[Clang] use unsigned integer constants in unit-test | fixes build error on ppc64le-lld-multistage-test

[kiloalphaindia]

Summary

Fixes:

/home/buildbots/ppc64le-lld-multistage-test/ppc64le-lld-multistage-test/llvm-project/third-party/unittest/googletest/include/gtest/gtest.h:1526:11: warning: comparison of integer expressions of different signedness: ‘const unsigned int’ and ‘const int’ [-Wsign-compare]
/home/buildbots/ppc64le-lld-multistage-test/ppc64le-lld-multistage-test/llvm-project/third-party/unittest/googletest/include/gtest/gtest.h:1526:11: warning: comparison of integer expressions of different signedness: ‘const long unsigned int’ and ‘const int’ [-Wsign-compare]

[kiloalphaindia]

See: https://lab.llvm.org/buildbot/#/builders/57/builds/28680

I couldn't reproduce this on my system. Anyone knows how to test it before merging?

[kiloalphaindia]

Also affects: clang-ppc64-aix
https://lab.llvm.org/buildbot/#/builders/214/builds/8649

[cor3ntin]

Thanks for the quick fix. Any idea why it would only show up on ppc64 @aaron.ballman ?

[kiloalphaindia]

It could also be a different version of gtest or so.

[aaron.ballman]

In D156224#4531659, @cor3ntin wrote:

Thanks for the quick fix. Any idea why it would only show up on ppc64 @aaron.ballman ?

Not certain, to be honest. The diagnostic is enabled by -Wextra so I would have imagined this was enabled all over. My guess is that this bot was configured with -Werror whereas other bots showed the issue but as a warning.

Changes LGTM

[cor3ntin]

I was able to confirm on my machine that the warnings are no longer emitted. They were there when i committed the initial patch, i just missed them. Sorry about that

[kiloalphaindia]

I had copied the warning-flags from the build-bot command line and ended up as follows:

-Werror -Werror=date-time -Werror=unguarded-availability-new -Wall -Wextra -Wno-unused-parameter -Wwrite-strings -Wcast-qual -Wmissing-field-initializers -pedantic -Wno-long-long -Wc++98-compat-extra-semi -Wimplicit-fallthrough -Wcovered-switch-default -Wno-noexcept-type -Wnon-virtual-dtor -Wdelete-non-virtual-dtor -Wsuggest-override -Wstring-conversion -Wmisleading-indentation -Wctad-maybe-unsupported -Woverloaded-virtual -Wno-nested-anon-types -O2 -DNDEBUG -Wno-variadic-macros -Wno-gnu-zero-variadic-macro-arguments -Wno-suggest-override

This obviously includes -Wextra ...
Still wondering, why I couldn't see the warning.
Which version of GTEST are you using?

[shafik]

clang/unittests/libclang/LibclangTest.cpp

1232

Above you used 0u but here you changed the type to int, is there a reason why not to use 0u here as well?

clang/unittests/libclang/LibclangTest.cpp

 Message message;
 static_assert(true, message);
 )cpp";
   std::string fileName = "main.cpp";
   WriteFile(fileName, testSource);
   const char *Args[] = {"-xc++", "-std=c++26"};
   ClangTU = clang_parseTranslationUnit(Index, fileName.c_str(), Args,
                                        std::size(Args), nullptr, 0, TUFlags);
-  ASSERT_EQ(clang_getNumDiagnostics(ClangTU), 0);
+  ASSERT_EQ(clang_getNumDiagnostics(ClangTU), 0u);
   std::optional<CXCursor> staticAssertCsr;
   Traverse([&](CXCursor cursor, CXCursor parent) -> CXChildVisitResult {
     if (cursor.kind == CXCursor_StaticAssert) {
       staticAssertCsr.emplace(cursor);
     }
     return CXChildVisit_Continue;
   });
   ASSERT_TRUE(staticAssertCsr.has_value());
-  size_t argCnt = 0;
+  int argCnt = 0;
   Traverse(*staticAssertCsr, [&argCnt](CXCursor cursor, CXCursor parent) {
     switch (argCnt) {
     case 0:
       EXPECT_EQ(cursor.kind, CXCursor_CXXBoolLiteralExpr);
       break;
     case 1:
       EXPECT_EQ(cursor.kind, CXCursor_DeclRefExpr);
       break;