win32: sun/security/ssl/SSLSocketImpl/SSLSocketLeak.java failed with jdk18 and jdk17

[sophia-guo]

sun/security/ssl/SSLSocketImpl/SSLSocketLeak.java failed with jdk18 and jdk17

13:57:25  ACTION: main -- Failed. Execution failed: `main' threw exception: java.lang.UnsatisfiedLinkError: 'long jdk.test.lib.util.FileUtils.getWinProcessHandleCount()'
13:57:25  REASON: User specified action: run main/othervm SSLSocketLeak 
13:57:25  TIME:   0.289 seconds
13:57:25  messages:
13:57:25  command: main SSLSocketLeak
13:57:25  reason: User specified action: run main/othervm SSLSocketLeak 
13:57:25  Mode: othervm [/othervm specified]
13:57:25  elapsed time (seconds): 0.289
13:57:25  configuration:
13:57:25  STDOUT:
13:57:25  STDERR:
13:57:25  java.lang.UnsatisfiedLinkError: 'long jdk.test.lib.util.FileUtils.getWinProcessHandleCount()'
13:57:25  	at jdk.test.lib.util.FileUtils.getWinProcessHandleCount(Native Method)
13:57:25  	at jdk.test.lib.util.FileUtils.getProcessHandleCount(FileUtils.java:374)
13:57:25  	at SSLSocketLeak.main(SSLSocketLeak.java:49)
13:57:25  	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
13:57:25  	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77)
13:57:25  	at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
13:57:25  	at java.base/java.lang.reflect.Method.invoke(Method.java:568)
13:57:25  	at com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:127)
13:57:25  	at java.base/java.lang.Thread.run(Thread.java:833)

Grinder with April release the error is a little bit different, which should be caused by no test image.

16:44:17  java.lang.UnsatisfiedLinkError: no FileUtils in java.library.path: E:\workspace\Grinder\openjdkbinary\j2sdk-image\bin;C:\Windows\Sun\Java\bin;C:\Windows\system32;C:\Windows;C:\cygwin64\bin;C:\Windows\system32;C:\Windows;C:\Windows\System32\Wbem;C:\Windows\System32\WindowsPowerShell\v1.0;C:\Program Files (x86)\Microsoft SQL Server\100\Tools\Binn;C:\Program Files\Microsoft SQL Server\100\Tools\Binn;C:\Program Files\Microsoft SQL Server\100\DTS\Binn;C:\Strawberry\perl\bin;C:\Program Files\Git\cmd;C:\openjdk\jdk-8\bin;C:\apache-ant\apache-ant-1.10.5\bin;C:\Program Files\Microsoft SQL Server\110\Tools\Binn;C:\Program Files (x86)\Microsoft SDKs\TypeScript\1.0;C:\Program Files\Microsoft SQL Server\120\Tools\Binn;C:\Program Files (x86)\IncrediBuild;C:\rust\bin;C:\Program Files\IcedTeaWEB\bin;.
16:44:17  	at java.base/java.lang.ClassLoader.loadLibrary(ClassLoader.java:2429)
16:44:17  	at java.base/java.lang.Runtime.loadLibrary0(Runtime.java:818)
16:44:17  	at java.base/java.lang.System.loadLibrary(System.java:1989)
16:44:17  	at jdk.test.lib.util.FileUtils.getProcessHandleCount(FileUtils.java:371)
16:44:17  	at SSLSocketLeak.main(SSLSocketLeak.java:49)
16:44:17  	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
16:44:17  	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:77)
16:44:17  	at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
16:44:17  	at java.base/java.lang.reflect.Method.invoke(Method.java:568)
16:44:17  	at com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:127)

Tests passed for jdk11 and passed for jdk17 windows64. jdk17 windows64 is using the same jni and test library.

[sophia-guo]

Windows uses FileUtils library and JNI getWinProcessHandleCount(). Passed on win64 and failed on win32. @jerboaa could you take a look?

[sophia-guo]

Could be windows32 compiler issue https://github.com/openjdk/jdk17u/blob/master/test/lib/jdk/test/lib/util/libFileUtils.c#L27-L43 ?

[jerboaa]

Upstream OpenJDK no longer tests on Win32, so this might be a test library issue. Either way, I have no way to test or look into this (not a Windows person). @karianna mentioned that there is interest from Microsoft to keep Win32 alive. Perhaps somebody from his team can look into it?

[karianna]

Will take a look internally (may be a July PSU target fix for us)

[smlambert]

ok, meantime I guess we exclude this testcase on win32