The expected output uses a hardcoded value for
maxBound :: Int.
This should fix one of circleci failures on i386.
Oh dear, I forgot all about architecture size differences. My bad.
That being said, I don't think it's necessary to skip this test on 32-bit architectures entirely. We could adopt a similar trick as other tests that print out maxBound: have different expected stdout files for difference architectures. For example, see T4340.stdout-ws-32 and T4340.stdout-ws-64.