Re: [PATCH 2/2] newlib: add AC_CACHE_CHECK sugar around preprocessor checks

[Corinna Vinschen <vinschen@redhat.com>]

On Jan 28 05:23, Mike Frysinger wrote:
> This isn't strictly necessary, but it makes for much clearer logs as
> to what the target is doing, and provides cache vars for anyone who
> wants to force the test a different way.
> ---
>  newlib/libc/configure                  | 48 ++++++++++++++++++--------
>  newlib/libc/machine/nds32/acinclude.m4 |  7 ++--
>  newlib/libc/machine/sh/acinclude.m4    |  7 ++--
>  newlib/libc/machine/spu/acinclude.m4   |  9 +++--
>  newlib/libm/configure                  | 32 ++++++++++++-----
>  newlib/libm/machine/nds32/acinclude.m4 | 14 ++++----
>  6 files changed, 78 insertions(+), 39 deletions(-)
> 
> diff --git a/newlib/libc/machine/nds32/acinclude.m4 b/newlib/libc/machine/nds32/acinclude.m4
> index 35c2afe992bf..ae8ea5d91ecf 100644
> --- a/newlib/libc/machine/nds32/acinclude.m4
> +++ b/newlib/libc/machine/nds32/acinclude.m4
> @@ -1,10 +1,11 @@
>  if test "${machine_dir}" = "nds32"; then
>    dnl Use builtin macro to detect if this is for "AndeStar ISA V3m".
> -  AC_PREPROC_IFELSE([AC_LANG_PROGRAM(
> +  AC_CACHE_CHECK([for nds32 V3M ISA], newlib_cv_nds32_isa_v3m, [dnl
> +    AC_PREPROC_IFELSE([AC_LANG_PROGRAM(
>  [[#ifdef __NDS32_ISA_V3M__
>  # error "This is nds32_isa_v3m."
>  #endif
> -]])], [is_nds32_isa_v3m="no"], [is_nds32_isa_v3m="yes"])
> +]])], [newlib_cv_nds32_isa_v3m="no"], [newlib_cv_nds32_isa_v3m="yes"])])
          ^^^^^^^^^^
          libc_cv, perhaps, to keep the number of changes smaller,
          depending on the outcome of the discussion you started in
	  https://sourceware.org/pipermail/newlib/2022/019018.html?


Corinna