bitstring_test: fix bit_nclear/bit_nset tests for last iteration

[rlibby]

Summary

Diagnosed by gcc -Wint-in-bool-context.

Test Plan

% kyua test -k /usr/tests/sys/Kyuafile sys/bitstring_test

[asomers]

What's the problem? Under what conditions is the current code incorrect?

[rlibby]

Apologies, trying to recall. IIRC...

GCC complained because the 2nd arg to the ternary is bool and the 3rd is int. Here it mean that the behavior for j + 1 == nbits wasn't being tested, but we should test that the result is -1.

(j + 1 < nbits) ? found_set_bit == j + 1 : -1,
                  ^^^^^^^^^^^^^^^^^^^^^^   ^^
                            |               |
true or false, depending on result          |
                                            |
always true---------------------------------|

[asomers]

So is -1 being cast to true? Thank you GCC for telling us that. If so, I think we should replace ATF_REQUIRE_MSG with ATF_REQUIRE_EQ_MSG. I think that would be more clear.

[rlibby]

In D45893#1207280, @asomers wrote:

So is -1 being cast to true?

Effectively, yes.

Thank you GCC for telling us that. If so, I think we should replace ATF_REQUIRE_MSG with ATF_REQUIRE_EQ_MSG. I think that would be more clear.

Sure, I'll work that up.

[rlibby]

asomers feedback: use ATF_REQUIRE_EQ family of tests.

I converted the nearby tests to ATF_REQUIRE_INTEQ_MSG, but I left the 40
other instances in this file alone.

I selected INTEQ instead of EQ as that variant outputs the actual
values on failure.