]> www.pilppa.org Git - linux-2.6-omap-h63xx.git/blobdiff - fs/ocfs2/file.c
ocfs2: Support creation of unwritten extents
[linux-2.6-omap-h63xx.git] / fs / ocfs2 / file.c
index 6745086da6fdae431cc35035d3371383cf223c89..3e21ad9a6dde43bf3aad4970712b307345e2702f 100644 (file)
@@ -425,6 +425,7 @@ int ocfs2_do_extend_allocation(struct ocfs2_super *osb,
                               struct inode *inode,
                               u32 *logical_offset,
                               u32 clusters_to_add,
+                              int mark_unwritten,
                               struct buffer_head *fe_bh,
                               handle_t *handle,
                               struct ocfs2_alloc_context *data_ac,
@@ -437,9 +438,13 @@ int ocfs2_do_extend_allocation(struct ocfs2_super *osb,
        enum ocfs2_alloc_restarted reason = RESTART_NONE;
        u32 bit_off, num_bits;
        u64 block;
+       u8 flags = 0;
 
        BUG_ON(!clusters_to_add);
 
+       if (mark_unwritten)
+               flags = OCFS2_EXT_UNWRITTEN;
+
        free_extents = ocfs2_num_free_extents(osb, inode, fe);
        if (free_extents < 0) {
                status = free_extents;
@@ -489,7 +494,7 @@ int ocfs2_do_extend_allocation(struct ocfs2_super *osb,
             num_bits, bit_off, (unsigned long long)OCFS2_I(inode)->ip_blkno);
        status = ocfs2_insert_extent(osb, handle, inode, fe_bh,
                                     *logical_offset, block, num_bits,
-                                    meta_ac);
+                                    flags, meta_ac);
        if (status < 0) {
                mlog_errno(status);
                goto leave;
@@ -522,9 +527,11 @@ leave:
  * For a given allocation, determine which allocators will need to be
  * accessed, and lock them, reserving the appropriate number of bits.
  *
- * Called from ocfs2_extend_allocation() for file systems which don't
- * support holes, and from ocfs2_write() for file systems which
- * understand sparse inodes.
+ * Sparse file systems call this from ocfs2_write_begin_nolock()
+ * and ocfs2_allocate_unwritten_extents().
+ *
+ * File systems which don't support holes call this from
+ * ocfs2_extend_allocation().
  */
 int ocfs2_lock_allocators(struct inode *inode, struct ocfs2_dinode *di,
                          u32 clusters_to_add, u32 extents_to_split,
@@ -595,14 +602,13 @@ out:
        return ret;
 }
 
-static int ocfs2_extend_allocation(struct inode *inode,
-                                  u32 clusters_to_add)
+static int __ocfs2_extend_allocation(struct inode *inode, u32 logical_start,
+                                    u32 clusters_to_add, int mark_unwritten)
 {
        int status = 0;
        int restart_func = 0;
-       int drop_alloc_sem = 0;
        int credits;
-       u32 prev_clusters, logical_start;
+       u32 prev_clusters;
        struct buffer_head *bh = NULL;
        struct ocfs2_dinode *fe = NULL;
        handle_t *handle = NULL;
@@ -617,7 +623,7 @@ static int ocfs2_extend_allocation(struct inode *inode,
         * This function only exists for file systems which don't
         * support holes.
         */
-       BUG_ON(ocfs2_sparse_alloc(osb));
+       BUG_ON(mark_unwritten && !ocfs2_sparse_alloc(osb));
 
        status = ocfs2_read_block(osb, OCFS2_I(inode)->ip_blkno, &bh,
                                  OCFS2_BH_CACHED, inode);
@@ -633,18 +639,9 @@ static int ocfs2_extend_allocation(struct inode *inode,
                goto leave;
        }
 
-       logical_start = OCFS2_I(inode)->ip_clusters;
-
 restart_all:
        BUG_ON(le32_to_cpu(fe->i_clusters) != OCFS2_I(inode)->ip_clusters);
 
