The standard build process works just as normal with a fresh source (September 13th) but fails with make installworld: ========= mkdir -p /tmp/install.H63Fp2Ty progs=$(for prog in [ awk cap_mkdb cat chflags chmod chown cmp cp date echo egrep find grep id install ln lockf make mkdir mtree mv pwd_mkdb rm sed services_mkdb sh strip sysctl test true uname wc zic tzsetup makewhatis; do if progpath=`which $prog`; then echo $progpath; else echo "Required tool $prog not found in PATH." >&2; exit 1; fi; done); libs=$(ldd -f "%o %p\n" -f "%o %p\n" $progs 2>/dev/null | sort -u | while read line; do set -- $line; if [ "$2 $3" != "not found" ]; then echo $2; else echo "Required library $1 not found." >&2; exit 1; fi; done); cp $libs $progs /tmp/install.H63Fp2Ty cp -R ${PATH_LOCALE:-"/usr/share/locale"} /tmp/install.H63Fp2Ty/locale cd /usr/src; MAKEOBJDIRPREFIX=/usr/obj MACHINE_ARCH=amd64 MACHINE=amd64 CPUTYPE= GROFF_BIN_PATH=/usr/obj/usr/src/tmp/legacy/usr/bin GROFF_FONT_PATH=/usr/obj/usr/src/tmp/legacy/usr/share/groff_font GROFF_TMAC_PATH=/usr/obj/usr/src/tmp/legacy/usr/share/tmac PATH=/usr/obj/usr/src/tmp/legacy/usr/sbin:/usr/obj/usr/src/tmp/legacy/usr/bin:/usr/obj/usr/src/tmp/legacy/bin:/usr/obj/usr/src/tmp/usr/sbin:/usr/obj/usr/src/tmp/usr/bin:/tmp/install.H63Fp2Ty LD_LIBRARY_PATH=/tmp/install.H63Fp2Ty PATH_LOCALE=/tmp/install.H63Fp2Ty/locale make -f Makefile.inc1 __MAKE_SHELL=/tmp/install.H63Fp2Ty/sh reinstall; MAKEOBJDIRPREFIX=/usr/obj MACHINE_ARCH=amd64 MACHINE=amd64 CPUTYPE= GROFF_BIN_PATH=/usr/obj/usr/src/tmp/legacy/usr/bin GROFF_FONT_PATH=/usr/obj/usr/src/tmp/legacy/usr/share/groff_font GROFF_TMAC_PATH=/usr/obj/usr/src/tmp/legacy/usr/share/tmac PATH=/usr/obj/usr/src/tmp/legacy/usr/sbin:/usr/obj/usr/src/tmp/legacy/usr/bin:/usr/obj/usr/src/tmp/legacy/bin:/usr/obj/usr/src/tmp/usr/sbin:/usr/obj/usr/src/tmp/usr/bin:/tmp/install.H63Fp2Ty LD_LIBRARY_PATH=/tmp/install.H63Fp2Ty PATH_LOCALE=/tmp/install.H63Fp2Ty/locale rm -rf /tmp/install.H63Fp2Ty make[2]: "/usr/src/share/mk/bsd.compiler.mk" line 48: Unable to determine compiler type for /usr/local/libexec/ccache/world/cc. Consider setting COMPILER_TYPE. *** Error code 1 Stop. make[1]: stopped in /usr/src *** Error code 1 Stop. make: stopped in /usr/src ========= Setting 'make COMPILER_TYPE=clang installworld' allows advancing further, but still fails: ========= install -s -o root -g wheel -m 555 h_memmove /usr/tests/lib/libc/ssp/h_memmove install -o root -g wheel -m 555 ssp_test /usr/tests/lib/libc/ssp/ssp_test (cd /usr/src/lib/libc/tests/ssp && DEPENDFILE=.depend.h_memset make -f /usr/src/lib/libc/tests/ssp/Makefile _RECURSING_PROGS= SUBDIR= PROG=h_memset install) install -s -o root -g wheel -m 555 h_memset /usr/tests/lib/libc/ssp/h_memset install -o root -g wheel -m 555 ssp_test /usr/tests/lib/libc/ssp/ssp_test (cd /usr/src/lib/libc/tests/ssp && DEPENDFILE=.depend.h_raw make -f /usr/src/lib/libc/tests/ssp/Makefile _RECURSING_PROGS= SUBDIR= PROG=h_raw install) install -s -o root -g wheel -m 555 h_raw /usr/tests/lib/libc/ssp/h_raw install: h_raw: No such file or directory *** Error code 71 Stop. make[8]: stopped in /usr/src/lib/libc/tests/ssp *** Error code 1 Stop. make[7]: stopped in /usr/src/lib/libc/tests/ssp *** Error code 1 Stop. make[6]: stopped in /usr/src/lib/libc/tests *** Error code 1 Stop. make[5]: stopped in /usr/src/lib/libc *** Error code 1 Stop. make[4]: stopped in /usr/src/lib *** Error code 1 Stop. make[3]: stopped in /usr/src *** Error code 1 Stop. make[2]: stopped in /usr/src *** Error code 1 Stop. make[1]: stopped in /usr/src *** Error code 1 Stop. make: stopped in /usr/src ========= I have to disable installing tests to make it work: 'make COMPILER_TYPE=clang -DWITHOUT_TESTS installworld' I use ccache, but I have tested without it and the result is the same.
