RFR(T) 8203005: The top-of-stack type specified for nofast_* bytecodes are wrong

[Yumin Qi]

Thanks Dan for the review!


On 6/25/20 12:17 PM, Daniel D. Daugherty wrote:
> On 6/25/20 1:33 PM, Yumin Qi wrote:
>> Hi, please review the tiny changes for
>>
>> bug: https://bugs.openjdk.java.net/browse/JDK-8203005
>>
>> webrev:http://cr.openjdk.java.net/~minqi/2020/8203005/webrev-00/
>
> src/hotspot/share/interpreter/bytecodes.cpp
>     No comments.
>
> src/jdk.hotspot.agent/share/classes/sun/jvm/hotspot/interpreter/Bytecodes.java 
>
>     No comments.
>
> Thumbs up. I agree that this is a trivial fix and you don't need to
> wait 24 hours to push.
>
> Dan
>
>
>
>
>>
>>
>> Summary: The change was left by 
>> 8074345(https://bugs.openjdk.java.net/browse/JDK-8074345), the types 
>> were wrongly put as T_ILLEGAL for T_OBJECT, and T_ILLEGAL for T_INT. 
>> This has not triggered any failures yet since the types stored in the 
>> type array for nofast version are never used, the used types are 
>> always the original types fortunately(unfortunately either).
>>
>>
>> tests: tier1,tier2,tier3
>>
>>
>> Thanks
>>
>> Yumin
>>
>