On Tue, Sep 07, 2021 at 06:08:44PM +0800, Hongtao Liu wrote:
> -On x86 targets with @code{target("sse2")} and above, GCC supports
> half-precision
> -(16-bit) floating point via the @code{_Float16} type which is defined by
> -18661-3:2015. For C++, x86 provide a builtin type named @code{_Float16}
> -which contains same data format as C.
> -
> -Without @option{-mavx512fp16}, @code{_Float16} type is storage only, all
> -operations will be emulated by software emulation and the @code{float}
> -instructions. The default behavior for @code{FLT_EVAL_METHOD} is to keep
> -the intermediate result of the operation as 32-bit precision. This may lead
> -to inconsistent behavior between software emulation and AVX512-FP16
> +On x86 targets, without @option{-mavx512fp16}, @code{_Float16} type is
I'd add write
targets with SSE2 enabled, without ...
> +storage only, all operations will be emulated by software emulation and the
> +@code{float} instructions. The default behavior for @code{FLT_EVAL_METHOD} is
> +to keep the intermediate result of the operation as 32-bit precision. This
> may
> +lead to inconsistent behavior between software emulation and AVX512-FP16
> instructions.
Ok for trunk with that change, thanks.
Jakub
