| commit | f16c4324a7e28c929f7226dd06fef87d951431c9 | [log] [tgz] |
|---|---|---|
| author | Patryk Duda <pdk@semihalf.com> | Mon Jul 17 16:23:55 2023 +0200 |
| committer | Carles Cufí <carles.cufi@nordicsemi.no> | Tue Aug 01 11:24:07 2023 +0200 |
| tree | 81a341aea7b077d6f31d3a2a2ad471784318277e | |
| parent | cf55f8329ede3fb533d7fac28bf95c3c13f04e2c [diff] |
clang: Don't specify FP16 format Support for 16 bit floats is enabled by default in Clang [1]. The ARM alternative format is not supported, so __fp16 always uses IEEE 754-2008 format [2]. [1] https://github.com/llvm/llvm-project/blob/main/llvm/lib/Target/ARM/ARMAsmPrinter.cpp#L750-L755 [2] https://clang.llvm.org/docs/LanguageExtensions.html#half-precision-floating-point Fixes: #55562 Signed-off-by: Patryk Duda <pdk@semihalf.com>