Re: RFR: JDK-8213001: vmTestbase/nsk/jvmti/ThreadStart/threadstart002/TestDescription.java debug agent times out
On 1/9/19 10:43 AM, Gary Adams wrote: I could use another reviewer, or an ok to push as a trivial change. Thumbs up! Yes, this qualifies as a trivial change. However, even a trivial change requires a (R)eviewer. JC is currently a committer. Dan On 1/8/19, 11:32 AM, Gary Adams wrote: A number failures of threadstart002 have been reported on windows-x64-debug builds. This configuration has a jtreg timeout factor assigned so the test can run 10x slower and not experience a timeout. Unfortunately, these older vmTestbase tests were not fully integrated with the jtreg timeout and factor capabilities. This test in particular uses a 2 second timeout when monitoring transitions between starting a thread and the check for the thread being resumed. This hard coded timeout is not scaled. As a quick workaround the timeout can be scaled up to allow 20 second window, as if the scaling factor had been applied as jtreg harness intended. Issue: https://bugs.openjdk.java.net/browse/JDK-8213001 diff --git a/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp b/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp --- a/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp +++ b/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2003, 2018, Oracle and/or its affiliates. All rights reserved. + * Copyright (c) 2003, 2019, Oracle and/or its affiliates. All rights reserved. * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This code is free software; you can redistribute it and/or modify it @@ -34,7 +34,7 @@ #define PASSED 0 #define STATUS_FAILED 2 -#define WAIT_TIME 2000 +#define WAIT_TIME 2 static jvmtiEnv *jvmti = NULL; static jvmtiCapabilities caps;
Re: RFR: JDK-8213001: vmTestbase/nsk/jvmti/ThreadStart/threadstart002/TestDescription.java debug agent times out
I could use another reviewer, or an ok to push as a trivial change. On 1/8/19, 11:32 AM, Gary Adams wrote: A number failures of threadstart002 have been reported on windows-x64-debug builds. This configuration has a jtreg timeout factor assigned so the test can run 10x slower and not experience a timeout. Unfortunately, these older vmTestbase tests were not fully integrated with the jtreg timeout and factor capabilities. This test in particular uses a 2 second timeout when monitoring transitions between starting a thread and the check for the thread being resumed. This hard coded timeout is not scaled. As a quick workaround the timeout can be scaled up to allow 20 second window, as if the scaling factor had been applied as jtreg harness intended. Issue: https://bugs.openjdk.java.net/browse/JDK-8213001 diff --git a/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp b/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp --- a/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp +++ b/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2003, 2018, Oracle and/or its affiliates. All rights reserved. + * Copyright (c) 2003, 2019, Oracle and/or its affiliates. All rights reserved. * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This code is free software; you can redistribute it and/or modify it @@ -34,7 +34,7 @@ #define PASSED 0 #define STATUS_FAILED 2 -#define WAIT_TIME 2000 +#define WAIT_TIME 2 static jvmtiEnv *jvmti = NULL; static jvmtiCapabilities caps;
RE: RFR: JDK-8213001: vmTestbase/nsk/jvmti/ThreadStart/threadstart002/TestDescription.java debug agent times out
Hi JC, I believe that’s covered by https://bugs.openjdk.java.net/browse/JDK-8206013. Regards, Marty From: JC Beyler Sent: Tuesday, January 8, 2019 9:14 AM To: Gary Adams Cc: OpenJDK Serviceability Subject: Re: RFR: JDK-8213001: vmTestbase/nsk/jvmti/ThreadStart/threadstart002/TestDescription.java debug agent times out Hi Gary, Looks good to me, is there a bug to track that we might want to modernize the timeouts for these tests? Thanks, Jc On Tue, Jan 8, 2019 at 8:30 AM Gary Adams mailto:gary.ad...@oracle.com"gary.ad...@oracle.com> wrote: A number failures of threadstart002 have been reported on windows-x64-debug builds. This configuration has a jtreg timeout factor assigned so the test can run 10x slower and not experience a timeout. Unfortunately, these older vmTestbase tests were not fully integrated with the jtreg timeout and factor capabilities. This test in particular uses a 2 second timeout when monitoring transitions between starting a thread and the check for the thread being resumed. This hard coded timeout is not scaled. As a quick workaround the timeout can be scaled up to allow 20 second window, as if the scaling factor had been applied as jtreg harness intended. Issue: https://bugs.openjdk.java.net/browse/JDK-8213001 diff --git a/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp b/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp --- a/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp +++ b/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2003, 2018, Oracle and/or its affiliates. All rights reserved. + * Copyright (c) 2003, 2019, Oracle and/or its affiliates. All rights reserved. * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This code is free software; you can redistribute it and/or modify it @@ -34,7 +34,7 @@ #define PASSED 0 #define STATUS_FAILED 2 -#define WAIT_TIME 2000 +#define WAIT_TIME 2 static jvmtiEnv *jvmti = NULL; static jvmtiCapabilities caps; -- Thanks, Jc
RFR: JDK-8213001: vmTestbase/nsk/jvmti/ThreadStart/threadstart002/TestDescription.java debug agent times out
A number failures of threadstart002 have been reported on windows-x64-debug builds. This configuration has a jtreg timeout factor assigned so the test can run 10x slower and not experience a timeout. Unfortunately, these older vmTestbase tests were not fully integrated with the jtreg timeout and factor capabilities. This test in particular uses a 2 second timeout when monitoring transitions between starting a thread and the check for the thread being resumed. This hard coded timeout is not scaled. As a quick workaround the timeout can be scaled up to allow 20 second window, as if the scaling factor had been applied as jtreg harness intended. Issue: https://bugs.openjdk.java.net/browse/JDK-8213001 diff --git a/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp b/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp --- a/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp +++ b/test/hotspot/jtreg/vmTestbase/nsk/jvmti/ThreadStart/threadstart002/threadstart002.cpp @@ -1,5 +1,5 @@ /* - * Copyright (c) 2003, 2018, Oracle and/or its affiliates. All rights reserved. + * Copyright (c) 2003, 2019, Oracle and/or its affiliates. All rights reserved. * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * * This code is free software; you can redistribute it and/or modify it @@ -34,7 +34,7 @@ #define PASSED 0 #define STATUS_FAILED 2 -#define WAIT_TIME 2000 +#define WAIT_TIME 2 static jvmtiEnv *jvmti = NULL; static jvmtiCapabilities caps;
