Re: RFR: 8240903: Add test to check that jmod hashes are reproducible [v5]
On Wed, 30 Mar 2022 01:55:30 GMT, Dongbo He wrote: >> This creates a regression test for >> [JDK-8240734](https://bugs.openjdk.java.net/browse/JDK-8240734). This was >> once blocked >> by a time stamp issue which has been resolved by: [JDK-8276766 >> ](https://bugs.openjdk.java.net/browse/JDK-8276766) >> >> We found the issue can be produced stably with at least 64 modules. >> Note that we need to add the --date jmod option to avoid the timestamp issue. >> >> Testing: >> Added test case fails without fix for JDK-8240734, and passes otherwise. >> Tested with tier2 on linux x86. > > Dongbo He has updated the pull request with a new target base due to a merge > or a rebase. The incremental webrev excludes the unrelated changes brought in > by the merge/rebase. The pull request contains six additional commits since > the last revision: > > - Merge branch 'master' into 8240903 > - Delete extra blank links > - Change shell test io java test > - Get date by 'date +%Y-%m-%dT%H:%M:%S%:z' > - Update Copyright > - 8240903: Add a regression test for JDK-8240734 Thank you for your review, Alan. - PR: https://git.openjdk.java.net/jdk/pull/7948
Re: RFR: 8240903: Add test to check that jmod hashes are reproducible [v5]
On Wed, 30 Mar 2022 01:55:30 GMT, Dongbo He wrote: >> This creates a regression test for >> [JDK-8240734](https://bugs.openjdk.java.net/browse/JDK-8240734). This was >> once blocked >> by a time stamp issue which has been resolved by: [JDK-8276766 >> ](https://bugs.openjdk.java.net/browse/JDK-8276766) >> >> We found the issue can be produced stably with at least 64 modules. >> Note that we need to add the --date jmod option to avoid the timestamp issue. >> >> Testing: >> Added test case fails without fix for JDK-8240734, and passes otherwise. >> Tested with tier2 on linux x86. > > Dongbo He has updated the pull request with a new target base due to a merge > or a rebase. The incremental webrev excludes the unrelated changes brought in > by the merge/rebase. The pull request contains six additional commits since > the last revision: > > - Merge branch 'master' into 8240903 > - Delete extra blank links > - Change shell test io java test > - Get date by 'date +%Y-%m-%dT%H:%M:%S%:z' > - Update Copyright > - 8240903: Add a regression test for JDK-8240734 Marked as reviewed by alanb (Reviewer). - PR: https://git.openjdk.java.net/jdk/pull/7948
Re: RFR: 8240903: Add test to check that jmod hashes are reproducible [v5]
On Wed, 30 Mar 2022 01:55:30 GMT, Dongbo He wrote: >> This creates a regression test for >> [JDK-8240734](https://bugs.openjdk.java.net/browse/JDK-8240734). This was >> once blocked >> by a time stamp issue which has been resolved by: [JDK-8276766 >> ](https://bugs.openjdk.java.net/browse/JDK-8276766) >> >> We found the issue can be produced stably with at least 64 modules. >> Note that we need to add the --date jmod option to avoid the timestamp issue. >> >> Testing: >> Added test case fails without fix for JDK-8240734, and passes otherwise. >> Tested with tier2 on linux x86. > > Dongbo He has updated the pull request with a new target base due to a merge > or a rebase. The incremental webrev excludes the unrelated changes brought in > by the merge/rebase. The pull request contains six additional commits since > the last revision: > > - Merge branch 'master' into 8240903 > - Delete extra blank links > - Change shell test io java test > - Get date by 'date +%Y-%m-%dT%H:%M:%S%:z' > - Update Copyright > - 8240903: Add a regression test for JDK-8240734 Thanks for the timings. I mostly asked because we've had issues with I/O heavy tests on older macOS tests in our CI. Several tests have had to be dialled down or configured to run with very large timeouts. We'll have to see how this one fares. - PR: https://git.openjdk.java.net/jdk/pull/7948
