RFR: 8240187: Test hotspot/jtreg/runtime/jni/terminatedThread/TestTerminatedThread.java fails on Alpine Linux

[David Holmes]

On 29/02/2020 8:29 pm, Mikael Vidstedt wrote:
> Is it only pthread_getcpuclockid() that is the problem, or are the other functions also problematic. Put differently, should the whole test be excluded, or are the parts of the test that still make sense to run on musl?

Anything that leads to calling a pthread function is potentially a 
problem. Doing the analysis to see which sub-tests might do that would 
be rather tedious IMO and not a particularly good use of ones time IMO.

Further the parts that aren't affected also aren't utilising musl, so 
the test adds nothing to musl testing in that regard.

Cheers,
David

> Cheers,
> Mikael
> 
>> On Feb 27, 2020, at 11:20 PM, Alexander Scherbatiy <alexander.scherbatiy at bell-sw.com> wrote:
>>
>> Hello,
>>
>> The hotspot test TestTerminatedThread.java fails on Alpine Linux because pthread_getcpuclockid() from musl libc library does not check that the passed thread id is valid (see discussion on musl mail list [1]).
>>
>> The proposed fix skips the test for the musl glibc library as it is suggested in the JDK-8240187 comments [1].
>>
>>    http://cr.openjdk.java.net/~alexsch/8240187/webrev.00/
>>
>>
>> [1] https://www.openwall.com/lists/musl/2020/02/10/6
>>
>> [2] https://bugs.openjdk.java.net/browse/JDK-8240187
>>
>>
>> Thanks,
>>
>> Alexander.
>>
>