hg: jdk7u/jdk7u-dev/jdk: 7088989: Improve the performance for T4 by utilizing the newly provided crypto APIs (original) (raw)

Valerie (Yu-Ching) Peng valerie.peng at oracle.com
Tue Feb 7 17:09:20 PST 2012

When I inspected make/closed, there are no directories containing Makefile. So, I still placed the build changes under "make" like all other components currently have, and it will be skipped for non-OpenJDK builds. If I move the build changes to make/closed, will it be built automatically if I start the build at the top level? I doubt it will be done with the current make mechanism? Valerie

On 02/07/12 16:51, David Holmes wrote:

Valerie,

I also have a concern here that if the source files for com/oracle/security/ucrupto are in src/closed then why weren't the bulk of the build changes in make/closed? David On 8/02/2012 9:41 AM, Valerie (Yu-Ching) Peng wrote:

I wasn't aware that explicit approval is needed for backporting a JDK8 fix into JDK7 update. I'll check w/ my co-workers and manager and see what I should do in this scenario. Thanks, Valerie On 02/07/12 13:47, Omair Majid wrote: On 02/07/2012 03:18 PM, Valerie (Yu-Ching) Peng wrote: This was discussed in JDK8 and then later backported to 7u. The majority of the code is under the closed tree, so the changes are reviewed internally. Ah, I was not aware of that. From what I have seen of other patches, the external changes are still reviewed externally: http://mail.openjdk.java.net/pipermail/hotspot-dev/2011-December/004805.html

Besides, I am concerned that Rule 4 of the JDK 7 Updates: Ground Rules [1] was violated and that this changeset was not approved publicly on the jdk7u-dev mailing list (nor was a url to the place where the patch was reviewed provided). Perhaps the rules need to be adjusted to clarify how cases like this should be handled? Cheers, Omair [1] http://openjdk.java.net/projects/jdk7u/groundrules.html

More information about the jdk7u-dev mailing list