config-patches
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Patch] configure: Also check C++11 (flags) for ${build} compiler no


From: Joseph Myers
Subject: Re: [Patch] configure: Also check C++11 (flags) for ${build} compiler not only for ${host}
Date: Thu, 20 Aug 2020 17:12:09 +0000
User-agent: Alpine 2.21 (DEB 202 2017-01-01)

On Thu, 13 Aug 2020, Tobias Burnus wrote:

> diff --git a/config/ax_cxx_compile_stdcxx.m4 b/config/ax_cxx_compile_stdcxx.m4
> index 9413da624d2..0cd515fc65b 100644
> --- a/config/ax_cxx_compile_stdcxx.m4
> +++ b/config/ax_cxx_compile_stdcxx.m4
> @@ -25,6 +25,10 @@
>  #   regardless, after defining HAVE_CXX${VERSION} if and only if a
>  #   supporting mode is found.
>  #
> +#   If the fourth argument is the CXX/CXXFLAG/CPPFLAG suffix, e.g.
> +#   "_FOR_BUILD".

It appears you're requiring _FOR_BUILD here and considering other suffixes 
invalid, which would prevent any other use, e.g. _FOR_TARGET.

When building GCC, _FOR_TARGET is of course irrelevant because the 
top-level build support in the source tree is only intended to work with 
the version of GCC in that source tree so can assume what language 
features it supports.  It's less clear that no other suffix will ever be 
relevant elsewhere, given that this is autoconf-archive code rather than 
just used by GCC.

> +  m4_if([$4], [], [],
> +        [$4], [_FOR_BUILD], [],
> +        [m4_fatal([invalid fourth argument `$4' to 
> AX_CXX_COMPILE_STDCXX])])dnl

So I'm not convinced this check that the suffix should be empty or 
_FOR_BUILD is a good idea.

> +  m4_if([$4], [_FOR_BUILD],
> +        [ax_cv_cxx_compile_cxx$1_orig_cxx="$CXX"
> +         ax_cv_cxx_compile_cxx$1_orig_cxxflags="$CXXFLAGS"
> +         ax_cv_cxx_compile_cxx$1_orig_cppflags="$CPPFLAGS"
> +         CXX="$CXX$4"
> +         CXXFLAGS="$CXXFLAGS$4"
> +         CPPFLAGS="$CPPFLAGS$4"])

And then it might be better for this to be a check for the suffix not 
being empty, rather than it being exactly _FOR_BUILD (even if you keep the 
check that other suffixes are invalid, cutting down the number of places 
hardcoding _FOR_BUILD seems a good idea).

>    m4_if([$2], [], [dnl
>      AC_CACHE_CHECK(whether $CXX supports C++$1 features by default,
> -                ax_cv_cxx_compile_cxx$1,
> +                ax_cv_cxx_compile_cxx$1$4,
>        
> [AC_COMPILE_IFELSE([AC_LANG_SOURCE([_AX_CXX_COMPILE_STDCXX_testbody_$1])],
>          [ax_cv_cxx_compile_cxx$1=yes],
>          [ax_cv_cxx_compile_cxx$1=no])])

I think this needs to update the variable name in the assignments of the 
result of the check, and then in the subsequent check for whether to set 
ac_success=yes, not just in the second argument to AC_CACHE_CHECK.

> +  m4_if([$4], [_FOR_BUILD],
> +        [CXX$4="$CXX"
> +         CXXFLAGS$4="$CXXFLAGS"
> +         CPPFLAGS$4="$CPPFLAGS"
> +         CXX="$ax_cv_cxx_compile_cxx$1_orig_cxx"
> +         CXXFLAGS="$ax_cv_cxx_compile_cxx$1_orig_cxxflags"
> +         CPPFLAGS="$ax_cv_cxx_compile_cxx$1_orig_cppflags"])

I think this also would be better checking for $4 not being empty rather 
than for it being exactly _FOR_BUILD.

-- 
Joseph S. Myers
joseph@codesourcery.com



reply via email to

[Prev in Thread] Current Thread [Next in Thread]