RFR(S): 8205615: Start of release updates for JDK 12 / 8205619: Bump maximum recognized class file version to 56 for JDK 12

[David Holmes]

On 26/06/2018 4:23 PM, Mikael Vidstedt wrote:
> 
> 
>> On Jun 25, 2018, at 10:38 PM, David Holmes <david.holmes at oracle.com> wrote:
>>
>> Hi Mikael,
>>
>> Generally looks good, only one issue ...
>>
>> On 26/06/2018 9:44 AM, Mikael Vidstedt wrote:
>>> All,
>>> Shamelessly stealing the background/intro from Joe’s email to build-dev[1] earlier today:
>>> With the JDK 11 and 12 split fast approaching [1], it is time to work on the various start of release update tasks for JDK 12. Those tasks are being tracked under the umbrella bug JDK-8205615: "Start of release updates for JDK 12".
>>> This thread is to review the hotspot-related portions of the work including JDK-8205619: "Bump maximum recognized class file version to 56 for JDK 12”.
>>> Bug: https://bugs.openjdk.java.net/browse/JDK-8205619
>>> CSR: https://bugs.openjdk.java.net/browse/JDK-8205642
>>> Webrev: http://cr.openjdk.java.net/~darcy/8205615.4/ <http://cr.openjdk.java.net/~darcy/8205615.4/>
>>> Of specific interest for hotspot are the following files:
>>> * src/hotspot/share/classfile/classFileParser.cpp
>>> Added a JAVA_12_VERSION definition which is, for now, unused.
>>
>> Good.
>>
>>> * src/jdk.internal.vm.compiler/share/classes/org.graalvm.compiler.replacements/src/org/graalvm/compiler/replacements/classfile/Classfile.java
>>> Bumped the maximum recognized version. Doug has promised to help upstream this to Graal.
>>
>> Good.
>>
>>> * test/hotspot/jtreg/runtime/CommandLine/VMDeprecatedOptions.java
>>> Removed verification of the previously deprecated, now obsolete, VM flags.
>>
>> Good.
>>
>>> * test/jdk/com/sun/management/HotSpotDiagnosticMXBean/CheckOrigin.java
>>> Removed the use of the previously deprecated, now obsolete, PrintSafepointStatistics flag.
>>
>> Bad. Sorry you need to replace the obsolete flag with something non-obsolete (and preferably not deprecated) and leave the test case in place. I suggest the innocuous PrintVMQWaitTime.
> 
> Oh, I seeeee. Thanks for catching! I’ll fix by reverting the changes and simply changing PrintSafepointStatistics to PrintVMQWaitTime as you suggest instead:
> 
> diff -r 356eaea05bf0 test/jdk/com/sun/management/HotSpotDiagnosticMXBean/CheckOrigin.java
> --- a/test/jdk/com/sun/management/HotSpotDiagnosticMXBean/CheckOrigin.java	Mon Jun 25 21:22:16 2018 +0300
> +++ b/test/jdk/com/sun/management/HotSpotDiagnosticMXBean/CheckOrigin.java	Mon Jun 25 23:21:49 2018 -0700
> @@ -58,7 +58,7 @@
>               File flagsFile = File.createTempFile("CheckOriginFlags", null);
>               try (PrintWriter pw =
>                      new PrintWriter(new FileWriter(flagsFile))) {
> -                pw.println("+PrintSafepointStatistics");
> +                pw.println("+PrintVMQWaitTime");
>               }
> 
>               ProcessBuilder pb = ProcessTools.
> @@ -108,7 +108,7 @@
>               checkOrigin("IgnoreUnrecognizedVMOptions", Origin.ENVIRON_VAR);
>               checkOrigin("PrintVMOptions", Origin.ENVIRON_VAR);
>               // Set in -XX:Flags file
> -            checkOrigin("PrintSafepointStatistics", Origin.CONFIG_FILE);
> +            checkOrigin("PrintVMQWaitTime", Origin.CONFIG_FILE);
>               // Set through j.l.m
>               checkOrigin("HeapDumpOnOutOfMemoryError", Origin.MANAGEMENT);
>               // Should be set by the VM, when we set UseConcMarkSweepGC
> 
> With that change the test (still) passes locally.

Looks good.

Thanks,
David

> Cheers,
> Mikael
> 
> 
>>
>> Thanks,
>> David
>>
>>> Testing:
>>> The webrev as a whole has been tested using tier1-3. There are no hotspot related failures.
>>> Cheers,
>>> Mikael
>>> [1] http://mail.openjdk.java.net/pipermail/build-dev/2018-June/022528.html
>>> [2] http://mail.openjdk.java.net/pipermail/jdk-dev/2018-June/001462.html <http://mail.openjdk.java.net/pipermail/jdk-dev/2018-June/001462.html>
>