[jdk21u-dev] RFR: 8321713: Harmonize executeTestJvm with create[Limited]TestJavaProcessBuilder

[Lutz Schmidt]

On Mon, 22 Apr 2024 13:45:39 GMT, Goetz Lindenmaier <goetz at openjdk.org> wrote:

> I backport this for parity with Oracle.
> 
> File not in 21:
> test/hotspot/jtreg/compiler/compilercontrol/commands/MemLimitTest.java
> was added by "8319301: Static analysis warnings after JDK-8318016" in 22.
> test/hotspot/jtreg/compiler/compilercontrol/commands/MemStatTest.java
> Was added by "8318671: Potential uninitialized uintx value after JDK-8317683".
> "8317683 Add JIT memory statistics" was added in 22.
> 
> test/hotspot/jtreg/compiler/compilercontrol/commands/OptionTest.java
> Resolved because "8282797: CompileCommand parsing errors should exit VM" is only in 22.
> It has a CSR and modifies behaviour, so not a good candidate to backport as prereq.
> 
> File not in 21:
> test/hotspot/jtreg/compiler/print/CompileCommandMemLimit.java
> Added by "8318016: Per-compilation memory ceiling" in 22.
> test/hotspot/jtreg/compiler/print/CompileCommandPrintMemStat.java
> Added by "8317683 Add JIT memory statistics" in 22.
> 
> test/hotspot/jtreg/compiler/vectorapi/TestVectorErgonomics.java
> Resolved because some Vector API changes are not in 21.
> 
> 
> File not in 21:
> test/hotspot/jtreg/runtime/ErrorHandling/UncaughtNativeExceptionTest.java
> Added in 22 by "8310228: Improve error reporting for uncaught native exceptions on Windows"
> 
> I had to edit two additional files. 
> I could not execute hotspot/jtreg/gc/z/TestHighUsage.java, but the edit is trivial enough to 
> skip this.
> compiler/cpuflags/TestAESIntrinsicsOnSupportedConfig.java fails in an unrelated way but it is on the
> ProblemList anyways.
> Other tests or tests using the edited class pass.

LGTM.

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

Marked as reviewed by lucy (Reviewer).

PR Review: https://git.openjdk.org/jdk21u-dev/pull/520#pullrequestreview-2022336222