[Bug driver/122429] [16 Regression] libatomic vs glibc build issue

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=122429

--- Comment #4 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #3)
> 
> The only special option I do for glibc is set `CXX=false` before configuring
> glibc (due to https://sourceware.org/bugzilla/show_bug.cgi?id=24183).

This is my stage1 gcc build:
```
../configure --target=${TARGET} --prefix=${STAGE1_DIR}  ${CONFG_OPTIONS}
--enable-languages=c \
  --disable-shared --with-sysroot=${STAGE1_SYSROOT} --enable-initfini-array \
  --disable-libssp --disable-libcilkrts --without-headers --with-newlib
--disable-threads \
  --disable-decimal-float --disable-gcov --disable-libffi --disable-libatomic \
  --disable-libgomp --disable-libitm --disable-libmpx --disable-libquadmath
--disable-libsanitizer

make -j${CPUS} || exit 1
make install || exit 1
```

glibc is:
```
# set CXX to workaround https://sourceware.org/bugzilla/show_bug.cgi?id=24183
CXX=false \
../configure --prefix=/usr --host=${TARGET}
make -j24
make install install_root=${SYSROOT}
```