GCStress=C not implemented for OS X

[RussKeldorph]

Environment
OS X x64
COMPlus_GCStress=C

Sample Output (all tests result in stack overflow)

BEGIN EXECUTION
/Users/russellk/git/coreclr/bin/tests/Windows_NT.x64.Checked/Tests/coreoverlay/corerun badcodeaftercatch_r.exe
Process is terminated due to StackOverflowException.
./badcodeaftercatch_r.sh: line 117: 45913 Abort trap: 6           $_DebuggerFullPath "$CORE_ROOT/corerun" badcodeaftercatch_r.exe $CLRTestExecutionArguments
Expected: 100
Actual: 134
END EXECUTION - FAILED

[RussKeldorph]

@janvorli @swaroop-sridhar Could one or both of you take a look?

[RussKeldorph]

Apparently GCStress=C isn't implemented for OS X yet.

[RussKeldorph]

@kouvel in case he can comment.

[janvorli]

So, does it mean that the test fails with stack overflow without any GC stress?

[RussKeldorph]

@janvorli Tests I tried passed under GCStress=3. C (presumably 4 and/or 8) were the probem.

[RussKeldorph]

@janvorli Tests also pass with no GCStress set.

[janvorli]

Oh, I guess I got confused a bit by you saying that GCStress=C isn't implemented for OS X yet. So what you meant was that we don't support the mode 0xC, but trying to run with it results in the stack overflow, right? So the fix should be to ignore the modes 4 and 8 on OSX for now.

[RussKeldorph]

@janvorli It is better to crash and burn than silently ignore, but if you would like to print an informative message that 4 and/or 8 aren't supported and refuse to initialize, that would be fine with me for now.

[kouvel]

@adityamandaleeka, it looks like HAVE_GCCOVER used to be defined only on Windows and Linux for CoreCLR and was changed in 0be6c2b. I don't have much context into what the build breaks were, could you please take a look?

[adityamandaleeka]

@kouvel The build breaks mentioned there were on ARM64 coresys builds.

At that time, it was decided that HAVE_GCCOVER being enabled doesn't mean it actually works on that platform, but just that the code compiles.

[janvorli]

I think that having GCStress mode C support is not urgent and we got all the coverage from Linux builds where the GC related stuff is the same as on OSX. So I'd suggest moving it to future.

[adityamandaleeka]

@janvorli Yup, I agree. I meant to move it earlier, but was busy with other work. Moved it now.

[dotnet-policy-service]

Due to lack of recent activity, this issue has been marked as a candidate for backlog cleanup. It will be closed if no further activity occurs within 14 more days. Any new comment (by anyone, not necessarily the author) will undo this process.

This process is part of our issue cleanup automation.

[dotnet-policy-service]

This issue will now be closed since it had been marked no-recent-activity but received no further activity in the past 14 days. It is still possible to reopen or comment on the issue, but please note that the issue will be locked if it remains inactive for another 30 days.