ports/187732: math/reduce compilation still needs MAKE_JOBS_UNSAFE=yes

[Mamoru Iwaki]

>Number:         187732
>Category:       ports
>Synopsis:       math/reduce compilation still needs MAKE_JOBS_UNSAFE=yes
>Confidential:   no
>Severity:       non-critical
>Priority:       low
>Responsible:    freebsd-ports-bugs
>State:          open
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   current-users
>Arrival-Date:   Wed Mar 19 11:20:01 UTC 2014
>Closed-Date:
>Last-Modified:
>Originator:     Mamoru Iwaki
>Release:        FreeBSD  10.0-STABLE FreeBSD 10.0-STABLE #0 r263331
>Organization:
>Environment:
FreeBSD  10.0-STABLE FreeBSD 10.0-STABLE #0 r263331: Wed Mar 19 14:26:49 JST 2014     root@:/mnt2/usr/obj/nanobsd.full/mnt2/usr/src/sys/MYGEN  amd64
>Description:
Re: # $FreeBSD: head/math/reduce/Makefile 344974 2014-02-18 22:16:33Z stephen $

Compilation of math/reduce has failed these days.  I'm using a multicore machine as follows: 

   FreeBSD/SMP: Multiprocessor System Detected: 8 CPUs
   FreeBSD/SMP: 1 package(s) x 4 core(s) x 2 SMT threads

I'm not sure but it seems to require MAKE_JOBS_UNSAFE=yes in Makefile.  


oot@:/usr/ports/math/reduce # make
===>  License BSD accepted by the user
===>   reduce-20110414 depends on file: /usr/local/sbin/pkg - found
===> Fetching all distfiles required by reduce-20110414 for building
===>  Extracting for reduce-20110414
=> SHA256 Checksum OK for reduce-src-20110414.tar.bz2.

(snip-snip-snip)

mv -f .deps/u01.Tpo .deps/u01.Po
cc -DHAVE_CONFIG_H -I. -I/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase  -Dfontsdir=reduce.fonts -I/usr/local/include -I/usr/local/include/freetype2 -I/usr/local/include -I/usr/local/include/freetype2 -D_THREAD_SAFE -I/usr/local/include   -DBINDIR=/usr/local/bin -DPKGDATADIR=/usr/local/share/reduce  -O2 -pipe -fno-strict-aliasing -I/usr/local/include -I/usr/local/include/freetype2 -I/usr/local/include -I/usr/local/include/freetype2 -D_THREAD_SAFE -I/usr/local/include -fno-strict-aliasing -O2 -Wall   -MT u06.o -MD -MP -MF .deps/u06.Tpo -c -o u06.o `test -f '/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u06.c' || echo '/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/'`/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u06.c
cc -DHAVE_CONFIG_H -I. -I/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase  -Dfontsdir=reduce.fonts -I/usr/local/include -I/usr/local/include/freetype2 -I/usr/local/include -I/usr/local/include/freetype2 -D_THREAD_SAFE -I/usr/local/include   -DBINDIR=/usr/local/bin -DPKGDATADIR=/usr/local/share/reduce  -O2 -pipe -fno-strict-aliasing -I/usr/local/include -I/usr/local/include/freetype2 -I/usr/local/include -I/usr/local/include/freetype2 -D_THREAD_SAFE -I/usr/local/include -fno-strict-aliasing -O2 -Wall   -MT u07.o -MD -MP -MF .deps/u07.Tpo -c -o u07.o `test -f '/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u07.c' || echo '/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/'`/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u07.c
mv -f .deps/u04.Tpo .deps/u04.Po
/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u06.c:3131:40: error: cc -DHAVE_CONFIG_H -I. -I/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase  -Dfontsdir=reduce.fonts -I/usr/local/include -I/usr/local/include/freetype2 -I/usr/local/include -I/usr/local/include/freetype2 -D_THREAD_SAFE -I/usr/local/include   -DBINDIR=/usr/local/bin -DPKGDATADIR=/usr/local/share/reduce  -O2 -pipe -fno-strict-aliasing -I/usr/local/include -I/usr/local/include/freetype2 -I/usr/local/include -I/usr/local/include/freetype2 -D_THREAD_SAFE -I/usr/local/include -fno-strict-aliasing -O2 -Wall   -MT u08.o -MD -MP -MF .deps/u08.Tpo -c -o u08.o `test -f '/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u08.c' || echo '/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/'`/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u08.c
expected
      expression
    v118 = (*qfn2(fn))(qenv(fn), v119,
                                       ^
/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u06.c:3131:40: error: expected '}'
/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u06.c:3095:1: note: to match this '{'
{
^
/mnt2/usr/ports/math/reduce/work/reduce-20110414/csl/cslbase/../../cslbuild/generated-c/u06.c:3121:27: error: use of
      undeclared label 'v42'
    if (v118 == nil) goto v42;
                          ^
3 errors generated.
gmake[4]: *** [u06.o] Error 1
gmake[4]: *** Waiting for unfinished jobs....
mv -f .deps/u07.Tpo .deps/u07.Po
mv -f .deps/u05.Tpo .deps/u05.Po
mv -f .deps/u08.Tpo .deps/u08.Po
gmake[3]: *** [all] Error 2
gmake[2]: *** [all] Error 2
gmake[1]: Leaving directory `/mnt2/usr/ports/math/reduce/work/reduce-20110414'



NOTE: /usr/ports is a symlink to /mnt2/usr/ports in my environment

>How-To-Repeat:
Simply try compiling the ports on a many core machine.

cd /usr/port/math/reduce
make
(No option is changed)

>Fix:
I made this patch to inhibit the number of jobs.  It should get right compilation order back as expected.  It works for me anyway.  


--- Makefile.orig	2014-03-12 20:32:27.000000000 +0900
+++ Makefile	2014-03-19 19:30:39.000000000 +0900
@@ -28,6 +28,8 @@
 CONFIGURE_ARGS=	--with-csl
 ALL_TARGET=
 
+MAKE_JOBS_UNSAFE=	yes
+
 .include <bsd.port.pre.mk>
 
 HMAN1=	redcsl.1


Patch attached with submission follows:

--- Makefile.orig	2014-03-12 20:32:27.000000000 +0900
+++ Makefile	2014-03-19 19:30:39.000000000 +0900
@@ -28,6 +28,8 @@
 CONFIGURE_ARGS=	--with-csl
 ALL_TARGET=
 
+MAKE_JOBS_UNSAFE=	yes
+
 .include <bsd.port.pre.mk>
 
 HMAN1=	redcsl.1


>Release-Note:
>Audit-Trail:
>Unformatted: