Bugids already used in this repository (Re: [FOR REVIEW] hs14 merge for OpenJDK6)

Thu Aug 6 17:49:15 PDT 2009

Andrew John Hughes wrote:
> 2009/8/5 Tim Bell <Tim.Bell at sun.com>:
>   
>> Andrew John Hughes wrote:
>>
>>     
>>> Thanks Tim, that seems to have fixed the permissions issue.
>>>       
>> Good.
>>
>>     
>>> There now seems to be an issue with jcheck, as the merge causes some
>>> bugids to be repeated:
>>>       
>> Hmmm - how did these changesets come in through two different paths?
>>
>>     
>>> pushing to ssh://hg.openjdk.java.net/jdk6/jdk6-gate/hotspot
>>> searching for changes
>>> remote: adding changesets
>>> remote: adding manifests
>>> remote: adding file changes
>>> remote: added 555 changesets with 4771 changes to 1453 files
>>> remote:
>>> remote: > Changeset: 100:d821d920b465
>>> remote: > Author:    kvn
>>> remote: > Date:      2008-03-11 11:04
>>> remote: >
>>> remote: > 6623167: C2 crashed in StoreCMNode::Value
>>> remote: > Summary: C2 crashed in StoreCMNode::Value because
>>> n->in(MemNode::OopStore) is 0.
>>> remote: > Reviewed-by: rasbold, never
>>> remote:
>>> remote: Bugid 6623167 already used in this repository, in revision 20
>>> remote:
>>> remote: > Changeset: 104:2c106685d6d0
>>> remote: > Author:    dcubed
>>> remote: > Date:      2008-03-12 18:06
>>> remote: >
>>> remote: > 6497639: 4/3 Profiling Swing application caused JVM crash
>>> remote: > Summary: Make RedefineClasses() interoperate better with
>>> class sharing.
>>> remote: > Reviewed-by: sspitsyn, jmasa
>>> remote:
>>> remote: Bugid 6497639 already used in this repository, in revision 20
>>> remote:
>>> remote: > Changeset: 105:d8b3ef7ee3e5
>>> remote: > Author:    dcubed
>>> remote: > Date:      2008-03-12 18:07
>>> remote: >
>>> remote: > 6599425: 4/3 OopMapCache::lookup() can cause later crash or
>>> assert() failure
>>> remote: > Summary: Add should_not_be_cached() to markOop and methodOop
>>> and query that status inOopMapCache::lookup()
>>> remote: > Reviewed-by: coleenp, sspitsyn, jmasa
>>> remote:
>>> remote: Bugid 6599425 already used in this repository, in revision 20
>>> remote:
>>> remote: > Changeset: 240:65fe2bd88839
>>> remote: > Author:    never
>>> remote: > Date:      2008-06-05 21:44
>>> remote: >
>>> remote: > 6614100: EXCEPTION_ACCESS_VIOLATION while running Eclipse
>>> with 1.6.0_05-ea
>>> remote: > Reviewed-by: kvn, jrose, rasbold
>>> remote:
>>> remote: Bugid 6614100 already used in this repository, in revision 20
>>> remote:
>>> remote: > Changeset: 286:3e82d72933d0
>>> remote: > Author:    xlu
>>> remote: > Date:      2008-06-26 14:15
>>> remote: >
>>> remote: > 6718830: Hotspot fails to build with gcc 4.3
>>> remote: > Summary: Fixed linux make file and couple adlc code to meet
>>> the changes of gcc 4.3
>>> remote: > Reviewed-by: kamg, igor
>>> remote:
>>> remote: Bugid 6718830 already used in this repository, in revision 32
>>> remote:
>>> remote: > Changeset: 289:551f4309f476
>>> remote: > Author:    ohair
>>> remote: > Date:      2008-07-03 10:46
>>> remote: >
>>> remote: > 6695777: Queens.class should be built from source, not put
>>> in source repo
>>> remote: > Reviewed-by: kvn
>>> remote:
>>> remote: Bugid 6695777 already used in this repository, in revision 20
>>> remote:
>>> remote: > Changeset: 314:54499b980c23
>>> remote: > Author:    swamyv
>>> remote: > Date:      2008-07-29 13:54
>>> remote: >
>>> remote: > 6710791: Remove files or build from source:maf-1_0.jar, jlfg-1_0.jar
>>> remote: > Summary: Removed maf-1_0.jar and jlfg-1_0.jar files.
>>> remote: > Reviewed-by: poonam, jjh
>>> remote:
>>> remote: Bugid 6710791 already used in this repository, in revision 20
>>> remote:
>>> remote: > Changeset: 360:fa4d1d240383
>>> remote: > Author:    never
>>> remote: > Date:      2008-08-26 15:49
>>> remote: >
>>> remote: > 6741642: bad enum definition in ciTypeFlow.hpp
>>> remote: > Reviewed-by: rasbold, martin
>>> remote: > Contributed-by: doko at ubuntu.com
>>> remote:
>>> remote: Bugid 6741642 already used in this repository, in revision 22
>>> remote:
>>> remote: > Changeset: 589:748572b86af6
>>> remote: > Author:    never
>>> remote: > Date:      2009-04-07 14:46
>>> remote: >
>>> remote: > 6636360: compiler/6595044/Main.java test fails with 64bit
>>> java on solaris-sparcv9 with SIGSEGV
>>> remote: > Reviewed-by: kvn, twisti
>>> remote:
>>> remote: Bugid 6636360 already used in this repository, in revision 29
>>> remote:
>>> remote: abort: pretxnchangegroup.0.jcheck hook failed
>>> remote: transaction abort!
>>> remote: rollback completed
>>> abort: unexpected response: empty string
>>>
>>> Is there a way of getting it to ignore these for this one push?  I
>>> don't know of a way to just pull out these nine changesets from the
>>> 555 waiting to go...
>>>       
>> I think they would need to be added to the jcheck whitelist for all time-
>> but that is more of a question for Mark or Kelly.
>>
>> Tim
>>
>>     
>
> The early revisions (20, 32) are from OpenJDK6 which was rebased to
> allow HotSpot patches to be applied on top.  So what happened here is
> that, as the fixes were already applied to OpenJDK6, the new
> changesets pulled in by the merge will be no-ops but still exist to
> keep the change history accurate.
>
> I think we just need a way of disabling this check.  OpenJDK6 already
> has whitespace and comment checks turned to lax.  See .jcheck:
> http://hg.openjdk.java.net/jdk6/jdk6/hotspot/rev/0282bf49b0f6.  There
> may already be an option for this bugid check too, but I have no idea
> what the format of that file is.
>   

*grumble*

I'll talk to Mark about this when he is available to discuss the best 
way to get these changes back.

-Joe