On Fri, 30 Oct 2026 05:12:35 GMT, Yasumasa Suenaga <[email protected]> wrote:
>> `TestDebugInfoDecode` checks for "java.lang.NullPointerException" in the >> output. At first, it checked just for "Exception", but the test failed a few >> times out of 300 for each OS due to the following exception in the output: >> >> Exception while decoding debug info for NMethod for >> jdk/internal/classfile/impl/StackMapGenerator$Type.<clinit>()V==>nsun.jvm.hotspot.code.NMethod@0x0000ffff8bfda748 >> sun.jvm.hotspot.utilities.AssertionFailure: inconsistent debug information >> at >> jdk.hotspot.agent/sun.jvm.hotspot.utilities.Assert.that(Assert.java:32) >> at >> jdk.hotspot.agent/sun.jvm.hotspot.code.ScopeDesc.decodeObjectValues(ScopeDesc.java:210) >> at >> jdk.hotspot.agent/sun.jvm.hotspot.code.ScopeDesc.<init>(ScopeDesc.java:75) >> at >> jdk.hotspot.agent/sun.jvm.hotspot.code.NMethod.decodeAllScopeDescs(NMethod.java:304) >> at TestDebugInfoDecode$1.visit(TestDebugInfoDecode.java:73) >> at >> jdk.hotspot.agent/sun.jvm.hotspot.memory.CodeHeap.iterate(CodeHeap.java:113) >> at >> jdk.hotspot.agent/sun.jvm.hotspot.code.CodeCache.iterate(CodeCache.java:135) >> at TestDebugInfoDecode.checkDecode(TestDebugInfoDecode.java:84) >> at TestDebugInfoDecode.main(TestDebugInfoDecode.java:119) >> >> >> Both jtreg SA tests and the manual clhsdb test pass after the new commit. >> TestDebugInfoDecode passed 300 times for each platform (Windows, Linux >> x64/aarch64, macOS). > >> but the test failed a few times out of 300 for each OS due to the following >> exception in the output: > > I think `ScopeDesc::decodeObjectValues` in SA is a mirror of > `ScopeDesc::decode_monitor_values` in HotSpot. If so, it seems to relate to > [JDK-8287061](https://bugs.openjdk.org/browse/JDK-8287061). It removed > assertion code which caused the exception. > > https://github.com/openjdk/jdk/commit/a53345ad03e07ab2a990721a506ebc25eed0f7c9#diff-909f858a8da29cce5b684e62f7e6649a3123d268e7171d1c16efed83f0dedfdbL117 > > I think it is out of scope from this issue, you can fix it on another issue. @YaSuenag You're right. JDK-8287061 removed the assert statement from decode_object_values() in HotSpot, but the corresponding one still remains in SA. https://github.com/openjdk/jdk/blob/6cf31a5e10084bd171a63908deb8bfce7443e448/src/jdk.hotspot.agent/share/classes/sun/jvm/hotspot/code/ScopeDesc.java#L210 ------------- PR Comment: https://git.openjdk.org/jdk/pull/33568#issuecomment-6516620037
