RFR: Backport of 6412968 and 6990419 to 7u

Poonam Bajaj poonam.bajaj at oracle.com
Mon Feb 10 21:00:06 PST 2014


Hi,

As per Jon Masamitsu's suggestion, we would be having the 
CMSParallelInitialMarkEnabled and CMSEdenChunksRecordAlways flags' value 
'false' by default for 7u so as not to introduce any behavior changes in 
7 update release. The updated webrev is here:
http://cr.openjdk.java.net/~poonam/6412968_6990419/webrev.01/

Thanks,
Poonam

On 2/9/2014 5:13 PM, Poonam Bajaj wrote:
> Hi All,
>
> Could I have some reviews for the backport of the following two fixes 
> from jdk8 to 7u:
> 6412968 <https://bugs.openjdk.java.net/browse/JDK-6412968>: CMS: Long 
> initial mark pauses
> 6990419 <https://bugs.openjdk.java.net/browse/JDK-6990419>: CMS: 
> Remaining work for 6572569: consistently skewed work distribution in 
> (long) re-mark pauses
>
> These changes parallelize the initial-mark phase of the CMS collector 
> and improve the work distribution among parallel worker threads.
> Webrev: http://cr.openjdk.java.net/~poonam/6412968_6990419/webrev.00/
>
> Original fixes:
> 6412968 : http://hg.openjdk.java.net/hsx/hsx25/hotspot/rev/ca9dedeebdec
> 6990419: http://hg.openjdk.java.net/hsx/hsx25/hotspot/rev/7b06ae405d7b
>
> Testing:
> JPRT, Customer verification testing
>
> Thanks,
> Poonam
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mail.openjdk.java.net/pipermail/hotspot-gc-dev/attachments/20140211/3a2ad8c3/attachment.html 


More information about the hotspot-gc-dev mailing list