- How did you do `make buildworld`? - Did you update your source tree between when you ran buildworld vs installworld? - How is your source tree/object tree mounted (noatime, atime, etc)?
(In reply to Garrett Cooper,425-314-3911 from comment #1) Exact commands are needed. In particular, if you are calling make installworld without sudo -E and are using env/export to export variables, that won't work necessarily as expected (but you would have run into issues later on with mismatched sources/objects ..)
Finally, this *should* not happen, but I'm wondering if somehow the build vs installed version of ${CC} is affecting what's being built: 29 # This testcase doesn't run properly when not compiled with -fsantize=bounds 30 # with clang, which is currently contingent on a compiler_rt update 31 .if ${COMPILER_TYPE} == "clang" && ${COMPILER_VERSION} < 30500 32 PROGS+= h_raw 33 .endif This would make sense and it would be a huge bug with whatever we do as far as COMPILER_VERSION gating logic is concerned (it affects a few handfuls of areas of the tree).
(In reply to Garrett Cooper,425-314-3911 from comment #1) First, I just did 'make -j2 buildworld' and it stopped on 'make installworld' as reported. When I saw that setting COMPILER_TYPE allowed the install to proceed further (till the h_raw error), I tried starting again with 'make COMPILER_TYPE=clang -j2 buildworld', which worked, but installworld failed again on the h_raw error. I have not tried disabling the tests at the buildworld stage.
(In reply to Garrett Cooper,425-314-3911 from comment #1) - Did you update your source tree between when you ran buildworld vs installworld? No. - How is your source tree/object tree mounted (noatime, atime, etc)? This is a plain vanilla install HEAD snapshot install downloaded Saturday: # cat /etc/fstab # Device Mountpoint FStype Options Dump Pass# /dev/ada0s1a / ufs rw 1 1 /dev/ada0s1b none swap sw 0 0
(In reply to Juan Ramón Molina Menor from comment #4) What version are you upgrading from/to?
(In reply to Garrett Cooper,425-314-3911 from comment #6) This is a fresh install of one of the latest snapshots: FreeBSD-11.0-CURRENT-amd64-20150903-r287437-mini-memstick.img.xz Then I download source: /usr/src # svnlite info Path: . Working Copy Root Path: /usr/src URL: https://svn.freebsd.org/base/head Relative URL: ^/head Repository Root: https://svn.freebsd.org/base Repository UUID: ccf9f872-aa2e-dd11-9fc8-001c23d0bc1f Revision: 287761 Node Kind: directory Schedule: normal Last Changed Author: jilles Last Changed Rev: 287761 Last Changed Date: 2015-09-13 19:17:52 +0200 (Sun, 13 Sep 2015) Then I followed the upgrade procedure stated in /usr/src/UPDATING: make -j2 buildworld make kernel <reboot in single user> mergemaster -Fp make installworld …and it failed as described. Only two things depart from a plain install: I installed devel/ccache and NFS-exported /usr/src and /usr/obj, as the aim is to upgrade a USB stick I’m using to test head on my ThinkPad S440. But I have tried the above procedure without ccache and without NFS exports (commented-out ccache in /etc/make.conf and NFS lines in /etc/rc.conf and /etc/exports, then rebooted computer) and the failure is still the same. The machine is on and I can make as many tests as you wish.
Still not enough info. What version are you upgrading the base system from? The version that's on the USB/CD isn't important, unless it's in sync with the version that you're installing to on the base system. Please note: building/installing on the base system requires some degree of finesse. Try using a kernel greater than the install version, chrooting into the install tree, then running make installworld (yes, it would be a doc bug, but as a rule of thumb I chroot into directories and run installworld because otherwise.. it's running binaries off the install media).
