Marco Rossignoli

> How common it is that APIs in question will complete synchronously? Do we have that data? We don't have info about it, also it's not easy, vast majority of...

I don't see the image and the mermaid rendering correctly is it related only to the "preview"?

> Maybe an additional test in [DotnetTool.cs](https://github.com/coverlet-coverage/coverlet/blob/master/test/coverlet.integration.tests/DotnetTool.cs) should verify the new command line option. I agree could be good add some tests for it

> Therfore I added now a new runsettings option called MergeDirectory to store intermediate results. The idea behind that is that everytime the attachment processor triggers all attachments are merged...

Hi Oren, I did some check and it's related to instrumentation, that test seem a performance test comparison https://github.com/onovotny/bc-csharp/blob/netstandard/crypto/test/src/math/ec/test/ECPointPerformanceTest.cs#L21 that lead to a lot of call to [instrumentation code](https://github.com/tonerdo/coverlet/blob/master/src/coverlet.core/Instrumentation/ModuleTrackerTemplate.cs#L68) ...

> If you run the tests several times, it'll eventually fail with that. You mean more than one time in same CI loop(to undestand if could be related to instrumenting...

Can you try to "cleanup"(git -fdx) between tests and check if it happens?

IIRW we had some issue where libs were instrumented more than once leading to weird results.

> Hmm....running twice without a clean does appear to trigger the errors. Seems like it's over-instrumenting or breaking something that way? At the moment we don't check if module is...

Tell me where you put `ExcludeFromCodeCoverage` I would try to repro on my local(or if you've branch/commit to clone with above setup).