diff --git a/libvmfs/vmfs_file.c b/libvmfs/vmfs_file.c
index 04f0698..62b2fa7 100644
--- a/libvmfs/vmfs_file.c
+++ b/libvmfs/vmfs_file.c
@@ -205,6 +205,11 @@ ssize_t vmfs_file_pread(vmfs_file_t *f,u_char *buf,size_t len,off_t pos)
             break;
          }
 
+         /* Pointer-Block, if we get here we failed to resolve pointer */
+         case VMFS_BLK_TYPE_PB:
+            fprintf(stderr, "VMFS: Pointer block at id 0x%8.8x\n", blk_id);
+            return(-EINVAL);
+
          /* Inline in the inode */
          case VMFS_BLK_TYPE_FD:
             if (blk_id == f->inode->id) {
diff --git a/libvmfs/vmfs_inode.c b/libvmfs/vmfs_inode.c
index 0596064..c6eaa9d 100644
--- a/libvmfs/vmfs_inode.c
+++ b/libvmfs/vmfs_inode.c
@@ -311,36 +311,20 @@ int vmfs_inode_get_block(const vmfs_inode_t *inode,off_t pos,uint32_t *blk_id)
 
       case VMFS_BLK_TYPE_PB:
       {
-         DECL_ALIGNED_BUFFER_WOL(buf,fs->pbc->bmh.data_size);
-         uint32_t pb_blk_id;
-         uint32_t blk_per_pb;
-         u_int pb_index;
-         u_int sub_index;
-
-         blk_per_pb = fs->pbc->bmh.data_size / sizeof(uint32_t);
-         blk_index = pos / inode->blk_size;
-
-         pb_index  = blk_index / blk_per_pb;
-         sub_index = blk_index % blk_per_pb;
-
-         if (pb_index >= VMFS_INODE_BLK_COUNT)
-            return(-EINVAL);
-
-         pb_blk_id = inode->blocks[pb_index];
-
-         if (!pb_blk_id)
-            break;
-
-         if (!vmfs_bitmap_get_item(fs->pbc,
-                                   VMFS_BLK_PB_ENTRY(pb_blk_id),
-                                   VMFS_BLK_PB_ITEM(pb_blk_id),
-                                   buf))
-            return(-EIO);
-
-         *blk_id = read_le32(buf,sub_index*sizeof(uint32_t));
+         int pointer_result = -EINVAL;
+         if (inode->blk_size == VMFS_DP_BLOCK_SZ &&
+                              inode->size > VMFS_SP_MAX_SZ) {
+            pointer_result = vmfs_dp_get_block(fs, inode, pos, blk_id);
+         } else {
+            pointer_result = vmfs_sp_get_block(fs, inode, pos, blk_id);
+         }
+         if (pointer_result) {
+            fprintf(stderr, "Error in resolving pointer block: %d\n",
+                              pointer_result);
+            return(pointer_result);
+         }
          break;
       }
-
       case VMFS_BLK_TYPE_FD:
          if (vmfs5_extension) {
             *blk_id = inode->id;
@@ -354,6 +338,84 @@ int vmfs_inode_get_block(const vmfs_inode_t *inode,off_t pos,uint32_t *blk_id)
    return(0);
 }
 
+/* read pointer to get block id */
+int vmfs_sp_get_block(const vmfs_fs_t *fs, const vmfs_inode_t *inode, off_t pos,
+                              uint32_t *blk_id) {
+   DECL_ALIGNED_BUFFER_WOL(buf,fs->pbc->bmh.data_size);
+   uint32_t pb_blk_id;
+   uint32_t blk_per_pb;
+   u_int pb_index;
+   u_int sub_index;
+   u_int blk_index;
+
+   blk_per_pb = fs->pbc->bmh.data_size / sizeof(uint32_t);
+   blk_index = pos / inode->blk_size;
+
+   pb_index  = blk_index / blk_per_pb;
+   sub_index = blk_index % blk_per_pb;
+
+   if (pb_index >= VMFS_INODE_BLK_COUNT)
+      return(-EINVAL);
+
+   pb_blk_id = inode->blocks[pb_index];
+
+   if (!pb_blk_id) {
+      return(0);
+   }
+
+   if (!vmfs_bitmap_get_item(fs->pbc,
+                              VMFS_BLK_PB_ENTRY(pb_blk_id),
+                              VMFS_BLK_PB_ITEM(pb_blk_id),
+                              buf)) {
+      return(-EIO);
+   }
+
+   *blk_id = read_le32(buf,sub_index*sizeof(uint32_t));
+
+   return(0);
+}
+
+/* read double pointer to get block */
+int vmfs_dp_get_block(const vmfs_fs_t *fs, const vmfs_inode_t *inode, off_t pos,
+                              uint32_t *blk_id) {
+   DECL_ALIGNED_BUFFER_WOL(buf,fs->pbc->bmh.data_size);
+   uint32_t pb_blk_id;
+   uint32_t di_blk_id;
+   uint32_t blk_per_pb;
+   u_int pb_index;
+   u_int sub_index;
+   u_int blk_index;
+   u_int di_pb_index;
+
+   blk_per_pb = fs->pbc->bmh.data_size / sizeof(uint32_t);
+   blk_index = pos / inode->blk_size;
+   pb_index = blk_index / (blk_per_pb * blk_per_pb);
+   di_pb_index = (blk_index / blk_per_pb) % blk_per_pb;
+   sub_index = blk_index % blk_per_pb;
+   pb_blk_id = inode->blocks[pb_index];
+
+   if (!pb_blk_id) {
+      return(0);
+   }
+   // read first pointer block
+   if (!vmfs_bitmap_get_item(fs->pbc,
+                              VMFS_BLK_PB_ENTRY(pb_blk_id),
+                              VMFS_BLK_PB_ITEM(pb_blk_id),
+                              buf)) {
+      return(-EIO);
+   }
+   di_blk_id = read_le32(buf,di_pb_index*sizeof(uint32_t));
+   // read second pointer block
+   if (!vmfs_bitmap_get_item(fs->pbc,
+                              VMFS_BLK_PB_ENTRY(di_blk_id),
+                              VMFS_BLK_PB_ITEM(di_blk_id),
+                              buf)) {
+      return(-EIO);
+   }
+   *blk_id = read_le32(buf, sub_index*sizeof(uint32_t));
+   return(0);
+}
+
 /* Aggregate a sub-block to a file block */
 static int vmfs_inode_aggregate_fb(vmfs_inode_t *inode)
 {
diff --git a/libvmfs/vmfs_inode.h b/libvmfs/vmfs_inode.h
index 14f303c..560fa52 100644
--- a/libvmfs/vmfs_inode.h
+++ b/libvmfs/vmfs_inode.h
@@ -26,6 +26,9 @@
 
 #define VMFS_INODE_MAGIC  0x10c00001
 
+#define VMFS_DP_BLOCK_SZ 0x100000
+#define VMFS_SP_MAX_SZ 0x4000000000
+
 struct vmfs_inode_raw {
    struct vmfs_metadata_hdr_raw mdh;
    uint32_t id;
@@ -136,6 +139,12 @@ int vmfs_inode_alloc(vmfs_fs_t *fs,u_int type,mode_t mode,vmfs_inode_t **inode);
  */
 int vmfs_inode_get_block(const vmfs_inode_t *inode,off_t pos,uint32_t *blk_id);
 
+/* Get block id from pointer block */
+int vmfs_sp_get_block(const vmfs_fs_t *fs, const vmfs_inode_t *inode, off_t pos, uint32_t *blk_id);
+
+/* Get block id from double indirect pointer block */
+int vmfs_dp_get_block(const vmfs_fs_t *fs, const vmfs_inode_t *inode, off_t pos, uint32_t *blk_id);
+
 /* Get a block for writing corresponding to the specified position */
 int vmfs_inode_get_wrblock(vmfs_inode_t *inode,off_t pos,uint32_t *blk_id);