Fix compilation of fullblock closures

[tesonep]

Taking all the changes I have done to fix some of the problems.

@MarcusDenker
@StevenCostiou
@guillep

[Ducasse]

Ouch not build passes.

[Ducasse]

None of the builds built.

and we have + ./pharo Pharo.image test --junit-xml-output --stage-name=Tests-osx-32 '.*'
2019-09-04 14:10:08.641 Pharo[21139:507] ApplePersistenceIgnoreState: Existing state will not be touched. New state will be written to /var/folders/w0/nn_htp7n6ddcydvkv21hjzhc0000gn/T/org.pharo.Pharo.savedState
2019-09-04 14:10:08.892 Pharo[21139:507] 14:10:08.892 ERROR: [0xa146e1a8] AQMEIO.cpp:377: _FindIOUnit: error -66680
2019-09-04 14:10:08.892 Pharo[21139:507] 14:10:08.892 ERROR: [0xa146e1a8] >aq> AudioQueueObject.cpp:1590: Prime: failed (-66680); will stop (5512/0 frames)
2019-09-04 14:10:08.893 Pharo[21139:507] 14:10:08.893 ERROR: [0xa146e1a8] AQMEIO.cpp:377: _FindIOUnit: error -66680
2019-09-04 14:10:08.893 Pharo[21139:507] 14:10:08.893 ERROR: [0xa146e1a8] >aq> AudioQueueObject.cpp:1590: Prime: failed (-66680); will stop (33075/0 frames)

[MarcusDenker]

There is a conflict, so the build builds the old version without the fix that was merged yesterday

[Ducasse]

ouch 99 broken tests, so what do we do?

[MarcusDenker]

a lot of them are the decompiler. The decompiler does not yet support fullblock closures.

[Ducasse]

So how do we proceed?
Do we have a doc, comment, description of the new blockclosure?

[MarcusDenker]

I will first make a list of all failing tests. (I will run the tests a couple of times, some seem to not be related)

I already saw that some tests in MethodMapTest are failing, I will look at those next

[MarcusDenker]

The crash on linux might be due to a not up-to-date implementation of #rfEnsure:, I did a PR to fix it: #4542

(when it is integrated I will close/reopen this PR so we build with the fix included)

[MarcusDenker]

There are more problems with this method... I will continue to debug.

[Ducasse]

It would be nice to see if we can pair program because I would like to learn the decompiler.

[MarcusDenker]

Yes, let's plan do so some pair programming from next week on on this

[Ducasse]

I cannot wednesday afternoon (CST thomas)and thursday full afternoon

[MarcusDenker]

The Reflectivity related crash is now fixed. This is a cleaned up list of the failing test:

FFI:

testCqsort – Unix32.UnifiedFFI.Tests.Tests.FFICallbackTest
testCqsortWithByteArray – Unix32.UnifiedFFI.Tests.Tests.FFICallbackTest

Debugger:

testInterruptedContext – Unix32.Debugger.Tests.DebugSessionContexts2Test
testContextsAfterStepInto – Unix32.Debugger.Tests.DebugSessionContexts2Test
testStepThrough – Unix32.Debugger.Tests.StepThroughTest

Decompiler:
- FBDDecompilerTest
- OCBytecodeDecompilerExamplesTest

Fuel:
- FLBlockClosureSerializationTest

Opal (Mapping):

testExampleTempNamedTempVectorNestedBlock – Unix32.OpalCompiler.Tests.Misc.MethodMapTest
testSourceMappingBlock – Unix32.OpalCompiler.Tests.Misc.MethodMapTest
testDNU – Unix32.OpalCompiler.Tests.Misc.MustBeBooleanTest

Misc:

testAllExamples – Unix32.Hiedra.Tests.Model.HiExamplesTest
testTerminatingBlockedCriticalSectionShouldNotUnblockAnotherWaitingSection –
Unix32.Kernel.Tests.Extended.Processes.MutexTest
testClosureRestart – Unix32.Kernel.Tests.Methods.ContextTest
testMixedMethod – Unix32.Kernel.Tests.Processes.LocalRecursionStopperTest
testPointersTo – Unix32.Kernel.Tests.Objects.ProtoObjectTest
testPointersToCycle – Unix32.Kernel.Tests.Objects.ProtoObjectTest
testMixedMethod – Unix32.Kernel.Tests.Processes.RecursionStopperTest
testInCriticalWait – Unix32.Kernel.Tests.Processes.SemaphoreTest
testMethodsEnumeration – Unix32.Calypso.SystemPlugins.Traits.Queries.Tests.ClyTraitUserScopeTest
testMethodsEnumerationWhenBothMetaLevels – Unix32.Calypso.SystemPlugins.Traits.Queries.Tests.ClyTraitUserScopeTest

[stale]

It is hard to review old PRs so this PR has been marked as stale since there has been no activity the last 20 days. It will be closed in 10 days if no further activity occurs. A simple comment will reactivate the PR, but please also consider updating the PR to the latest SNAPSHOT build to make it easier for reviewers.

[MarcusDenker]

I suggest to do these things:

1. commit all but the setup change. I will make a new PR for that.

2. do the switch to the Sista bytemode backend first without new blocks.

3. then we make a new PR for testing the change to new blocks

[Ducasse]

Gogogo

[MarcusDenker]

I have made a PR with all the image level changes: #5076

I will close this and open a new one for actually testing making it default.