# HG changeset patch
# User dl
# Date 1299606752 0
# Node ID f34ee71fc38a47a427d84f7a5cff6e1b49fe876a
# Parent  d1585a92e376a46da754d8e997f08597961d0cb0
6495521: (doc) ConcurrentSkipListMap links to web page NOT describing SkipList
Summary: update link, and minor implementation comment in Exchanger
Reviewed-by: chegar, mduigou

diff -r d1585a92e376 -r f34ee71fc38a jdk/src/share/classes/java/util/concurrent/ConcurrentSkipListMap.java
--- a/jdk/src/share/classes/java/util/concurrent/ConcurrentSkipListMap.java	Tue Mar 08 10:32:18 2011 +0000
+++ b/jdk/src/share/classes/java/util/concurrent/ConcurrentSkipListMap.java	Tue Mar 08 17:52:32 2011 +0000
@@ -44,8 +44,8 @@
  * creation time, depending on which constructor is used.
  *
  * <p>This class implements a concurrent variant of <a
- * href="http://www.cs.umd.edu/~pugh/">SkipLists</a> providing
- * expected average <i>log(n)</i> time cost for the
+ * href="http://en.wikipedia.org/wiki/Skip_list" target="_top">SkipLists</a>
+ * providing expected average <i>log(n)</i> time cost for the
  * <tt>containsKey</tt>, <tt>get</tt>, <tt>put</tt> and
  * <tt>remove</tt> operations and their variants.  Insertion, removal,
  * update, and access operations safely execute concurrently by
diff -r d1585a92e376 -r f34ee71fc38a jdk/src/share/classes/java/util/concurrent/Exchanger.java
--- a/jdk/src/share/classes/java/util/concurrent/Exchanger.java	Tue Mar 08 10:32:18 2011 +0000
+++ b/jdk/src/share/classes/java/util/concurrent/Exchanger.java	Tue Mar 08 17:52:32 2011 +0000
@@ -164,8 +164,8 @@
      * races between two threads or thread pre-emptions occurring
      * between reading and CASing.  Also, very transient peak
      * contention can be much higher than the average sustainable
-     * levels.  The max limit is decreased on average 50% of the times
-     * that a non-slot-zero wait elapses without being fulfilled.
+     * levels.  An attempt to decrease the max limit is usually made
+     * when a non-slot-zero wait elapses without being fulfilled.
      * Threads experiencing elapsed waits move closer to zero, so
      * eventually find existing (or future) threads even if the table
      * has been shrunk due to inactivity.  The chosen mechanics and