Thanks! Do let me know if I can help you with figuring this out. -- adrian
> On Aug 17, 2018, at 5:34 PM, Greg Clayton <clayb...@gmail.com> wrote: > > $ svn commit > Sending > packages/Python/lldbsuite/test/tools/lldb-vscode/attach/TestVSCode_attach.py > Sending > packages/Python/lldbsuite/test/tools/lldb-vscode/breakpoint/TestVSCode_setBreakpoints.py > Sending > packages/Python/lldbsuite/test/tools/lldb-vscode/breakpoint/TestVSCode_setExceptionBreakpoints.py > Sending > packages/Python/lldbsuite/test/tools/lldb-vscode/breakpoint/TestVSCode_setFunctionBreakpoints.py > Sending > packages/Python/lldbsuite/test/tools/lldb-vscode/launch/TestVSCode_launch.py > Sending > packages/Python/lldbsuite/test/tools/lldb-vscode/stackTrace/TestVSCode_stackTrace.py > Sending > packages/Python/lldbsuite/test/tools/lldb-vscode/step/TestVSCode_step.py > Sending > packages/Python/lldbsuite/test/tools/lldb-vscode/variables/TestVSCode_variables.py > Transmitting file data ........done > Committing transaction... > Committed revision 340112. > > >> On Aug 17, 2018, at 3:43 PM, Adrian Prantl <apra...@apple.com >> <mailto:apra...@apple.com>> wrote: >> >> We really need to do *something*. Normally our lldb-cmake bot churns out a >> fresh run every 10 minutes, now it is timing out after 1.5 hours: >> >> http://green.lab.llvm.org/green/view/LLDB/job/lldb-cmake/9021/consoleFull#console-section-10 >> >> <http://green.lab.llvm.org/green/view/LLDB/job/lldb-cmake/9021/consoleFull#console-section-10> >> Build timed out (after 90 minutes). Marking the build as aborted. >> http://green.lab.llvm.org/green/view/LLDB/job/lldb-cmake/buildTimeTrend >> >> Until we figured this out, would you mind setting all tests that use the >> vs_code shim to skipIfDarwin? >> >> I'm happy to help you with collecting additional diagnostics, also feel free >> to enable them again if you want to test a fix. I can also run commands for >> you directly on the bot if that helps. >> >> -- adrian >> >> >>> On Aug 17, 2018, at 1:52 PM, Adrian Prantl <apra...@apple.com> wrote: >>> >>> >>> >>>> On Aug 17, 2018, at 1:49 PM, Adrian Prantl <apra...@apple.com> wrote: >>>> >>>> >>>> On Aug 17, 2018, at 1:44 PM, Greg Clayton <clayb...@gmail.com> wrote: >>>> >>>>> I run everything on a mac and there are no failures on my machine. >>>>> Can you find out exactly how the test suite is invoked so I can try to >>>>> reproduce? >>>> >>>> You can find the exact invocation in the build logs: >>>> >>>> cmake /Users/buildslave/jenkins/workspace/lldb-cmake/src/llvm >>>> -DCMAKE_BUILD_TYPE=Release -DLLVM_ENABLE_ASSERTIONS=ON >>>> -DCMAKE_EXPORT_COMPILE_COMMANDS=ON >>>> -DCMAKE_INSTALL_PREFIX=/Users/buildslave/jenkins/workspace/lldb-cmake/results/lldb >>>> >>>> '-DLLDB_TEST_USER_ARGS=--arch;x86_64;--build-dir;/Users/buildslave/jenkins/workspace/lldb-cmake/build/lldb-test-build.noindex;--session-file-format;fm;-s=/Users/buildslave/jenkins/workspace/lldb-cmake/test/logs;--rerun-all-issues;--env;TERM=vt100' >>>> >>>> '-DLLVM_LIT_ARGS=--xunit-xml-output=/Users/buildslave/jenkins/workspace/lldb-cmake/test/results.xml >>>> -v' -G Ninja >>>> >>>> ninja >>>> >>>> env TERM=vt100 ninja -v check-lldb >>>> >>>> This is running on a 2013 Mac Pro with 24 threads on 10.13.5 with Xcode >>>> 9.2. >>> >>> I would not be surprised if the fact that lit launches 24 jobs >>> simultaneously is significant for reproducing the issue; it does look like >>> some sort of synchronization bug that may need a really loaded machine to >>> trigger. >>> >>> -- adrian >> >
_______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
