NTFS: Fix cluster (de)allocators to work when the runlist is NULL and more

      importantly to take a locked runlist rather than them locking it
      which leads to lock reversal.

Signed-off-by: Anton Altaparmakov <aia21@cantab.net>

4 files changed, 32 insertions, 33 deletions

diff --git a/fs/ntfs/ChangeLog b/fs/ntfs/ChangeLog
index b0cc43bd2974..beed90c9f551 100644
--- a/fs/ntfs/ChangeLog
+++ b/fs/ntfs/ChangeLog
@@ -70,6 +70,9 @@ ToDo/Notes:
         - Add BUG() checks to ntfs_attr_make_non_resident() and ntfs_attr_set()
           to ensure that these functions are never called for compressed or
           encrypted attributes.
+        - Fix cluster (de)allocators to work when the runlist is NULL and more
+          importantly to take a locked runlist rather than them locking it
+          which leads to lock reversal.
 
 2.1.23 - Implement extension of resident files and make writing safe as well as
          many bug fixes, cleanups, and enhancements...
diff --git a/fs/ntfs/lcnalloc.c b/fs/ntfs/lcnalloc.c
index a4bc07616e5d..7b5934290685 100644
--- a/fs/ntfs/lcnalloc.c
+++ b/fs/ntfs/lcnalloc.c
@@ -54,6 +54,8 @@ int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol,
         int ret = 0;
 
         ntfs_debug("Entering.");
+        if (!rl)
+                return 0;
         for (; rl->length; rl++) {
                 int err;
 
@@ -163,17 +165,9 @@ runlist_element *ntfs_cluster_alloc(ntfs_volume *vol, const VCN start_vcn,
         BUG_ON(zone < FIRST_ZONE);
         BUG_ON(zone > LAST_ZONE);
 
-        /* Return empty runlist if @count == 0 */
-        // FIXME: Do we want to just return NULL instead? (AIA)
-        if (!count) {
-                rl = ntfs_malloc_nofs(PAGE_SIZE);
-                if (!rl)
-                        return ERR_PTR(-ENOMEM);
-                rl[0].vcn = start_vcn;
-                rl[0].lcn = LCN_RL_NOT_MAPPED;
-                rl[0].length = 0;
-                return rl;
-        }
+        /* Return NULL if @count is zero. */
+        if (!count)
+                return NULL;
         /* Take the lcnbmp lock for writing. */
         down_write(&vol->lcnbmp_lock);
         /*
@@ -788,7 +782,8 @@ out:
  * @vi:         vfs inode whose runlist describes the clusters to free
  * @start_vcn:  vcn in the runlist of @vi at which to start freeing clusters
  * @count:      number of clusters to free or -1 for all clusters
- * @is_rollback:        if TRUE this is a rollback operation
+ * @write_locked:       true if the runlist is locked for writing
+ * @is_rollback:        true if this is a rollback operation
  *
  * Free @count clusters starting at the cluster @start_vcn in the runlist
  * described by the vfs inode @vi.
@@ -806,17 +801,17 @@ out:
  * Return the number of deallocated clusters (not counting sparse ones) on
  * success and -errno on error.
  *
- * Locking: - The runlist described by @vi must be unlocked on entry and is
- *            unlocked on return.
- *          - This function takes the runlist lock of @vi for reading and
- *            sometimes for writing and sometimes modifies the runlist.
+ * Locking: - The runlist described by @vi must be locked on entry and is
+ *            locked on return.  Note if the runlist is locked for reading the
+ *            lock may be dropped and reacquired.  Note the runlist may be
+ *            modified when needed runlist fragments need to be mapped.
  *          - The volume lcn bitmap must be unlocked on entry and is unlocked
  *            on return.
  *          - This function takes the volume lcn bitmap lock for writing and
  *            modifies the bitmap contents.
  */
 s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count,
-                const BOOL is_rollback)
+                const BOOL write_locked, const BOOL is_rollback)
 {
         s64 delta, to_free, total_freed, real_freed;
         ntfs_inode *ni;
@@ -848,8 +843,7 @@ s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count,
 
         total_freed = real_freed = 0;
 
-        down_read(&ni->runlist.lock);
-        rl = ntfs_attr_find_vcn_nolock(ni, start_vcn, FALSE);
+        rl = ntfs_attr_find_vcn_nolock(ni, start_vcn, write_locked);
         if (IS_ERR(rl)) {
                 if (!is_rollback)
                         ntfs_error(vol->sb, "Failed to find first runlist "
@@ -903,7 +897,7 @@ s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count,
 
                         /* Attempt to map runlist. */
                         vcn = rl->vcn;
-                        rl = ntfs_attr_find_vcn_nolock(ni, vcn, FALSE);
+                        rl = ntfs_attr_find_vcn_nolock(ni, vcn, write_locked);
                         if (IS_ERR(rl)) {
                                 err = PTR_ERR(rl);
                                 if (!is_rollback)
@@ -950,7 +944,6 @@ s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count,
                 /* Update the total done clusters. */
                 total_freed += to_free;
         }
-        up_read(&ni->runlist.lock);
         if (likely(!is_rollback))
                 up_write(&vol->lcnbmp_lock);
 
