Re: Possible bug in configure script

[Simon Josefsson]

Paul Eggert <eggert@cs.ucla.edu> writes:

> On 1/25/21 7:55 PM, Jeffrey Walton wrote:
>
>>> Jeffrey Walton wrote:
>>>> Placing flags in CC breaks other scripts. For example, this no longer 
>>>> works:
>>>>
>>>>      if [[ ! $(command -v "${CC}") ]]; then
>>>>          echo "The compiler is not valid. Please install a compiler."
>>>>          exit 1
>>>>      fi
>
> You can easily work around this problem by creating whatever compiler
> you like as a script. For example, create an executable file mycc in
> your PATH with this contents:
>
> #!/bin/sh
> exec gcc -m32 "$@"
>
> and then use './configure CC=mycc'.
>
>>> Some flags, like '-m32' or '-m64', MUST be
>>> put in $CC. [1]
>>
>> I place them where they belong - in CFLAGS and CXXFLAGS.
>
> Where they belong is a matter of opinion. In practice, I prefer the
> method that Bruno suggested, as it follows the usual practice for
> cross-compiling. Flags like -m32 are significant enough that they are
> more like using a cross-compiler and so "belong" in CC more than in
> CFLAGS.

There is a long-standing libtool issue with the work-around to use
CC='gcc -static-libgcc', and it is still the preferred way (as far as I
know) to build usable executables for Mingw32:

http://mingw-users.1079350.n2.nabble.com/Building-DLLs-that-don-t-depend-on-libgcc-s-dw2-1-dll-td7582938.html#a7582940

I think it should be fine to have parameters in CC, and I often use that
when I want to force certain compilation behaviour.

/Simon

signature.asc
Description: PGP signature