(11) RFR (M) JDK-8186211: Negative tests for CONSTANT_Dynamic

[Lois Foltan]

On 2/7/2018 6:24 PM, Paul Sandoz wrote:

> Hi Lois,
>
> This looks good to me. It might be useful to embed the Java source (or byte code generator) that you used to generate the jcod/jasm which was then tweaked, if you think that’s useful then please don’t let it block this commit as it could be a separate issue.
>
> Paul.

Thanks for the review Paul!  Good idea, I have added the Java pseudo 
code as a comment to all the jcod files, the jasm files seem pretty 
straightforward.  I have also added @requires os.arch != "sparcv9" which 
will be removed once condy is supported on SPARC.

New webrev at http://cr.openjdk.java.net/~lfoltan/bug_jdk8186211.1/webrev/

Lois

>
>> On Feb 7, 2018, at 12:19 PM, Lois Foltan <lois.foltan at oracle.com> wrote:
>>
>> Please review this webrev of several additional tests for the new constant pool form, CONSTANT_Dynamic.  The majority of these cover negative testing for a poorly formed CONSTANT_Dynamic_info structure.  A positive test to ensure that a condy's Bootstrap method can be invoked via a REF_newInvokeSpecial is also included.
>>
>> open webrev at http://cr.openjdk.java.net/~lfoltan/bug_jdk8186211/webrev/
>> bug link https://bugs.openjdk.java.net/browse/JDK-8186211
>>
>> Testing in progress (hs-tier1-3, jdk-tier1-3)
>>
>> Thanks,
>> Lois