tar: document xgetcwd test case better

* src/misc.c (normalize_filename): Add commentary for clarity.
* tests/extrac09.at: Retitle test case and add comments for clarity.

src/misc.c

@@ -288,7 +288,8 @@ normalize_filename (int cdidx, const char *name)
          this following approach may lead to situations where the same
          file or directory is processed twice under different absolute
          paths without that duplication being detected.  Perhaps we
-         should use dev+ino pairs instead of names?  */
+         should use dev+ino pairs instead of names?  (See listed03.at for
+         a related test case.) */
       const char *cdpath = tar_getcdpath (cdidx);
       size_t copylen;
       bool need_separator;

tests/extrac09.at

@@ -18,10 +18,20 @@
 # You should have received a copy of the GNU General Public License
 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
-# This checks for the --listed-incremental bug reported by J Chapman Flack at
-# http://lists.gnu.org/archive/html/bug-tar/2010-06/msg00000.html
+# This attempts to cause xgetcwd() to fail, and then checks to see if
+# such failure causes tar to abort even in a case where the results of
+# the call aren't actually needed.
+#
+# (xgetcwd() may fail e.g. on Solaris 10 when "." or ".." are unreadable.
+# On most systems xgetcwd() won't fail even in that situation, but
+# on those systems this test will simply succeed without actually testing
+# anything within tar.)
+#
+# http://lists.gnu.org/archive/html/bug-tar/2010-07/msg00045.html
+#
+# (See also 'listed03.at'.)
 
-AT_SETUP([no need to save dir with unreadable . and ..])
+AT_SETUP([extracting even when . and .. are unreadable])
 AT_KEYWORDS([extract extrac09])
 
 AT_TAR_CHECK([