Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

macOS: Use Clang for JDK8 JIT builds #3629

Merged
merged 1 commit into from
Nov 13, 2018

Conversation

nbhuiyan
Copy link
Member

@nbhuiyan nbhuiyan commented Nov 9, 2018

Necessary for removing the JIT's dependency on
libstdc++.6.0.9.dylib that results in failing to load libj9jit on
newer macOS versions not having the right version of the library.

Signed-off-by: Nazim Uddin Bhuiyan [email protected]

Fixes: #3496
Depends on: ibmruntimes/openj9-openjdk-jdk8#206
Issue: #36

Necessary for removing the JIT's dependency on
libstdc++.6.0.9.dylib that results in failing to load libj9jit on
newer macOS versions not having the right version of the library.

Signed-off-by: Nazim Uddin Bhuiyan <[email protected]>
@nbhuiyan nbhuiyan changed the title Use Clang for JDK8 JIT builds macOS: Use Clang for JDK8 JIT builds Nov 9, 2018
@DanHeidinga
Copy link
Member

Depends on: ibmruntimes/openj9-openjdk-jdk8#206

@nbhuiyan can this be delivered before the extensions change? Is the extensions change safe to deliver before this? Trying to understand which way the dependency goes so we can get these delivered without breaking the builds.

@nbhuiyan
Copy link
Member Author

@DanHeidinga , The extensions change would need to go first if we want to prevent breaking the OSX JDK8 builds. Without ibmruntimes/openj9-openjdk-jdk8#206 , we have to build the OpenJ9 portion using gcc-4.9 on macOS. This patch would add clang-specific compiler flags that causes gcc to error out.

If there are currently no automated CI and builds set up for OSX jdk8 (my assumption), then it should not matter which PR gets delivered first.

Copy link
Member

@DanHeidinga DanHeidinga left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

lgtm

@DanHeidinga
Copy link
Member

Jenkins test sanity osx jdk11

@DanHeidinga
Copy link
Member

Running a jdk11 pr build to validate the file is still well formatted as we don't have jdk8 mac builds yet

@DanHeidinga
Copy link
Member

ibmruntimes/openj9-openjdk-jdk8#206 has been merged

@DanHeidinga DanHeidinga self-assigned this Nov 12, 2018
@nbhuiyan
Copy link
Member Author

@DanHeidinga sanity test seems to have hanged after successfully building. Is that expected?

@babsingh
Copy link
Contributor

babsingh commented Nov 12, 2018

11:14:35 Testing: abort  
11:14:35 Test start time: 2018/11/12 11:14:37 Eastern Standard Time
11:14:35 Running command: "/Users/jenkins/workspace/PullRequest-Sanity-JDK11-osx_x86-64_cmprssptrs-OpenJ9/build/macosx-x86_64-normal-server-release/images/jdk/bin/java" -Xcompressedrefs  -Xdump:system:none -cp "/Users/jenkins/workspace/PullRequest-Sanity-JDK11-osx_x86-64_cmprssptrs-OpenJ9/openj9/test/TestConfig/scripts/testKitGen/../../../../jvmtest/functional/cmdline_options_testresources/cmdlinetestresources.jar" VMBench.GPTests.GPTest   abort
11:14:35 Time spent starting: 5 milliseconds
11:19:38 ***[TEST INFO 2018/11/12 11:19:37] ProcessKiller detected a timeout after 300000 milliseconds!***
11:19:38 INFO: getUnixPID() has failed indicating this is not a UNIX System.'Debug on timeout' is currently only supported on Linux.
11:19:38 Output from test:
11:19:38  [ERR] JVMDUMP039I Processing dump event "abort", detail "" at 2018/11/12 11:14:39 - please wait.
11:19:38  [ERR] JVMDUMP032I JVM requested Java dump using '/Users/jenkins/workspace/PullRequest-Sanity-JDK11-osx_x86-64_cmprssptrs-OpenJ9/openj9/test/TestConfig/test_output_15420384313143/cmdLineTest_gpTest_0/javacore.20181112.111439.16125.0001.txt' in response to an event
12:11:36 Cannot contact osx1013-x86-3: java.lang.InterruptedException

12:11:36 Cannot contact osx1013-x86-3: java.lang.InterruptedException - The CI build job has disconnected from the machine. This is an infrastructure failure; unrelated to the pull request changes. These changes only impact the build phase. The JDK build was successful. Since the failure happened in the test phase, we can ignore the failure and merge the changes.

@DanHeidinga
Copy link
Member

Jenkins test sanity osx jdk11

@DanHeidinga
Copy link
Member

No, that's not expected. Rekicked the build and will check it again tonight.

@babsingh
Copy link
Contributor

babsingh commented Nov 13, 2018

CI job failed again. This time ant is unavailable on osx1013-x86-6:

18:47:01 ant -f scripts/build_test.xml -DTEST_ROOT=/Users/jenkins/workspace/PullRequest-Sanity-JDK11-osx_x86-64_cmprssptrs-OpenJ9/openj9/test/TestConfig/.. -DBUILD_ROOT=/Users/jenkins/workspace/PullRequest-Sanity-JDK11-osx_x86-64_cmprssptrs-OpenJ9/openj9/test/TestConfig/../../jvmtest -DJAVA_BIN=/Users/jenkins/workspace/PullRequest-Sanity-JDK11-osx_x86-64_cmprssptrs-OpenJ9/build/macosx-x86_64-normal-server-release/images/jdk/bin -DJDK_VERSION=11 -DJCL_VERSION=latest -DBUILD_LIST= -DRESOURCES_DIR=/Users/jenkins/workspace/PullRequest-Sanity-JDK11-osx_x86-64_cmprssptrs-OpenJ9/openj9/test/TestConfig/../../jvmtest/TestConfig/resources -DPLATFORM=osx_x86-64_cr -DJDK_HOME=/Users/jenkins/workspace/PullRequest-Sanity-JDK11-osx_x86-64_cmprssptrs-OpenJ9/build/macosx-x86_64-normal-server-release/images/jdk/bin/.. -DJVM_VERSION=openjdk11-openj9
18:47:01 make: ant: No such file or directory
18:47:01 make: *** [compile] Error 1

fyi - @jdekonin

@DanHeidinga DanHeidinga merged commit fda5677 into eclipse-openj9:master Nov 13, 2018
@DanHeidinga
Copy link
Member

merged as failures are machine configuration, not related to this pr

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants