RFR 8209443: [lworld] byteMutation test asserts intermittently in the verifier

[Harold David Seigel]

Thanks Fred!

Harold


On 8/14/2018 3:20 PM, Frederic Parain wrote:
> Looks good to me.
>
> Fred
>
>
>> On Aug 14, 2018, at 14:26, Harold David Seigel <harold.seigel at oracle.com> wrote:
>>
>> Hi,
>>
>> Please review this LWorld change for bug JDK-8209443.  The fix checks that the verificationType is not an uninitialized type before calling VerificationType.name(), preventing VerificationType::name() from asserting.  The fix also includes two small changes to use the constant pool index version of is_declared_value_type(), instead of the class name version.
>>
>> Open Webrev: http://cr.openjdk.java.net/~hseigel/bug_8209443/webrev/
>>
>> JBS Bug:  https://bugs.openjdk.java.net/browse/JDK-8209443
>>
>> The changes were tested with Mach5 tiers 1 - 5 on Linux and Mac. Additionally, the byteMutation test was run over 25 times against the fix.
>>
>> Thanks, Harold
>>