RFR: 8333130: MakeJAR2.sh uses hard-coded JDK version [v5]

[Serguei Spitsyn]

On Tue, 4 Jun 2024 08:54:40 GMT, Jaikiran Pai <jpai at openjdk.org> wrote:

>> test/jdk/java/lang/instrument/NativeMethodPrefixAgent.java line 34:
>> 
>>> 32:  * @modules java.management
>>> 33:  *          java.instrument
>>> 34:  * @run shell/timeout=240 MakeJAR2.sh NativeMethodPrefixAgent NativeMethodPrefixApp 'Can-Retransform-Classes: true' 'Can-Set-Native-Method-Prefix: true'
>> 
>> Just a question. The original test had a timeout:`timeout=240`. My guess is that the default timeout was not enough. Do you think, it is not needed anymore or you just missed to add it?
>> The same question applies to the other test.
>
> Hello Serguei, that `timeout=240` appears to have been added as part of https://bugs.openjdk.org/browse/JDK-6528548 to several of these tests back in Java 7 days because the `shell` test which was creating the jar file was timing out. The actual test itself, the one which uses the generated jar to run the agent, was completing within a second or two. 
> 
> With the current state of the PR I have run tier3 in our CI and there too both these tests complete within a second on all platforms. So I decided not to copy over this timeout as part of this change.

My guess is these tests were timed out when executed with some specific options like -Xint, -Xcomp or any other that can slow down the execution (e.g. DeoptimizeALot).

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

PR Review Comment: https://git.openjdk.org/jdk/pull/19495#discussion_r1625855573