[PATCH FOR REVIEW] 7036559: ConcurrentHashMap footprint and contention improvements

[Andrew Hughes]

----- Original Message -----
> On 05/31/2013 05:38 PM, Andrew Hughes wrote:
> > This fix was determined to be a prerequisite backport for the 2013-04
> > security fixes.
> > 
> > http://cr.openjdk.java.net/~andrew/jdk6/7036559/
> 
> I did a diff between the result of this patch and the file in jdk7u
> after revision 005c0c85b0de and it looks identical, except for minor
> changes:
> 
> > --- jdk7u/jdk/src/share/classes/java/util/concurrent/ConcurrentHashMap.java
> > 	2013-05-31 18:32:28.179433269 -0400
> > +++ temp/ConcurrentHashMap.java	2013-05-31 17:34:06.000000000 -0400
> > @@ -30,7 +30,7 @@
> >   *
> >   * Written by Doug Lea with assistance from members of JCP JSR-166
> >   * Expert Group and released to the public domain, as explained at
> > - * http://creativecommons.org/publicdomain/zero/1.0/
> > + * http://creativecommons.org/licenses/publicdomain
> >   */
> >  
> >  package java.util.concurrent;
> > @@ -1412,7 +1412,7 @@
> >       * for each key-value mapping, followed by a null pair.
> >       * The key-value mappings are emitted in no particular order.
> >       */
> > -    private void writeObject(java.io.ObjectOutputStream s) throws
> > IOException {
> > +    private void writeObject(java.io.ObjectOutputStream s) throws
> > IOException  {
> >          // force all segments for serialization compatibility
> >          for (int k = 0; k < segments.length; ++k)
> >              ensureSegment(k);
> > @@ -1446,7 +1446,7 @@
> >       */
> >      @SuppressWarnings("unchecked")
> >      private void readObject(java.io.ObjectInputStream s)
> > -        throws IOException, ClassNotFoundException {
> > +        throws IOException, ClassNotFoundException  {
> >          s.defaultReadObject();
> >  
> >          // Re-initialize segments to be minimally sized, and let grow.
> 

These are:

changeset:   3984:31619dfa6a4a
user:        dl
date:        Thu Apr 07 15:06:32 2011 +0100
summary:     7034657: Update Creative Commons license URL in legal notices

changeset:   3203:9e494de19690
user:        dl
date:        Wed Dec 01 21:46:52 2010 +0000
summary:     7003745: Code style cleanups (sync from Dougs CVS)

> I don't see any API changes in the webrev; that's good.
> 

Well, these changes have already been shipped in IcedTea and we've had builds
in Fedora and RHEL pass the TCK.  This is just a matter of syncing them to
OpenJDK 6.  We'll not be changing them in the process, as it would affect our
existing results.

> Do you also intend to backport the following?
> 
> changeset:   4110:e9ec52c63a9f
> user:        dl
> date:        Thu Apr 21 17:00:23 2011 +0100
> summary:     7038542: Small performance regression in ConcurrentHashMap
> on c1 since CR 703655
> 

Good spot.

We can consider this after a security update is released (which will probably
be after the June one is in too now).

We can look at including it in IcedTea first to give it some testing.

> Thanks,
> Omair
> --
> PGP Key: 66484681 (http://pgp.mit.edu/)
> Fingerprint = F072 555B 0A17 3957 4E95  0056 F286 F14F 6648 4681
> 

-- 
Andrew :)

Free Java Software Engineer
Red Hat, Inc. (http://www.redhat.com)

PGP Key: 248BDC07 (https://keys.indymedia.org/)
Fingerprint = EC5A 1F5E C0AD 1D15 8F1F  8F91 3B96 A578 248B DC07