On 7/22/20 11:47 AM, Dieter, William R wrote:
Is splitting compilation into two steps to eliminate the compiler output a reasonable fix for this?
It might be, yes. Though that will take a bit of work to get right. In the meantime (since we're trying to get a new release out) I installed the attached, which I hope works around your particular problem. Please give it a try.
>From ae26b9b44c183c41b84fc36ff70082713295d621 Mon Sep 17 00:00:00 2001 From: Paul Eggert <egg...@cs.ucla.edu> Date: Wed, 22 Jul 2020 12:56:19 -0700 Subject: [PATCH] Port AC_F77_LIBRARY_LDFLAGS to oneAPI HPC Toolkit MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Problem reported by Bill Dieter in: https://lists.gnu.org/r/bug-autoconf/2020-07/msg00089.html * lib/autoconf/fortran.m4 (_AC_FC_LIBRARY_LDFLAGS): Defend against ‘clang -mllvm -loopopt=0’. --- lib/autoconf/fortran.m4 | 1 + 1 file changed, 1 insertion(+) diff --git a/lib/autoconf/fortran.m4 b/lib/autoconf/fortran.m4 index 0a0c4168..59a8859b 100644 --- a/lib/autoconf/fortran.m4 +++ b/lib/autoconf/fortran.m4 @@ -698,6 +698,7 @@ while test $[@%:@] != 1; do -zallextract*| -zdefaultextract) ac_cv_[]_AC_LANG_ABBREV[]_libs="$ac_cv_[]_AC_LANG_ABBREV[]_libs $ac_arg" ;; + -mllvm) ${2+shift};; # Defend against 'clang -mllvm -loopopt=0'. # Ignore everything else. esac done -- 2.17.1
