RFR: 8255992: JFR EventWriter does not use first string from StringPool with id 0 [v3]

[Erik Gahlin]

On Thu, 12 Nov 2020 20:18:37 GMT, Jie Kang <jkang at openjdk.org> wrote:

>> Does the fix pass tests in test/jdk/jdk/jfr?
>
> @egahlin 
> My system is:
> Linux (Fedora 31), jtreg 5.1-b01 configured with flags `--disable-warnings-as-errors --with-jtreg=/path/to/jtreg-5.1-b01/`
> 
> Running `make run-test TEST=":jdk_jfr"` I see one failure in TestShouldCommit with output:
> ----------System.out:(4/378)----------
> [0.001s][warning][logging] No tag set matches selection: jfr+event+setting. Did you mean any of the following? jfr+setting* jfr+event* jfr+system+setting jfr+system+event jfr*
> threshold=100000, duration=[93391-97400], shouldCommit()=true, expected=false
> Unexpected value of shouldCommit(). Searching for active threshold...
> shouldCommit()=true after 90,464 ms, expected 100,000
> ----------System.err:(17/1184)----------
> java.lang.RuntimeException: Wrong shouldCommit(): expected false to equal true
> 	at jdk.test.lib.Asserts.fail(Asserts.java:594)
> 	at jdk.test.lib.Asserts.assertEquals(Asserts.java:205)
> 	at jdk.jfr.api.event.TestShouldCommit.verifyThreshold(TestShouldCommit.java:151)
> 	at jdk.jfr.api.event.TestShouldCommit.verifyThreshold(TestShouldCommit.java:107)
> 	at jdk.jfr.api.event.TestShouldCommit.main(TestShouldCommit.java:55)
> 	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> 	at java.base/jdk.internal.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:64)
> 	at java.base/jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> 	at java.base/java.lang.reflect.Method.invoke(Method.java:564)
> 	at com.sun.javatest.regtest.agent.MainWrapper$MainThread.run(MainWrapper.java:127)
> 	at java.base/java.lang.Thread.run(Thread.java:832)
> 
> I don't think this is related to the patch.
> 
> When running `make run-test-tier1` I see 4 failures:
> gtest/GTestWrapper.java: a jtreg wrapper for gtest tests
> gtest/MetaspaceGtests.java#balanced-no-ccs: Run metaspace-related gtests with compressed class pointers off
> gtest/MetaspaceGtests.java#reclaim-aggressive-ndebug: Run metaspace-related gtests for reclaim policy aggressive
> gtest/MetaspaceGtests.java#reclaim-none-ndebug: Run metaspace-related gtests for reclaim policy none
> 
> I don't think these are related either.

The test looks fragile and failure unrelated.

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

PR: https://git.openjdk.java.net/jdk/pull/1097