lock);
-	ret = kvm_pgtable_stage2_find_range(&host_kvm.pgt, addr, prot, &range);
+	ret = host_stage2_adjust_range(addr, &range);
 	if (ret)
 		goto unlock;
 
diff --git a/arch/arm64/kvm/hyp/pgtable.c b/arch/arm64/kvm/hyp/pgtable.c
index 49d768b92997..4dff2ad39ee4 100644
--- a/arch/arm64/kvm/hyp/pgtable.c
+++ b/arch/arm64/kvm/hyp/pgtable.c
@@ -1102,77 +1102,3 @@ void kvm_pgtable_stage2_destroy(struct kvm_pgtable *pgt)
 	pgt->mm_ops->free_pages_exact(pgt->pgd, pgd_sz);
 	pgt->pgd = NULL;
 }
-
-#define KVM_PTE_LEAF_S2_COMPAT_MASK	(KVM_PTE_LEAF_ATTR_S2_PERMS | \
-					 KVM_PTE_LEAF_ATTR_LO_S2_MEMATTR | \
-					 KVM_PTE_LEAF_ATTR_S2_IGNORED)
-
-static int stage2_check_permission_walker(u64 addr, u64 end, u32 level,
-					  kvm_pte_t *ptep,
-					  enum kvm_pgtable_walk_flags flag,
-					  void * const arg)
-{
-	kvm_pte_t old_attr, pte = *ptep, *new_attr = arg;
-
-	/*
-	 * Compatible mappings are either invalid and owned by the page-table
-	 * owner (whose id is 0), or valid with matching permission attributes.
-	 */
-	if (kvm_pte_valid(pte)) {
-		old_attr = pte & KVM_PTE_LEAF_S2_COMPAT_MASK;
-		if (old_attr != *new_attr)
-			return -EEXIST;
-	} else if (pte) {
-		return -EEXIST;
-	}
-
-	return 0;
-}
-
-int kvm_pgtable_stage2_find_range(struct kvm_pgtable *pgt, u64 addr,
-				  enum kvm_pgtable_prot prot,
-				  struct kvm_mem_range *range)
-{
-	kvm_pte_t attr;
-	struct kvm_pgtable_walker check_perm_walker = {
-		.cb		= stage2_check_permission_walker,
-		.flags		= KVM_PGTABLE_WALK_LEAF,
-		.arg		= &attr,
-	};
-	u64 granule, start, end;
-	u32 level;
-	int ret;
-
-	ret = stage2_set_prot_attr(pgt, prot, &attr);
-	if (ret)
-		return ret;
-	attr &= KVM_PTE_LEAF_S2_COMPAT_MASK;
-
-	for (level = pgt->start_level; level < KVM_PGTABLE_MAX_LEVELS; level++) {
-		granule = kvm_granule_size(level);
-		start = ALIGN_DOWN(addr, granule);
-		end = start + granule;
-
-		if (!kvm_level_supports_block_mapping(level))
-			continue;
-
-		if (start < range->start || range->end < end)
-			continue;
-
-		/*
-		 * Check the presence of existing mappings with incompatible
-		 * permissions within the current block range, and try one level
-		 * deeper if one is found.
-		 */
-		ret = kvm_pgtable_walk(pgt, start, granule, &check_perm_walker);
-		if (ret != -EEXIST)
-			break;
-	}
-
-	if (!ret) {
-		range->start = start;
-		range->end = end;
-	}
-
-	return ret;
-}