@@ -960,7 +953,6 @@ s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn, s64 count,
         ntfs_debug("Done.");
         return real_freed;
 err_out:
-        up_read(&ni->runlist.lock);
         if (is_rollback)
                 return err;
         /* If no real clusters were freed, no need to rollback. */
@@ -973,7 +965,8 @@ err_out:
          * If rollback fails, set the volume errors flag, emit an error
          * message, and return the error code.
          */
-        delta = __ntfs_cluster_free(vi, start_vcn, total_freed, TRUE);
+        delta = __ntfs_cluster_free(vi, start_vcn, total_freed, write_locked,
+                        TRUE);
         if (delta < 0) {
                 ntfs_error(vol->sb, "Failed to rollback (error %i).  Leaving "
                                 "inconsistent metadata!  Unmount and run "
diff --git a/fs/ntfs/lcnalloc.h b/fs/ntfs/lcnalloc.h
index 4cac1c024af6..e4d7fb98d685 100644
--- a/fs/ntfs/lcnalloc.h
+++ b/fs/ntfs/lcnalloc.h
@@ -43,13 +43,14 @@ extern runlist_element *ntfs_cluster_alloc(ntfs_volume *vol,
                 const NTFS_CLUSTER_ALLOCATION_ZONES zone);
 
 extern s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn,
-                s64 count, const BOOL is_rollback);
+                s64 count, const BOOL write_locked, const BOOL is_rollback);
 
 /**
  * ntfs_cluster_free - free clusters on an ntfs volume
  * @vi:         vfs inode whose runlist describes the clusters to free
  * @start_vcn:  vcn in the runlist of @vi at which to start freeing clusters
  * @count:      number of clusters to free or -1 for all clusters
+ * @write_locked:       true if the runlist is locked for writing
  *
  * Free @count clusters starting at the cluster @start_vcn in the runlist
  * described by the vfs inode @vi.
@@ -64,19 +65,19 @@ extern s64 __ntfs_cluster_free(struct inode *vi, const VCN start_vcn,
  * Return the number of deallocated clusters (not counting sparse ones) on
  * success and -errno on error.
  *
- * Locking: - The runlist described by @vi must be unlocked on entry and is
- *            unlocked on return.
- *          - This function takes the runlist lock of @vi for reading and
- *            sometimes for writing and sometimes modifies the runlist.
+ * Locking: - The runlist described by @vi must be locked on entry and is
+ *            locked on return.  Note if the runlist is locked for reading the
+ *            lock may be dropped and reacquired.  Note the runlist may be
+ *            modified when needed runlist fragments need to be mapped.
  *          - The volume lcn bitmap must be unlocked on entry and is unlocked
  *            on return.
  *          - This function takes the volume lcn bitmap lock for writing and
  *            modifies the bitmap contents.
  */
 static inline s64 ntfs_cluster_free(struct inode *vi, const VCN start_vcn,
-                s64 count)
+                s64 count, const BOOL write_locked)
 {
-        return __ntfs_cluster_free(vi, start_vcn, count, FALSE);
+        return __ntfs_cluster_free(vi, start_vcn, count, write_locked, FALSE);
 }
 
 extern int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol,
@@ -93,8 +94,10 @@ extern int ntfs_cluster_free_from_rl_nolock(ntfs_volume *vol,
  *
  * Return 0 on success and -errno on error.
  *
- * Locking: This function takes the volume lcn bitmap lock for writing and
- *          modifies the bitmap contents.
+ * Locking: - This function takes the volume lcn bitmap lock for writing and
+ *            modifies the bitmap contents.
+ *          - The caller must have locked the runlist @rl for reading or
+ *            writing.
  */
 static inline int ntfs_cluster_free_from_rl(ntfs_volume *vol,
                 const runlist_element *rl)
diff --git a/fs/ntfs/mft.c b/fs/ntfs/mft.c
index e27651db4534..2c32b84385a8 100644
--- a/fs/ntfs/mft.c
+++ b/fs/ntfs/mft.c
@@ -1953,7 +1953,7 @@ restore_undo_alloc:
         a = ctx->attr;
         a->data.non_resident.highest_vcn = cpu_to_sle64(old_last_vcn - 1);
 undo_alloc:
-        if (ntfs_cluster_free(vol->mft_ino, old_last_vcn, -1) < 0) {
+        if (ntfs_cluster_free(vol->mft_ino, old_last_vcn, -1, TRUE) < 0) {
                 ntfs_error(vol->sb, "Failed to free clusters from mft data "
                                 "attribute.%s", es);
                 NVolSetErrors(vol);