RFR(XS): 8061785: [TEST_BUG] serviceability/sa/jmap-hashcode/Test8028623.java has utf8 character corrupted by earlier merge

KEVIN WALLS kevin.walls at oracle.com
Wed Dec 3 22:06:02 UTC 2014


Hi,

This is a review request for a changing a couple of characters in a 
test.  It's the same test I just changed with another review, I kept it 
separate so as to complicate the review already in progress (it had gone 
on long enough..).


The test contains an embedded utf8 character, to test a jmap/SA heap 
dumping fix.  The test is not broken here in 9/hs-rt, but in another 
clone I saw this character get corrupted by a merge: it was rewritten as 
literal ?? question marks.  It wasn't clear if it was an hg issue (I 
couldn't reproduce that) or perhaps more likely the editor used to do a 
merge.

To avoid the same thing happening again we can change the utf8 char to 
\uXXXX format.

The change below I tested to check the test still correctly fails on a 
non-fixed JVM, and passes on a fixed one

bug:
https://bugs.openjdk.java.net/browse/JDK-8061785

change:
$ hg diff  test/serviceability/sa/jmap-hashcode/Test8028623.java
diff --git a/test/serviceability/sa/jmap-hashcode/Test8028623.java 
b/test/serviceability/sa/jmap-hashcode/Test8028623.java
--- a/test/serviceability/sa/jmap-hashcode/Test8028623.java
+++ b/test/serviceability/sa/jmap-hashcode/Test8028623.java
@@ -41,12 +41,12 @@

  public class Test8028623 {

-  public static int à = 1;
+  public static int \u00CB = 1;
    public static String dumpFile = "heap.out";

    public static void main (String[] args) {

-    System.out.println(Ã);
+    System.out.println(\u00CB);

      try {
          if (!Platform.shouldSAAttach()) {

---------------------------------------------------------------------------------------------- 

Thanks
Kevin



More information about the serviceability-dev mailing list