RFR(M) : 8010124 : JVM_GetClassContext: use GrowableArray instead of KlassLink

[Christian Thalinger]

On Jun 13, 2013, at 11:44 AM, David Chase <david.r.chase at oracle.com> wrote:

> So where do we stand on this?  Do I need to prepare a patch and forward it to Vladimir K?

Right.  You can also send me the exported changeset.

> If so, is it a dual contributed by (twisti & drchase)?

Maybe have me as contributor and you as author.

-- Chris

> 
> David
> 
> On 2013-06-11, at 4:21 PM, David Chase <david.r.chase at oracle.com> wrote:
>> On 2013-06-11, at 2:23 PM, Christian Thalinger <christian.thalinger at oracle.com> wrote:
>> 
>>> Looks good.  For testing, the only real user of this method is SecurityManager.checkMemberAccess where the access type is not public.
>> 
>> Off in jdk land, I did (jtreg) all of
>> 
>> closed/java/lang/reflect
>> java/lang/reflect
>> closed/javax/management
>> java/security
>> closed/java/security
>> 
>> plus a couple of others that included uses of getField.
>> 
>>> Thanks for pushing that cleanup.
>>> 
>>> -- Chris
>>> 
>>> On Jun 11, 2013, at 10:52 AM, David Chase <david.r.chase at oracle.com> wrote:
>>> 
>>>> This is an update and application of a leftover patch from Christian Thalinger's bug 7198429 .
>>>> I believe it is intended to accumulate call stacks slightly more quickly; it is not a correctness bug.
>>>> 
>>>> There were some minor differences and some parts had already been applied, so it required a hand-apply and line-by-line inspection.
>>>> 
>>>> http://cr.openjdk.java.net/~drchase/8010124/webrev.00/
>>>> 
>>>> Testing: jtreg locally, JPRT, also hand-tested some applets that attempt to do Bad Things, and they failed.
>>>> 
>>>> David
>>>> 
>>> 
>> 
>