RFR: Store cpu features in AOTCodeCache header [v9]

Vladimir Kozlov kvn at openjdk.org
Mon Jul 28 21:08:11 UTC 2025


On Mon, 28 Jul 2025 16:27:48 GMT, Ashutosh Mehra <asmehra at openjdk.org> wrote:

> I see some failures on macos-aarch64 in GHA testing:
> 
> * tier1:  `jdk/java/lang/Thread/virtual/stress/PingPong.java` and `java/lang/Thread/virtual/stress/Skynet.java#default` failing with time outs on macos-aarch64
> * tier2: `java/util/concurrent/BlockingQueue/ProducerConsumerLoops.java`
> * tier3: `java/foreign/TestConcurrentClose.java`
> 
> They all fail due to timeouts. Has anyone seen these failures before? I don't think these can be related to my changes.

I see timeout only for this test.in latest testing of premain:
   - runtime/Monitor/ConcurrentDeflation.java

My latest testing of your changes showed timeouts:
 - tier1 the same as you see and java/util/concurrent/BlockingQueue/SingleProducerMultipleConsumerLoops.java
 - tier3: several serviceability/jvmti/stress/StackTrace tests

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

PR Comment: https://git.openjdk.org/leyden/pull/84#issuecomment-3129886581


More information about the leyden-dev mailing list