-       /* blocks peope in read/write from reading our allocation
-        * until we're done changing it. We depend on i_mutex to block
-        * other extend/truncate calls while we're here. Ordering wrt
-        * start_trans is important here -- always do it before! */
-       down_write(&OCFS2_I(inode)->ip_alloc_sem);
-       drop_alloc_sem = 1;
-
        status = ocfs2_lock_allocators(inode, fe, clusters_to_add, 0, &data_ac,
                                       &meta_ac);
        if (status) {
@@ -678,6 +675,7 @@ restarted_transaction:
                                            inode,
                                            &logical_start,
                                            clusters_to_add,
+                                           mark_unwritten,
                                            bh,
                                            handle,
                                            data_ac,
@@ -730,10 +728,6 @@ restarted_transaction:
             OCFS2_I(inode)->ip_clusters, i_size_read(inode));
 
 leave:
-       if (drop_alloc_sem) {
-               up_write(&OCFS2_I(inode)->ip_alloc_sem);
-               drop_alloc_sem = 0;
-       }
        if (handle) {
                ocfs2_commit_trans(osb, handle);
                handle = NULL;
@@ -759,6 +753,25 @@ leave:
        return status;
 }
 
+static int ocfs2_extend_allocation(struct inode *inode, u32 logical_start,
+                                  u32 clusters_to_add, int mark_unwritten)
+{
+       int ret;
+
+       /*
+        * The alloc sem blocks peope in read/write from reading our
+        * allocation until we're done changing it. We depend on
+        * i_mutex to block other extend/truncate calls while we're
+        * here.
+        */
+       down_write(&OCFS2_I(inode)->ip_alloc_sem);
+       ret = __ocfs2_extend_allocation(inode, logical_start, clusters_to_add,
+                                       mark_unwritten);
+       up_write(&OCFS2_I(inode)->ip_alloc_sem);
+
+       return ret;
+}
+
 /* Some parts of this taken from generic_cont_expand, which turned out
  * to be too fragile to do exactly what we need without us having to
  * worry about recursive locking in ->prepare_write() and
@@ -900,7 +913,9 @@ static int ocfs2_extend_file(struct inode *inode,
        }
 
        if (clusters_to_add) {
-               ret = ocfs2_extend_allocation(inode, clusters_to_add);
+               ret = ocfs2_extend_allocation(inode,
+                                             OCFS2_I(inode)->ip_clusters,
+                                             clusters_to_add, 0);
                if (ret < 0) {
                        mlog_errno(ret);
                        goto out_unlock;
@@ -1176,6 +1191,64 @@ out:
        return ret;
 }
 
+/*
+ * Allocate enough extents to cover the region starting at byte offset
+ * start for len bytes. Existing extents are skipped, any extents
+ * added are marked as "unwritten".
+ */
+static int ocfs2_allocate_unwritten_extents(struct inode *inode,
+                                           u64 start, u64 len)
+{
+       int ret;
+       u32 cpos, phys_cpos, clusters, alloc_size;
+
+       /*
+        * We consider both start and len to be inclusive.
+        */
+       cpos = start >> OCFS2_SB(inode->i_sb)->s_clustersize_bits;
+       clusters = ocfs2_clusters_for_bytes(inode->i_sb, start + len);
+       clusters -= cpos;
+
+       while (clusters) {
+               ret = ocfs2_get_clusters(inode, cpos, &phys_cpos,
+                                        &alloc_size, NULL);
+               if (ret) {
+                       mlog_errno(ret);
+                       goto out;
+               }
+
+               /*
+                * Hole or existing extent len can be arbitrary, so
+                * cap it to our own allocation request.
+                */
+               if (alloc_size > clusters)
+                       alloc_size = clusters;
+
+               if (phys_cpos) {
+                       /*
+                        * We already have an allocation at this
+                        * region so we can safely skip it.
+                        */
+                       goto next;
+               }
+
+               ret = __ocfs2_extend_allocation(inode, cpos, alloc_size, 1);
+               if (ret) {
+                       if (ret != -ENOSPC)
+                               mlog_errno(ret);
+                       goto out;
+               }
+
+next:
+               cpos += alloc_size;
+               clusters -= alloc_size;
+       }
+
+       ret = 0;
+out:
+       return ret;
+}
+
 static int ocfs2_prepare_inode_for_write(struct dentry *dentry,
                                         loff_t *ppos,
                                         size_t count,