Date: Thu, 19 Apr 2012 22:11:34 -0500 From: "Conrad J. Sabatier" <conrads@cox.net> To: Jason Evans <jasone@freebsd.org> Cc: svn-src-all@freebsd.org, src-committers@freebsd.org Subject: Re: svn commit: r234370 - in head: contrib/jemalloc contrib/jemalloc/doc contrib/jemalloc/include contrib/jemalloc/include/jemalloc contrib/jemalloc/include/jemalloc/internal contrib/jemalloc/src inclu... Message-ID: <20120419221134.76128edc@cox.net> In-Reply-To: <D08BCC8E-C20C-46BF-984F-C43876A0E7AB@freebsd.org> References: <201204170722.q3H7ME6A070934@svn.freebsd.org> <F43DDEEA-375B-4F1C-80A3-AAECFDFF414A@FreeBSD.org> <AECE03EF-CDEB-43DE-9222-CD94CC07CC95@FreeBSD.org> <CAJ-VmomRqGRbDD3B=NhtKY=E2pb1Lqrr9AcwNGtD91U_GfBi8Q@mail.gmail.com> <20120419091558.2ae655a0@cox.net> <A85EFF6D-6250-4A37-A051-6F852CB2FBF5@freebsd.org> <D08BCC8E-C20C-46BF-984F-C43876A0E7AB@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, 19 Apr 2012 15:23:39 -0700 Jason Evans <jasone@freebsd.org> wrote: > On Apr 19, 2012, at 11:44 AM, Jason Evans wrote: > > On Apr 19, 2012, at 7:15 AM, Conrad J. Sabatier wrote: > >> Actually, I keep getting "undefined reference to _malloc_options" > >> while building libexec/atrun on amd64. Not sure what the problem > >> may be. Could it have something to do with doing a debug build > >> (with -g)? > >> > >> ===> libexec/atrun (all) > >> […] > >> -lutil /usr/obj/usr/src/tmp/usr/lib/libc.so: undefined reference to > >> `_malloc_options' clang: error: linker command failed with exit > >> code 1 (use -v to see invocation) *** [atrun] Error code 1 > > > > Perhaps this is related to using clang rather than gcc. I'll start > > a clang-based buildworld to see if it reproduces here. > > > > Jason > > > With /etc/make.conf as: > > CFLAGS+=-g > STRIP= > CC=clang > CXX=clang++ > CPP=clang-cpp > > I got this buildworld failure: > --- > clang -O2 -pipe -g -I/usr/src/lib/csu/amd64/../common > -I/usr/src/lib/csu/amd64/../../libc/include -fno-omit-frame-pointer > -std=gnu99 -Qunused-arguments -Wsystem-headers -Werror -Wall > -Wno-format-y2k -W -Wno-unused-parameter -Wstrict-prototypes > -Wmissing-prototypes -Wpointer-arith -Wreturn-type -Wcast-qual > -Wwrite-strin gs -Wswitch -Wshadow -Wunused-parameter -Wcast-align > -Wchar-subscripts -Winline -Wnested-externs -Wredundant-decls > -Wold-style-definition -Wno-pointer-sign -Wno-empty-body > -Wno-string-plus-int -c -o crt1.o crt1.scrt1.s:8:2: error: input > can't have .file dwarf directives when -g is used to generate dwarf > debug info for assembly code .file 1 > "/usr/src/lib/csu/amd64/crt1.c" ^ crt1.s:8:8: error: file > number already allocated .file 1 > "/usr/src/lib/csu/amd64/crt1.c" ^crt1.s:9:2: error: input can't > have .file dwarf directives when -g is used to generate dwarf debug > info for assembly code .file 2 > "/usr/src/lib/csu/amd64/../common/crtbrand.c" ^crt1.s:10:2: > error: input can't have .file dwarf directives when -g is used to > generate dwarf debug info for assembly code .file 3 > "/usr/src/lib/csu/amd64/../common/ignore_init.c" ^ --- > > I removed the CFLAGS+=-g line, and buildworld succeeded. > > Jason Yes, I've seen exactly the same on other, earlier builds (I've been using essentially exactly the same make.conf settings as what you just posted, as well as using "make -j8"). That other error output I posted earlier in this thread resulted after I had already seen the above, and tried another buildworld without -j8. So, apparently, the thing to do for now is *not* attempt to build a debug world, it would seem. I suspected as much. :-) Thank you very much for giving your attention to this matter. -- Conrad J. Sabatier conrads@cox.net
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20120419221134.76128edc>