BUG: fixed test that made bad assumptions of GetGlobalDefaultNumberOf…

[seanm]

…Threads()

[seanm]

@dzenanz how would I solve the case with the ITK_TEST_SET_GET_VALUE macro? I'm not sure what to use instead...

[N-Dekker]

Thanks @seanm !

• For the record, a slightly related issue: GlobalDefaultNumberOfThreads should not be greater than GlobalMaximumNumberOfThreads #4773

[seanm]

There's one more failing test that I hope to figure out and add here...

[seanm]

OK, I don't get it... need help :)

It's this code:

ITK/Modules/Core/Common/test/itkMultiThreaderExceptionsTest.cxx

Line 102 in 5df209f

ITK_TRY_EXPECT_EXCEPTION(dummySrc->Update());

The test fails with Failed to catch expected exception.

The call to dummySrc->SetNumberOfWorkUnits(4) will be clamped down from 4 to 2 (my ITK_DEFAULT_MAX_THREADS), so I thought changing the hardedcoded for loop from 4 would fix it, but alas no (though that change maybe necessary too, but insufficient).

[dzenanz]

Nothing jumps at me. I guess some digging is needed into DummyImageSource.

[seanm]

I took a wild guess: there was another 4 in the code. Changing that fixed it.

[seanm]

If this looks good to y'all, would be nice to merge, as it will green up cdash a bit.

[seanm]

Seems this broke the build, depending on build options: https://open.cdash.org/viewBuildError.php?buildid=10179364

fix: #5200