Re: [PATCH] LoongArch: Add soft floating-point fe* function implementations.

[Joseph Myers <josmyers@redhat.com>]

[-- Attachment #1: Type: text/plain, Size: 1636 bytes --]

On Sun, 31 Mar 2024, caiyinyu wrote:

> LoongArch soft float support was added in version 2.37, but the fe* functions
> were not completely added. I want to backport this patch if possible
> (it requires some modifications) to the 2.37/38/39/master branch, so the
> version of
> the relevant symbols in  Versions is GLIBC_2.37.

I don't think such new symbol additions are appropriate for backporting; 
the GLIBC_2.37 symbol version contents should have been fixed at the time 
2.37 was released.

> diff --git a/sysdeps/loongarch/nofpu/Versions
> b/sysdeps/loongarch/nofpu/Versions
> index 6d7f6aae6c..73b6a71bbd 100644
> --- a/sysdeps/loongarch/nofpu/Versions
> +++ b/sysdeps/loongarch/nofpu/Versions
> @@ -1,5 +1,13 @@
>  libc {
> +  GLIBC_2.37 {
> +    __adddf3; __addsf3; __divdf3; __divsf3; __eqdf2; __eqsf2;
> +    __extendsfdf2; __fixdfsi; __fixsfsi;
> +    __fixunsdfsi; __fixunssfsi;
> +    __floatsidf; __floatsisf;
> +    __gedf2; __gesf2; __ledf2; __lesf2; __muldf3; __mulsf3;
> +    __negdf2; __negsf2; __subdf3; __subsf3; __truncdfsf2;
> +  }

I'd tend to expect you also to have the symbols that powerpc-nofpu has at 
GLIBC_2.4 version.

    __floatundidf; __floatundisf;
    __floatunsidf; __floatunsisf;
    __unorddf2; __unordsf2;
    __nedf2; __nesf2;
    __gtdf2; __gtsf2;
    __ltdf2; __ltsf2;

Also, as LoongArch uses binary128 long double, I'd expect all the TFmode 
symbols as well.  And as a 64-bit architecture, I'd expect TImode symbols 
(conversions between floating modes and 128-bit signed and unsigned 
integers, in both directions) as well.

-- 
Joseph S. Myers
josmyers@redhat.com