This test is failing in a way that seems unrelated to what it is trying to test:

/opt/jprt/T/P1/121941.tbell/s/jdk/test/java/math/BigInteger/BigIntegerTest.java:292: error: unmappable character for encoding ASCII
 ????????* ensures that {@code v} is just under the B-Z threshold, that {@code z} is
 ^
/opt/jprt/T/P1/121941.tbell/s/jdk/test/java/math/BigInteger/BigIntegerTest.java:292: error: unmappable character for encoding ASCII
 ????????* ensures that {@code v} is just under the B-Z threshold, that {@code z} is
  ^

[...] - refer to the attached test log file: solaris_x64-product-c2-jdk_math.log.FAILED.log


When I look at the .java file I see non-whitespace chars in a comment block, and this seems to be causing the headache:

cat -n test/java/math/BigInteger/BigIntegerTest.java | less
 [... snip ...]
   291 * {@code q1 = q2*pow(2,a-b)} and {@code r1 = r2*pow(2,b)}. The test
   292     * ensures that {@code v} is just under the B-Z threshold, that {@code z} is
   293     * over the threshold and {@code w} is much larger than {@code z}. This
   294     * implies that {@code u/v} uses the standard division algorithm and
   295     * {@code w/z} uses the B-Z algorithm.  The results of the two algorithms
   296     * are then compared using the observation described in the foregoing and
   297     * if they are not equal a failure is logged.