RFR 8211438: [Testbug] runtime/XCheckJniJsig/XCheckJSig.java looks for libjsig in wrong location

Harold David Seigel harold.seigel at oracle.com
Fri Oct 5 11:42:55 UTC 2018


Hi David, Lois,

Thanks for reviewing this.  I'll remove the jre path from the test 
before pushing the change.

JNI_EXPORT did fix the Mac OS issue.

Thanks, Harold


On 10/5/2018 7:25 AM, Lois Foltan wrote:
> +1. Looks good Harold.
> Lois
>
> On 10/4/2018 5:13 PM, David Holmes wrote:
>> Hi Harold,
>>
>> There's no more jre in the path either so this can be further 
>> simplified.
>>
>> Otherwise all looks good. I hope the JNI_EXPORT fixes the Mac OS issue.
>>
>> Thanks,
>> David
>>
>> On 5/10/2018 6:38 AM, Harold David Seigel wrote:
>>> Hi,
>>>
>>> Please review this fix for JDK-8211438 (and probably JDK-8211084 
>>> <https://bugs.openjdk.java.net/browse/JDK-8211084>). The fix changes 
>>> the test to not use the os_arch string when constructing the path of 
>>> libjsig.so.  It also causes the test to fail if libjsig.so is not 
>>> found.
>>>
>>> The fix also adds JNIEXPORT to the functions in 
>>> src/java.base/unix/native/libjsig/jsig.c that the JVM tries to 
>>> access by calling dlsym().
>>>
>>> Open Webrev: http://cr.openjdk.java.net/~hseigel/bug_8211438/webrev/
>>>
>>> JBS Bug:  https://bugs.openjdk.java.net/browse/JDK-8211438
>>>
>>> The fix was tested by running the XCheckJSig.java test on Linux, 
>>> Mac, and Solaris.  The fix was regression tested by running Mach5 
>>> tiers 1 and 2 tests and builds on Linux-x64, Windows, and Mac OS X, 
>>> running tiers 3-5 tests on Linux-x64, and by running JCK-12 Lang and 
>>> VM tests on Linux-x64.
>>>
>>> Thanks, Harold
>>>
>



More information about the hotspot-runtime-dev mailing list