RFR 8144279 [TESTBUG] hotspot/runtime/jsig/Test8017498.sh should use native library build support

[Chris Plummer]

On 5/5/16 1:25 PM, Gerald Thornbrugh wrote:
> Hi Chris,te:
>> He Jerry,
>>
>> You dropped the "-z noexecstack" part. Isn't this needed? I think the 
>> default is not always the same for all version of gcc and all linux 
>> platforms, which is why it was explicitly added.
> If the "-z noexecstack" option is not used the object is not tagged 
> and is basically treated as "-z noexecstack".
This is the part I'm not so certain about. I have this vague 
recollection of the default not always being noexecstack. I've cc'd 
Dean, since he's dealt with this topic before.

thanks,

Chris
> When loading occurs if no objects have been tagged with "execstack" 
> the stack is set to non-executable.
> If only one object is tagged with "execstack" then it sets the stack 
> as an executable stack.
>
> I can put it back if you like.
>
> Thanks for the review.
>
> Jerry
>>
>> thanks,
>>
>> Chris
>>
>> On 5/3/16 2:54 PM, Gerald Thornbrugh wrote:
>>> Hi Everyone,
>>>
>>> I would like to have the following change reviewed:
>>>
>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8144279
>>> Webrev: http://cr.openjdk.java.net/~gthornbr/8144279/webrev.00/ 
>>> <http://cr.openjdk.java.net/%7Egthornbr/8144279/webrev.00/>
>>>
>>> JPRT testing completed without errors and I verified that the test 
>>> was run.
>>>
>>>
>>> The changes consist of:
>>>
>>> Replacing the existing Test8017498.sh script file with a java test 
>>> source file Testjsig.java.
>>>
>>> Adding the test directory to the "BUILD_HOTSPOT_JTREG_NATIVE_SRC" 
>>> list for Linux builds
>>> so the native libTestJNI.c source file would be compiled into a 
>>> native library.
>>>
>>> Renaming the existing TestJNI.c native source file to libTestJNI.c 
>>> so the build process would
>>> identify the file as a native library source file and build the 
>>> library.
>>>
>>> Please let me know if you have any questions or concerns.
>>>
>>> Thanks,
>>>
>>> Jerry
>>
>