Re: RFR: 8240903: Add test to check that jmod hashes are reproducible [v5]
On Fri, 8 Apr 2022 09:26:08 GMT, Alan Bateman wrote: > I think this looks okay. Can you say how long the test runs for you? It > creates 130 JMOD files and I'm just wondering if the default timeout is > enough when running with debug builds. The following is the test result on linux x86 with 128 cores, it looks ok. time make CONF=release run-test TEST=test/jdk/tools/jmod/hashes/HashesOrderTest.java Finished building target 'run-test' in configuration 'linux-x86_64-server-release' real0m24.235s user0m52.196s sys 0m7.170s time make CONF=slowdebug run-test TEST=test/jdk/tools/jmod/hashes/HashesOrderTest.java real2m2.877s user5m58.887s sys 0m10.474s - PR: https://git.openjdk.java.net/jdk/pull/7948
Re: RFR: 8240903: Add test to check that jmod hashes are reproducible [v5]
On Wed, 30 Mar 2022 01:55:30 GMT, Dongbo He wrote: >> This creates a regression test for >> [JDK-8240734](https://bugs.openjdk.java.net/browse/JDK-8240734). This was >> once blocked >> by a time stamp issue which has been resolved by: [JDK-8276766 >> ](https://bugs.openjdk.java.net/browse/JDK-8276766) >> >> We found the issue can be produced stably with at least 64 modules. >> Note that we need to add the --date jmod option to avoid the timestamp issue. >> >> Testing: >> Added test case fails without fix for JDK-8240734, and passes otherwise. >> Tested with tier2 on linux x86. > > Dongbo He has updated the pull request with a new target base due to a merge > or a rebase. The incremental webrev excludes the unrelated changes brought in > by the merge/rebase. The pull request contains six additional commits since > the last revision: > > - Merge branch 'master' into 8240903 > - Delete extra blank links > - Change shell test io java test > - Get date by 'date +%Y-%m-%dT%H:%M:%S%:z' > - Update Copyright > - 8240903: Add a regression test for JDK-8240734 I think this looks okay. Can you say how long the test runs for you? It creates 130 JMOD files and I'm just wondering if the default timeout is enough when running with debug builds. - PR: https://git.openjdk.java.net/jdk/pull/7948
Re: RFR: 8240903: Add test to check that jmod hashes are reproducible [v5]
> This creates a regression test for > [JDK-8240734](https://bugs.openjdk.java.net/browse/JDK-8240734). This was > once blocked > by a time stamp issue which has been resolved by: [JDK-8276766 > ](https://bugs.openjdk.java.net/browse/JDK-8276766) > > We found the issue can be produced stably with at least 64 modules. > Note that we need to add the --date jmod option to avoid the timestamp issue. > > Testing: > Added test case fails without fix for JDK-8240734, and passes otherwise. > Tested with tier2 on linux x86. Dongbo He has updated the pull request with a new target base due to a merge or a rebase. The incremental webrev excludes the unrelated changes brought in by the merge/rebase. The pull request contains six additional commits since the last revision: - Merge branch 'master' into 8240903 - Delete extra blank links - Change shell test io java test - Get date by 'date +%Y-%m-%dT%H:%M:%S%:z' - Update Copyright - 8240903: Add a regression test for JDK-8240734 - Changes: - all: https://git.openjdk.java.net/jdk/pull/7948/files - new: https://git.openjdk.java.net/jdk/pull/7948/files/6b185934..34b8008d Webrevs: - full: https://webrevs.openjdk.java.net/?repo=jdk=7948=04 - incr: https://webrevs.openjdk.java.net/?repo=jdk=7948=03-04 Stats: 3635 lines in 166 files changed: 2489 ins; 556 del; 590 mod Patch: https://git.openjdk.java.net/jdk/pull/7948.diff Fetch: git fetch https://git.openjdk.java.net/jdk pull/7948/head:pull/7948 PR: https://git.openjdk.java.net/jdk/pull/7948
