OSDN Git Service

xfs: update ctime and remove suid before cloning files
authorDarrick J. Wong <darrick.wong@oracle.com>
Fri, 5 Oct 2018 09:05:41 +0000 (19:05 +1000)
committerDave Chinner <david@fromorbit.com>
Fri, 5 Oct 2018 09:05:41 +0000 (19:05 +1000)
Before cloning into a file, update the ctime and remove sensitive
attributes like suid, just like we'd do for a regular file write.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Dave Chinner <david@fromorbit.com>
fs/xfs/xfs_reflink.c

index f135748..59da970 100644 (file)
@@ -1264,6 +1264,7 @@ xfs_reflink_zero_posteof(
  * Prepare two files for range cloning.  Upon a successful return both inodes
  * will have the iolock and mmaplock held, the page cache of the out file
  * will be truncated, and any leases on the out file will have been broken.
+ * This function borrows heavily from xfs_file_aio_write_checks.
  */
 STATIC int
 xfs_reflink_remap_prep(
@@ -1327,6 +1328,30 @@ xfs_reflink_remap_prep(
        /* Zap any page cache for the destination file's range. */
        truncate_inode_pages_range(&inode_out->i_data, pos_out,
                                   PAGE_ALIGN(pos_out + *len) - 1);
+
+       /* If we're altering the file contents... */
+       if (!is_dedupe) {
+               /*
+                * ...update the timestamps (which will grab the ilock again
+                * from xfs_fs_dirty_inode, so we have to call it before we
+                * take the ilock).
+                */
+               if (!(file_out->f_mode & FMODE_NOCMTIME)) {
+                       ret = file_update_time(file_out);
+                       if (ret)
+                               goto out_unlock;
+               }
+
+               /*
+                * ...clear the security bits if the process is not being run
+                * by root.  This keeps people from modifying setuid and setgid
+                * binaries.
+                */
+               ret = file_remove_privs(file_out);
+               if (ret)
+                       goto out_unlock;
+       }
+
        return 1;
 out_unlock:
        xfs_reflink_remap_unlock(file_in, file_out);