diff options
| author |  Niels Dekker <N.Dekker@lumc.nl> | 2020-07-11 12:50:46 +0200 |
|---|---|---|
| committer | Niels Dekker <N.Dekker@lumc.nl> | 2020-07-11 13:32:28 +0200 |
| commit | 4ab32e2de2511746e2108563a43cbbeb1922fbf2 (patch) | |
| tree | a06f6a180f538ce9dfd79138bd7c2efbf5c1e65b /test/bfloat16_float.cpp | |
| parent | dcf7655b3d469a399c1182f350c9009e13ad8654 (diff) | |
Allow implicit conversion from bfloat16 to float and double
Conversion from `bfloat16` to `float` and `double` is lossless. It seems natural to allow the conversion to be implicit, as the C++ language also support implicit conversion from a smaller to a larger floating point type.
Intel's OneDLL bfloat16 implementation also has an implicit `operator float()`: https://github.com/oneapi-src/oneDNN/blob/v1.5/src/common/bfloat16.hpp
Diffstat (limited to 'test/bfloat16_float.cpp')
| -rw-r--r-- | test/bfloat16_float.cpp | 6 |
1 files changed, 3 insertions, 3 deletions
diff --git a/test/bfloat16_float.cpp b/test/bfloat16_float.cpp index eb55f7d45..478aef3a3 100644 --- a/test/bfloat16_float.cpp +++ b/test/bfloat16_float.cpp @@ -53,9 +53,9 @@ void test_conversion() VERIFY_IS_EQUAL(bfloat16(3.40e38f).value, 0x7f80); // Becomes infinity. // Verify round-to-nearest-even behavior. - float val1 = static_cast<float>(bfloat16(__bfloat16_raw(0x3c00))); - float val2 = static_cast<float>(bfloat16(__bfloat16_raw(0x3c01))); - float val3 = static_cast<float>(bfloat16(__bfloat16_raw(0x3c02))); + float val1 = bfloat16(__bfloat16_raw(0x3c00)); + float val2 = bfloat16(__bfloat16_raw(0x3c01)); + float val3 = bfloat16(__bfloat16_raw(0x3c02)); VERIFY_IS_EQUAL(bfloat16(0.5f * (val1 + val2)).value, 0x3c00); VERIFY_IS_EQUAL(bfloat16(0.5f * (val2 + val3)).value, 0x3c02); |