[ADT] [NFC] Resolve a potential overload ambiguity in the ADT unit tests

[wolfy1961]

Some compilers (ex. VS2019 and VS2022) report an overload resolution error with the APInt constructor.
Resolve the ambiguity with a static_cast.

[llvmbot]

@llvm/pr-subscribers-llvm-adt

Author: Wolfgang Pieb (wolfy1961)

Changes

Some compilers (ex. VS2019 and VS2022) report an overload resolution error with the APInt constructor.
Resolve the ambiguity with a static_cast.

---

Full diff: https://github.com/llvm/llvm-project/pull/85341.diff

1 Files Affected:

• (modified) llvm/unittests/ADT/APIntTest.cpp (+1-1)

diff --git a/llvm/unittests/ADT/APIntTest.cpp b/llvm/unittests/ADT/APIntTest.cpp
index 400313a6234677..c3805431e6b2b6 100644
--- a/llvm/unittests/ADT/APIntTest.cpp
+++ b/llvm/unittests/ADT/APIntTest.cpp
@@ -2918,7 +2918,7 @@ TEST(APIntTest, Average) {
   APInt A100(32, 100);
   APInt A101(32, 101);
   APInt A200(32, 200, false);
-  APInt ApUMax(32, UINT_MAX, false);
+  APInt ApUMax(32, static_cast<uint64_t>(UINT_MAX), false);
 
   EXPECT_EQ(APInt(32, 150), APIntOps::avgFloorU(A100, A200));
   EXPECT_EQ(APIntOps::RoundingUDiv(A100 + A200, A2, APInt::Rounding::DOWN),

[kuhar]

How about

Suggested change

	APInt ApUMax(32, static_cast<uint64_t>(UINT_MAX), false);
	APInt ApUMax(32, static_cast<uint64_t>(std::numeric_limits<uint32_t>::max()), false);

or

Suggested change

	APInt ApUMax(32, static_cast<uint64_t>(UINT_MAX), false);
	APInt ApUMax(32, (~(uint64_t(0)) >> 32, false);

to avoid a system-specific definition of uint?

[RKSimon]

This would be my preferred option:

APInt ApUMax = APInt::getMaxValue(32);

[RKSimon]

I ended up pushing a fix here to unblock some other people upstream: 41bdcaa