aboutsummaryrefslogtreecommitdiffstats
path: root/include/linux/mmu_notifier.h
diff options
context:
space:
mode:
Diffstat (limited to 'include/linux/mmu_notifier.h')
-rw-r--r--include/linux/mmu_notifier.h147
1 files changed, 118 insertions, 29 deletions
diff --git a/include/linux/mmu_notifier.h b/include/linux/mmu_notifier.h
index 1bd8e6a09a3c..9e6caa8ecd19 100644
--- a/include/linux/mmu_notifier.h
+++ b/include/linux/mmu_notifier.h
@@ -6,9 +6,12 @@
#include <linux/spinlock.h>
#include <linux/mm_types.h>
#include <linux/srcu.h>
+#include <linux/interval_tree.h>
+struct mmu_notifier_mm;
struct mmu_notifier;
-struct mmu_notifier_ops;
+struct mmu_notifier_range;
+struct mmu_interval_notifier;
/**
* enum mmu_notifier_event - reason for the mmu notifier callback
@@ -31,6 +34,9 @@ struct mmu_notifier_ops;
* access flags). User should soft dirty the page in the end callback to make
* sure that anyone relying on soft dirtyness catch pages that might be written
* through non CPU mappings.
+ *
+ * @MMU_NOTIFY_RELEASE: used during mmu_interval_notifier invalidate to signal
+ * that the mm refcount is zero and the range is no longer accessible.
*/
enum mmu_notifier_event {
MMU_NOTIFY_UNMAP = 0,
@@ -38,38 +44,11 @@ enum mmu_notifier_event {
MMU_NOTIFY_PROTECTION_VMA,
MMU_NOTIFY_PROTECTION_PAGE,
MMU_NOTIFY_SOFT_DIRTY,
-};
-
-#ifdef CONFIG_MMU_NOTIFIER
-
-#ifdef CONFIG_LOCKDEP
-extern struct lockdep_map __mmu_notifier_invalidate_range_start_map;
-#endif
-
-/*
- * The mmu notifier_mm structure is allocated and installed in
- * mm->mmu_notifier_mm inside the mm_take_all_locks() protected
- * critical section and it's released only when mm_count reaches zero
- * in mmdrop().
- */
-struct mmu_notifier_mm {
- /* all mmu notifiers registerd in this mm are queued in this list */
- struct hlist_head list;
- /* to serialize the list modifications and hlist_unhashed */
- spinlock_t lock;
+ MMU_NOTIFY_RELEASE,
};
#define MMU_NOTIFIER_RANGE_BLOCKABLE (1 << 0)
-struct mmu_notifier_range {
- struct vm_area_struct *vma;
- struct mm_struct *mm;
- unsigned long start;
- unsigned long end;
- unsigned flags;
- enum mmu_notifier_event event;
-};
-
struct mmu_notifier_ops {
/*
* Called either by mmu_notifier_unregister or when the mm is
@@ -249,6 +228,41 @@ struct mmu_notifier {
unsigned int users;
};
+/**
+ * struct mmu_interval_notifier_ops
+ * @invalidate: Upon return the caller must stop using any SPTEs within this
+ * range. This function can sleep. Return false only if sleeping
+ * was required but mmu_notifier_range_blockable(range) is false.
