Review Request (S) 8015436: compiler/ciReplay/TestSA.sh fails with assert() index is out of bounds

[Vladimir Kozlov]

Sergey,

The fix is good.

You don't need to show the failure output in the test sources since you 
have bugid in it.

Also you don't need /othervm if you don't have command line flags.

Vladimir

On 5/30/13 10:29 AM, serguei.spitsyn at oracle.com wrote:
>
> Thanks, Chris!
> Serguei
>
> On 5/30/13 8:41 AM, Christian Thalinger wrote:
>> Looks good.  -- Chris
>>
>> On May 29, 2013, at 9:08 PM, serguei.spitsyn at oracle.com wrote:
>>
>>> Please, review the fix and unit test for:
>>>   bug: http://bugs.sun.com/view_bug.do?bug_id=8015436
>>>   jbs:  https://jbs.oracle.com/bugs/browse/JDK-8015436
>>>
>>> Open webrev:
>>> http://cr.openjdk.java.net/~sspitsyn/webrevs/2013/hotspot/8015436-JVMTI-JSR292.1
>>>
>>>
>>> Summary:
>>>   The InstanceKlass _initial_method_idnum value must be adjusted as
>>> necessary if the overpass methods are added.
>>>   I guess, it is Ok to integrate a unit test covering the fix under
>>> the same bug id.
>>>
>>> Testing:
>>>   Newly added hotspot unit test: compiler/8015436/Test8015436.java
>>>   The originally failed test:    compiler/ciReplay/TestSA.sh
>>>   The vm/mlvm tests
>>>
>>> Thanks,
>>> Serguei
>