Re: Fwd: What is realpath in tools-src? (gcc-12 devel branch)

[john o goyo]

On 2022-02-15 12:34, Gaius Mulley wrote:
> john o goyo <jog37@riddermarkfarm.ca> writes:
>
> > Oops -- I meant to the list, now corrected.
> >
> >
> > -------- Forwarded Message --------
> > Subject:        Re: What is realpath in tools-src? (gcc-12 devel branch)
> > Date:   Mon, 14 Feb 2022 17:03:33 -0500
> > From:   john o goyo <jog37@riddermarkfarm.ca>
> > Reply-To:       jog37@riddermarkfarm.ca
> > To:     Gaius Mulley <mulley.gaius@gmail.com>
> >
> >
> >
> > Greetings, Gaius.
> >
> > On 2022-02-14 13:09, Gaius Mulley wrote:
> > > Hi John,
> > >
> > > sure realpath can be used to determine the absolute path. Is it a GNU ism?
> > Thank you, Gaius.
> >
> > It is indeed a GNUism and part of GNU core-utils.  (There is a POSIX
> > realpath() function.)
> > NetBSD installs it as grealpath.  I modified the calcpath script to
> > call grealpath.
> >
> > But now I have a subsequent build problem:
> >
> > checking for flex... flex
> > checking for gcc... /home/build/opt/gcc/bld/gm2/./prev-gcc/xgcc
> > -B/home/build/opt/gcc/bld/gm2/./prev-gcc/
> > -B/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/bin/
> >   -B/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/bin/
> > -B/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/lib/ -isystem
> >   /home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/include -isystem
> > /home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/sys-include
> > -fno-checking
> > checking whether the C compiler works... no
> > configure: error: in `/home/build/opt/gcc/bld/gm2/gcc/m2':
> > configure: error: C compiler cannot create executables
> > See `config.log' for more details
> > gmake[3]: *** [../../../src/gcc-git/gcc/m2/Make-lang.in:1204:
> > m2/gm2config.h] Error 7
> >
> > The file gcc/m2/config.log aprubtly ends "configure: exit 77" with no
> > indication of what it tried to compile.
> >
> > Sincerely,
> > john
> Hi John,
>
> I wonder if the above error might be related to the other .so.6 C++
> problem?
>
> What does:
>
> cat << EOF > t.c
> main () {}
> EOF
>
> /home/build/opt/gcc/bld/gm2/./prev-gcc/xgcc \
>   -B/home/build/opt/gcc/bld/gm2/./prev-gcc/  \
>   -B/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/bin/ \
>    -B/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/bin/  \
>   -B/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/lib/ -isystem \
>    /home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/include -isystem \
>   /home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/sys-include \
>   -fno-checking t.c
> generate?
Well, that is the odd thing.  Witness:

duplex$ cat compile.gm2
/home/build/opt/gcc/bld/gm2/./prev-gcc/xgcc 
-B/home/build/opt/gcc/bld/gm2/./prev-gcc/ 
-B/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/bin/ 
-B/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/bin/ 
-B/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/lib/ -isystem 
/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/include -isystem 
/home/build/gm2/2022-02-11/sparc64-unknown-netbsd9.1/sys-include 
-fno-checking $*
duplex$ cat what.c
#include <stdio.h>

int
main(void)
{
  puts("What?");
  return 0;
}
duplex$ ./compile.gm2 what.c
duplex$ file a.out
a.out: ELF 64-bit MSB executable, SPARC V9, relaxed memory ordering, 
version 1 (SYSV), dynamically linked, interpreter 
/usr/libexec/ld.elf_so, for NetBSD 9.1, not stripped
duplex$ ./a.out
What?
duplex$ echo $?
0

Sincerely,
john