RFR: 8292695: SIGQUIT and jcmd attaching mechanism does not work with signal chaining library

[David Holmes]

On Mon, 22 Aug 2022 16:41:20 GMT, Xin Liu <xliu at openjdk.org> wrote:

>> There are a few serviceability dcmd jtreg tests that already depend on /vmTestbase:
>> - test/hotspot/jtreg/serviceability/dcmd/framework/HelpTest.java
>> - test/hotspot/jtreg/serviceability/dcmd/framework/VMVersionTest.java
>> - test/hotspot/jtreg/serviceability/dcmd/framework/Invalidcommandtest.java
>> 
>> Is there a real downside to depend on /vmTestbase?
>> 
>> Agree that it is a good suggestion to add a parameter to ProcessTools.sendCtrlBreak() and move it to test/lib. This function is only used in one other place datadumpreq001.java, so it's probably easy to change.
>> However, it's probably better to make a separate test-only RFE to move this function and add a test for JDK-8279124 together.
>
> make sense.

There can be issues using vmTestbase as a library, simply due to the fact that jtreg does not handle libraries particularly well and we can end up with NoClassDefFoundError at runtime due to "split packages".

-------------

PR: https://git.openjdk.org/jdk/pull/9955