[PATCH] dax: Fix missed PMD wakeups

[Jan Kara]

Signed-off-by: Jan Kara <jack@xxxxxxx>
---
 fs/dax.c | 46 +++++++++++++++++++++++++---------------------
 1 file changed, 25 insertions(+), 21 deletions(-)

diff --git a/fs/dax.c b/fs/dax.c
index fe5e33810cd4..3fe655d38c7a 100644
--- a/fs/dax.c
+++ b/fs/dax.c
@@ -191,15 +191,18 @@ static void dax_wake_entry(struct xa_state *xas, void *entry, bool wake_all)
 		__wake_up(wq, TASK_NORMAL, wake_all ? 0 : 1, &key);
 }
 
+#define DAX_ENTRY_CONFLICT dax_make_entry(pfn_to_pfn_t(1), DAX_EMPTY)
 /*
  * Look up entry in page cache, wait for it to become unlocked if it
  * is a DAX entry and return it.  The caller must subsequently call
  * put_unlocked_entry() if it did not lock the entry or dax_unlock_entry()
- * if it did.
+ * if it did. 'order' is the minimum order of entry to return. If there's no
+ * entry of sufficiently large order but there are some entries of lower order
+ * in the range described by xas, return special DAX_ENTRY_CONFLICT value.
  *
  * Must be called with the i_pages lock held.
  */
-static void *get_unlocked_entry(struct xa_state *xas)
+static void *get_unlocked_entry(struct xa_state *xas, unsigned int order)
 {
 	void *entry;
 	struct wait_exceptional_entry_queue ewait;
@@ -210,6 +213,8 @@ static void *get_unlocked_entry(struct xa_state *xas)
 
 	for (;;) {
 		entry = xas_find_conflict(xas);
+		if (dax_entry_order(entry) < order)
+			return DAX_ENTRY_CONFLICT;
 		if (!entry || WARN_ON_ONCE(!xa_is_value(entry)) ||
 				!dax_is_locked(entry))
 			return entry;
@@ -254,7 +259,7 @@ static void wait_entry_unlocked(struct xa_state *xas, void *entry)
 static void put_unlocked_entry(struct xa_state *xas, void *entry)
 {
 	/* If we were the only waiter woken, wake the next one */
-	if (entry)
+	if (entry && entry != DAX_ENTRY_CONFLICT)
 		dax_wake_entry(xas, entry, false);
 }
 
@@ -461,7 +466,7 @@ void dax_unlock_page(struct page *page, dax_entry_t cookie)
  * overlap with xarray value entries.
  */
 static void *grab_mapping_entry(struct xa_state *xas,
-		struct address_space *mapping, unsigned long size_flag)
+		struct address_space *mapping, unsigned int order)
 {
 	unsigned long index = xas->xa_index;
 	bool pmd_downgrade = false; /* splitting PMD entry into PTE entries? */
@@ -469,20 +474,16 @@ static void *grab_mapping_entry(struct xa_state *xas,
 
 retry:
 	xas_lock_irq(xas);
-	entry = get_unlocked_entry(xas);
-
+	entry = get_unlocked_entry(xas, order);
+	if (entry == DAX_ENTRY_CONFLICT)
+		goto fallback;
 	if (entry) {
 		if (!xa_is_value(entry)) {
 			xas_set_err(xas, EIO);
 			goto out_unlock;
 		}
 
-		if (size_flag & DAX_PMD) {
-			if (dax_is_pte_entry(entry)) {
-				put_unlocked_entry(xas, entry);
-				goto fallback;
-			}
-		} else { /* trying to grab a PTE entry */
+		if (order == 0) {
 			if (dax_is_pmd_entry(entry) &&
 			    (dax_is_zero_entry(entry) ||
 			     dax_is_empty_entry(entry))) {
@@ -523,7 +524,11 @@ static void *grab_mapping_entry(struct xa_state *xas,
 	if (entry) {
 		dax_lock_entry(xas, entry);
 	} else {
-		entry = dax_make_entry(pfn_to_pfn_t(0), size_flag | DAX_EMPTY);
+		unsigned long flags = DAX_EMPTY;
+
+		if (order > 0)
+			flags |= DAX_PMD;
+		entry = dax_make_entry(pfn_to_pfn_t(0), flags);
 		dax_lock_entry(xas, entry);
 		if (xas_error(xas))
 			goto out_unlock;
@@ -594,7 +599,7 @@ struct page *dax_layout_busy_page(struct address_space *mapping)
 		if (WARN_ON_ONCE(!xa_is_value(entry)))
 			continue;
 		if (unlikely(dax_is_locked(entry)))
-			entry = get_unlocked_entry(&xas);
+			entry = get_unlocked_entry(&xas, 0);
 		if (entry)
 			page = dax_busy_page(entry);
 		put_unlocked_entry(&xas, entry);
@@ -621,7 +626,7 @@ static int __dax_invalidate_entry(struct address_space *mapping,
 	void *entry;
 
 	xas_lock_irq(&xas);
-	entry = get_unlocked_entry(&xas);
+	entry = get_unlocked_entry(&xas, 0);
 	if (!entry || WARN_ON_ONCE(!xa_is_value(entry)))
 		goto out;
 	if (!trunc &&
@@ -848,7 +853,7 @@ static int dax_writeback_one(struct xa_state *xas, struct dax_device *dax_dev,
 	if (unlikely(dax_is_locked(entry))) {
 		void *old_entry = entry;
 
-		entry = get_unlocked_entry(xas);
+		entry = get_unlocked_entry(xas, 0);
 
 		/* Entry got punched out / reallocated? */
 		if (!entry || WARN_ON_ONCE(!xa_is_value(entry)))
@@ -1509,7 +1514,7 @@ static vm_fault_t dax_iomap_pmd_fault(struct vm_fault *vmf, pfn_t *pfnp,
 	 * entry is already in the array, for instance), it will return
 	 * VM_FAULT_FALLBACK.
 	 */
-	entry = grab_mapping_entry(&xas, mapping, DAX_PMD);
+	entry = grab_mapping_entry(&xas, mapping, PMD_ORDER);
 	if (xa_is_internal(entry)) {
 		result = xa_to_internal(entry);
 		goto fallback;
@@ -1658,11 +1663,10 @@ dax_insert_pfn_mkwrite(struct vm_fault *vmf, pfn_t pfn, unsigned int order)
 	vm_fault_t ret;
 
 	xas_lock_irq(&xas);
-	entry = get_unlocked_entry(&xas);
+	entry = get_unlocked_entry(&xas, order);
 	/* Did we race with someone splitting entry or so? */
-	if (!entry ||
-	    (order == 0 && !dax_is_pte_entry(entry)) ||
-	    (order == PMD_ORDER && !dax_is_pmd_entry(entry))) {
+	if (!entry || entry == DAX_ENTRY_CONFLICT ||
+	    (order == 0 && !dax_is_pte_entry(entry))) {
 		put_unlocked_entry(&xas, entry);
 		xas_unlock_irq(&xas);
 		trace_dax_insert_pfn_mkwrite_no_entry(mapping->host, vmf,
-- 
2.16.4


--Nq2Wo0NMKNjxTN9z--