+ */
+struct mmu_interval_notifier_ops {
+ bool (*invalidate)(struct mmu_interval_notifier *mni,
+ const struct mmu_notifier_range *range,
+ unsigned long cur_seq);
+};
+
+struct mmu_interval_notifier {
+ struct interval_tree_node interval_tree;
+ const struct mmu_interval_notifier_ops *ops;
+ struct mm_struct *mm;
+ struct hlist_node deferred_item;
+ unsigned long invalidate_seq;
+};
+
+#ifdef CONFIG_MMU_NOTIFIER
+
+#ifdef CONFIG_LOCKDEP
+extern struct lockdep_map __mmu_notifier_invalidate_range_start_map;
+#endif
+
+struct mmu_notifier_range {
+ struct vm_area_struct *vma;
+ struct mm_struct *mm;
+ unsigned long start;
+ unsigned long end;
+ unsigned flags;
+ enum mmu_notifier_event event;
+};
+
static inline int mm_has_notifiers(struct mm_struct *mm)
{
return unlikely(mm->mmu_notifier_mm);
@@ -275,6 +289,81 @@ extern int __mmu_notifier_register(struct mmu_notifier *mn,
struct mm_struct *mm);
extern void mmu_notifier_unregister(struct mmu_notifier *mn,
struct mm_struct *mm);
+
+unsigned long mmu_interval_read_begin(struct mmu_interval_notifier *mni);
+int mmu_interval_notifier_insert(struct mmu_interval_notifier *mni,
+ struct mm_struct *mm, unsigned long start,
+ unsigned long length,
+ const struct mmu_interval_notifier_ops *ops);
+int mmu_interval_notifier_insert_locked(
+ struct mmu_interval_notifier *mni, struct mm_struct *mm,
+ unsigned long start, unsigned long length,
+ const struct mmu_interval_notifier_ops *ops);
+void mmu_interval_notifier_remove(struct mmu_interval_notifier *mni);
+
+/**
+ * mmu_interval_set_seq - Save the invalidation sequence
+ * @mni - The mni passed to invalidate
+ * @cur_seq - The cur_seq passed to the invalidate() callback
+ *
+ * This must be called unconditionally from the invalidate callback of a
+ * struct mmu_interval_notifier_ops under the same lock that is used to call
+ * mmu_interval_read_retry(). It updates the sequence number for later use by
+ * mmu_interval_read_retry(). The provided cur_seq will always be odd.
+ *
+ * If the caller does not call mmu_interval_read_begin() or
+ * mmu_interval_read_retry() then this call is not required.
+ */
+static inline void mmu_interval_set_seq(struct mmu_interval_notifier *mni,
+ unsigned long cur_seq)
+{
+ WRITE_ONCE(mni->invalidate_seq, cur_seq);
+}
+
+/**
+ * mmu_interval_read_retry - End a read side critical section against a VA range
+ * mni: The range
+ * seq: The return of the paired mmu_interval_read_begin()
+ *
+ * This MUST be called under a user provided lock that is also held
+ * unconditionally by op->invalidate() when it calls mmu_interval_set_seq().
+ *
+ * Each call should be paired with a single mmu_interval_read_begin() and
+ * should be used to conclude the read side.
+ *
+ * Returns true if an invalidation collided with this critical section, and
+ * the caller should retry.
+ */
+static inline bool mmu_interval_read_retry(struct mmu_interval_notifier *mni,
+ unsigned long seq)
+{
+ return mni->invalidate_seq != seq;
+}
+
+/**
+ * mmu_interval_check_retry - Test if a collision has occurred
+ * mni: The range
+ * seq: The return of the matching mmu_interval_read_begin()
+ *
+ * This can be used in the critical section between mmu_interval_read_begin()
+ * and mmu_interval_read_retry(). A return of true indicates an invalidation
+ * has collided with this critical region and a future
+ * mmu_interval_read_retry() will return true.
+ *
+ * False is not reliable and only suggests a collision may not have
+ * occured. It can be called many times and does not have to hold the user
+ * provided lock.
+ *
+ * This call can be used as part of loops and other expensive operations to
+ * expedite a retry.
+ */
+static inline bool mmu_interval_check_retry(struct mmu_interval_notifier *mni,
+ unsigned long seq)
+{
+ /* Pairs with the WRITE_ONCE in mmu_interval_set_seq() */
+ return READ_ONCE(mni->invalidate_seq) != seq;
+}
+
extern void __mmu_notifier_mm_destroy(struct mm_struct *mm);
extern void __mmu_notifier_release(struct mm_struct *mm);
extern int __mmu_notifier_clear_flush_young(struct mm_struct *mm,