Revert "Add environment variable (COMPlus_EnableDiagnostics) to disab…#16007
Conversation
|
|
||
| RaiseStartupNotification(); | ||
|
|
||
| // Also initialize the AppDomainEnumerationIPCBlock |
There was a problem hiding this comment.
@mikem8361 Notice that FEATURE_DBGIPC_TRANSPORT_VM block above uses m_pAppDomainCB - that is initialized too late now.
There was a problem hiding this comment.
My change that is being reverted moved the m_pAppDomainCB initialization to almost the beginning of Debugger::Startup() so it should be initialized.
The change is moving the m_AppDomainCB back to where it was (later in Startup()).
Am I missing something?
There was a problem hiding this comment.
Sorry, I go it backward ... you are right. But I do think that the change of the order here caused the regression somehow.
There was a problem hiding this comment.
And it is likely that we have the same regression on other Linux flavors and just do not see it as often. The ARM emulator is a lot slower than a real machine, so it hits races that are hard to hit elsewhere.
|
Sort of worked; but not reporting success to CI |
|
The error you hit seems to be a build infrastructure problem. I really hope we won't have to merge this reversion. I don't see how this could affect only Linux Arm tests. I don't know how to debug this environment . |
|
CentOS7.1 x64 Release Innerloop Build and Test seems infrastructure |
|
Another infrastructure problem.
…-------- Original message --------
From: Ben Adams <notifications@github.com>
Date: 1/25/18 7:01 AM (GMT-08:00)
To: dotnet/coreclr <coreclr@noreply.github.com>
Cc: Mike McLaughlin <mikem@microsoft.com>, Mention <mention@noreply.github.com>
Subject: Re: [dotnet/coreclr] [WIP] Revert "Add environment variable (COMPlus_EnableDiagnostics) to disab… (#16007)
CentOS7.1 x64 Release Innerloop Build and Test seems infrastructure
First time build. Skipping changelog.
parallel {
Schedule job dotnet_coreclr » master » x64_release_centos7.1_innerloop_prtest
Build dotnet_coreclr » master » x64_release_centos7.1_innerloop_prtest #12 started
dotnet_coreclr » master » x64_release_centos7.1_innerloop_prtest #12 completed
ERROR: Failed to run DSL Script
java.util.concurrent.ExecutionException:
dbees.plugins.flow.JobNotFoundException:
net_coreclr/master/x64_release_windows_nt_innerloop_bld_prtest not found (or isn't a job).
at java.util.concurrent.FutureTask.report(FutureTask.java:122)
at java.util.concurrent.FutureTask.get(FutureTask.java:192)
at java_util_concurrent_Future$get.call(Unknown Source)
at com.cloudbees.plugins.flow.FlowDelegate$_parallel_closure6.doCall(FlowDSL.groovy:456)
at sun.reflect.GeneratedMethodAccessor1199.invoke(Unknown Source)
at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
at java.lang.reflect.Method.invoke(Method.java:498)
at org.codehaus.groovy.reflection.CachedMethod.invoke(CachedMethod.java:93)
at groovy.lang.MetaMethod.doMethodInvoke(MetaMethod.java:325)
at org.codehaus.groovy.runtime.metaclass.ClosureMetaClass.invokeMethod(ClosureMetaClass.java:294)
at groovy.lang.MetaClassImpl.invokeMethod(MetaClassImpl.java:1022)
at groovy.lang.Closure.call(Closure.java:414)
at groovy.lang.Closure.call(Closure.java:430)
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdotnet%2Fcoreclr%2Fpull%2F16007%23issuecomment-360491512&data=02%7C01%7Cmikem%40microsoft.com%7C6795908e93424e10711908d5640475f9%7Cee3303d7fb734b0c8589bcd847f1c277%7C1%7C0%7C636524892675215453&sdata=RqFXbJcZSFNnbI5lBipuUt8U170MZpOrygQxLnhGdPU%3D&reserved=0>, or mute the thread<https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FAJ-Fa2NQM5U6jHBr4Dz1Qohu3HFMQxtGks5tOJcxgaJpZM4RsK0w&data=02%7C01%7Cmikem%40microsoft.com%7C6795908e93424e10711908d5640475f9%7Cee3303d7fb734b0c8589bcd847f1c277%7C1%7C0%7C636524892675215453&sdata=0QVbXG3oMCfybibx8e%2BMKgANLnlF5Qz9VXHM7H5pOkE%3D&reserved=0>.
|
There is a strong evidence that it did affect Linux Arm tests. They started consistently failing right after your change went in; and this trial revert made them pass again. |
@mikem8361 I'm not insisting that merging this is the right fix. I just wanted to demonstrate that #15878 is the likely culprit. I think @jashook or @Samsung can help get you set up to debug in this environment if you can't figure it out with @jkotas thoughts. I would ask that you make it your top priority to investigate this and consider merging this PR if you can't, however. |
| CONTRACTL_END; | ||
|
|
||
| // Don't do anything if there is a native debugger already attached or the debugging support has been disabled. | ||
| if (IsDebuggerPresent() || m_pRCThread == NULL) |
There was a problem hiding this comment.
Aside, should this be reversed?
if (m_pRCThread == NULL || IsDebuggerPresent())|
It shouldn't matter. IsDebuggerPresent a API call to Windows.
|
|
CentOS Release failure is #16016. Merging per above. We can debug as time allows and make another pull request when ready. @lt72 FYI |
…le debugging and profiling. (#15878)"
This reverts commit 5bcfde4.
If the armel CI jobs don't hang on this, then that suggests #15878 is responsible for #15914.
@mikem8361 FYI