[libc-commits] [libc] [libc][mathvec] Vectorise cosf (PR #224290)

[via libc-commits]

================
@@ -17,12 +18,56 @@
 
 #include "hdr/math_macros.h"
 
-#define EXPECT_SIMD_EQ(REF, RES)                                               \
+namespace LIBC_NAMESPACE_DECL {
+namespace testing {
+
+template <typename T>
+inline bool within_ulp_tolerance(T expected, T actual, uint64_t tolerance) {
+  fputil::FPBits<T> expected_bits(expected), actual_bits(actual);
+
+  // Handle inf and nan cases.
+  if (expected_bits.is_inf() || expected_bits.is_inf())
----------------
DylanFleming-arm wrote:

Oh whoops. I completely did mean that yes, That's my bad :')

I'll fix it now, thanks.

https://github.com/llvm/llvm-project/pull/224290