Code Review Request: CR 6988099 Add version information to jvmti demos.
david.holmes at oracle.com
Thu Nov 3 00:50:55 UTC 2011
On 3/11/2011 6:14 AM, Daniel D. Daugherty wrote:
> On 11/2/11 1:29 PM, Kelly O'Hair wrote:
>> This change is both a build change and a serviceability change, but I
>> would argue that it is more
>> of a build change since the functionality of the jvmti demos shouldn't
>> have changed with this.
>> Although I am listed as a member of the serviceability "group", I'm
>> not exactly sure who leads that
>> group now, Daniel D. Daugherty is listed in the census but I'm not
>> sure if that is correct.
> It is not correct. I think the entire set of census entries for
> Serviceability are now stale. They need to be updated with the
> folks from the new Serviceability Team that Tuva Palm is building.
This is getting somewhat off-topic with regard to pushing changesets,
but there needs to be a formal voting process for adding/removing
members of a group or appointing a new group lead. But do not confuse
the OpenJDK Serviceability group with Oracle's internal "serviceability
team" (as I will call them). These are not the same thing and team
members need not be Serviceability Group members to work on changes in
the serviceability area.
The Serviceability group, as such, is not particularly relevant (no
offence intended!). All that counts is Projects and their Authors,
Committers and Reviewers. Regardless of whether this is a "build" change
or a "serviceability" change, the repositories are in the jdk8 forest
and so belong to the jdk8 project. Consequently it is the roles in that
project which count. The JDK8 project page makes no mention of any
Reviewer requirements, but the fact the Reviewers exist means that
formal review is necessary.
Arguably all discussion of changes intended for jdk8 should be held on
the jdk8-dev mailing list.
>> Some groups have forests for integration changes into the primary
>> releases, like jdk8 right now.
>> I am the lead on the build "group" and I can speak to that.
>> For jdk8 build changes, those being pushed into jdk8 (or the
>> http://hg.openjdk.java.net/jdk8/build forest)
>> only one reviewer is generally required, I have not heard anything
>> about more than one, although more
>> is always better, especially if there is any paranoia over the changes.
>> In this case, and with most build changes, someone should verify that
>> all platforms still build ok.
>> It will be important that you create the changeset using the latest
>> jdk8 build
>> forest, e.g. using a clones from http://hg.openjdk.java.net/jdk8/build
>> Once you are happy that the change is ok on your system, then bundle
>> up the changeset
>> with 'hg bundle' and I will run it through our internal JPRT build
>> system to insure all build
>> platforms are ok, then I can push it into the
>> http://hg.openjdk.java.net/jdk8/build repos.
>> On Nov 2, 2011, at 11:54 AM, Neil Richards wrote:
>>> On Wed, 2011-11-02 at 14:07 -0400, Zhengyu Gu wrote:
>>>> Hi Neil,
>>>> My understanding is that you need to have at least two reviewers who
>>>> have 'reviewer' privilege, (and I don't have 'reviewer' privilege).
>>> I had a quick leaf through the OpenJDK Bylaws.
>>> It looks like the number of necessary approvals from 'reviewers' vs
>>> 'committers' (vs anyone else) is project-specific, so it isn't directly
>>> defined there.
>>> Now you raise it, I wasn't able to track down where in the JDK 8 project
>>> documentation this is defined either.
>>> The log of (at least) the jdk repository suggests that this limit isn't
>>> universally applied (though in many cases, the submitter is of
>>> 'reviewer' status themselves - perhaps that counts towards the total ?).
>>> Kelly, can you give guidance / pointers here ?
>>> (So far, this change has found favor by one 'reviewer' - you - and two
>>> 'committers' - Zhengyu and myself).
>>>> Yes, I am tied up, if you can handle the push, that will be great. What
>>>> java bug id you need? can you use CR 6988099?
>>> Doh! Of course I can.
>>> (I blame it all on coding-blindness - that's the same as snow-blindness,
>>> but in hex).
>>> Regards, Neil
>>> Unless stated above:
>>> IBM email: neil_richards at uk.ibm.com
>>> IBM United Kingdom Limited - Registered in England and Wales with
>>> number 741598.
>>> Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire
>>> PO6 3AU
More information about the build-dev