(In reply to Garrett Cooper,425-314-3911 from comment #8) > Still not enough info. What version are you upgrading the base system from? > The version that's on the USB/CD isn't important, unless it's in sync with > the version that you're installing to on the base system. Sorry, I must be obtuse, but I don’t understand your question: This is a just a FreeBSD-11.0-CURRENT-amd64-20150903-r287437 mini-memstick image booted-up and installed on a wiped-out hard disk. Then I downloaded sources from the SVN repository (2015-09-13 r287761, ten days older) with svnlite, then tried to upgrade.
(In reply to Juan Ramón Molina Menor from comment #9) Just to be clear -- did you build/install from the rebooted system or the USB live image?
(In reply to Garrett Cooper,425-314-3911 from comment #10) I did build/install from the rebooted system, the USB (mini-memstick, whithout sources) was not even connected.
This is still happening with the system updated to base r288424. Process as usual: # cd /usr/src # svnlite up …updates from r287761 to r288424… # make -j2 buildworld # make -j2 kernel # shutdown now # make -j2 installworld …fails as previously reported… # make COMPILER_TYPE=clang -j2 -DWITHOUT_TESTS installworld …works… It’s really curious I am the only one seeing this.
I'm able to confirm and reproduce (a similar/this) problem during the build-/installworld process in combination with ccache. 1. TEST-ENV - ThinkPad X220 (4290-DD1) - Fresh installed FreeBSD 11.0-CURRENT amd64 r296485 2. WORKING REFERENCE - check out freebsd-current (r297375) # svnlite co -r 297375 https://svn.freebsd.org/base/head/ /usr/src - modify /etc/make.conf # vi /etc/make.conf <-- MALLOC_PRODUCTION=YES --> - start build and install process # make -j16 buildworld # make -j16 buildkernel KERNCONF=GENERIC-NODEBUG <reboot in s.u. mode> # mergemaster -Fp # make installworld # mergemaster -Fi # make delete-old <reboot> - Everything works as usual 2. NON-WORKING EXAMPLE - install ccache version 3.2.4 # pkg install ccache - modify /etc/make.conf # vi /etc/make.conf <-- MALLOC_PRODUCTION=YES WITH_CCACHE_BUILD=YES .if (!empty(.CURDIR:M/usr/src*) || !empty(.CURDIR:M/usr/obj*)) .if !defined(NOCCACHE) && exists(/usr/local/libexec/ccache/world/cc) CC:=${CC:C,^cc,/usr/local/libexec/ccache/world/cc,1} CXX:=${CXX:C,^c\+\+,/usr/local/libexec/ccache/world/c++,1} .endif .endif --> - modify /etc/csh.cshrc # vi /etc/csh.cshrc <-- setenv PATH /usr/local/libexec/ccache:$PATH setenv CCACHE_PATH /usr/bin:/usr/local/bin --> - check out freebsd-current (r297375) # svnlite co -r 297375 https://svn.freebsd.org/base/head/ /usr/src - start build and install process # make -j16 buildworld # make -j16 buildkernel KERNCONF=GENERIC-NODEBUG <reboot in s.u. mode> # mergemaster -Fp # make installworld - the problem occurs <-- mkdir -p /tmp/install.KI6wjBV6 progs=$(for prog in [ awk cap_mkdb cat chflags chmod chown cmp cp date echo egrep find grep id install ln make mkdir mtree mv pwd_mkdb rm sed services_mkdb sh strip sysctl test true uname wc zic tzsetup makewhatis; do if progpath=`which $prog`; then echo $progpath; else echo "Required tool $prog not found in PATH." >&2; exit 1; fi; done); libs=$(ldd -f "%o %p\n" -f "%o %p\n" $progs 2>/dev/null | sort -u | while read line; do set -- $line; if [ "$2 $3" != "not found" ]; then echo $2; else echo "Required library $1 not found." >&2; exit 1; fi; done); cp $libs $progs /tmp/install.KI6wjBV6 cp -R ${PATH_LOCALE:-"/usr/share/locale"} /tmp/install.KI6wjBV6/locale cd /usr/src; MAKEOBJDIRPREFIX=/usr/obj MACHINE_ARCH=amd64 MACHINE=amd64 CPUTYPE= GROFF_BIN_PATH=/usr/obj/usr/src/tmp/legacy/usr/bin GROFF_FONT_PATH=/usr/obj/usr/src/tmp/legacy/usr/share/groff_font GROFF_TMAC_PATH=/usr/obj/usr/src/tmp/legacy/usr/share/tmac CC="/usr/local/libexec/ccache/world/cc -target x86_64-unknown-freebsd11.0 --sysroot=/usr/obj/usr/src/tmp -B/usr/obj/usr/src/tmp/usr/bin" CXX="/usr/local/libexec/ccache/world/c++ -target x86_64-unknown-freebsd11.0 --sysroot=/usr/obj/usr/src/tmp -B/usr/obj/usr/src/tmp/usr/bin " DEPFLAGS="" CPP="cpp -target x86_64-unknown-freebsd11.0 --sysroot=/usr/obj/usr/src/tmp -B/usr/obj/usr/src/tmp/usr/bin" AS="as" AR="ar" LD="ld" NM=nm OBJDUMP=objdump OBJCOPY="objcopy" RANLIB=ranlib STRINGS= SIZE="size" PATH=/usr/obj/usr/src/tmp/legacy/usr/sbin:/usr/obj/usr/src/tmp/legacy/usr/bin:/usr/obj/usr/src/tmp/legacy/bin:/usr/obj/usr/src/tmp/usr/sbin:/usr/obj/usr/src/tmp/usr/bin:/tmp/install.KI6wjBV6 LD_LIBRARY_PATH=/tmp/install.KI6wjBV6 PATH_LOCALE=/tmp/install.KI6wjBV6/locale /usr/obj/usr/src/make.amd64/bmake -f Makefile.inc1 __MAKE_SHELL=/tmp/install.KI6wjBV6/sh reinstall; MAKEOBJDIRPREFIX=/usr/obj MACHINE_ARCH=amd64 MACHINE=amd64 CPUTYPE= GROFF_BIN_PATH=/usr/obj/usr/src/tmp/legacy/usr/bin GROFF_FONT_PATH=/usr/obj/usr/src/tmp/legacy/usr/share/groff_font GROFF_TMAC_PATH=/usr/obj/usr/src/tmp/legacy/usr/share/tmac CC="/usr/local/libexec/ccache/world/cc -target x86_64-unknown-freebsd11.0 --sysroot=/usr/obj/usr/src/tmp -B/usr/obj/usr/src/tmp/usr/bin" CXX="/usr/local/libexec/ccache/world/c++ -target x86_64-unknown-freebsd11.0 --sysroot=/usr/obj/usr/src/tmp -B/usr/obj/usr/src/tmp/usr/bin " DEPFLAGS="" CPP="cpp -target x86_64-unknown-freebsd11.0 --sysroot=/usr/obj/usr/src/tmp -B/usr/obj/usr/src/tmp/usr/bin" AS="as" AR="ar" LD="ld" NM=nm OBJDUMP=objdump OBJCOPY="objcopy" RANLIB=ranlib STRINGS= SIZE="size" PATH=/usr/obj/usr/src/tmp/legacy/usr/sbin:/usr/obj/usr/src/tmp/legacy/usr/bin:/usr/obj/usr/src/tmp/legacy/bin:/usr/obj/usr/src/tmp/usr/sbin:/usr/obj/usr/src/tmp/usr/bin:/tmp/install.KI6wjBV6 LD_LIBRARY_PATH=/tmp/install.KI6wjBV6 PATH_LOCALE=/tmp/install.KI6wjBV6/locale rm -rf /tmp/install.KI6wjBV6 ccache: error: Could not find compiler "cc" in PATH bmake[2]: "/usr/src/share/mk/bsd.compiler.mk" line 141: Unable to determine compiler type for /usr/local/libexec/ccache/world/cc -target x86_64-unknown-freebsd11.0 --sysroot=/usr/obj/usr/src/tmp -B/usr/obj/usr/src/tmp/usr/bin. Consider setting COMPILER_TYPE. *** Error code 1 Stop. bmake[1]: stopped in /usr/src *** Error code 1 Stop. make: stopped in /usr/src --> - NOTE: to proceed at this step it is necessary to set "COMPILER_TYPE=clang -DWITHOUT_TESTS" as Juan wrote before. # make COMPILER_TYPE=clang -DWITHOUT_TESTS installworld ...
During my research I've found SVN:290526 (https://svnweb.freebsd.org/base?view=revision&revision=290526). Bryan added ccache build support via WITH_CCACHE_BUILD option to "base". This removes the need to add the following lines (as suggested in /usr/local/share/doc/ccache/ccache-howto-freebsd.txt) to /etc/make.conf: <-- .if (!empty(.CURDIR:M/usr/src*) || !empty(.CURDIR:M/usr/obj*)) .if !defined(NOCCACHE) && exists(/usr/local/libexec/ccache/world/cc) CC:=${CC:C,^cc,/usr/local/libexec/ccache/world/cc,1} CXX:=${CXX:C,^c\+\+,/usr/local/libexec/ccache/world/c++,1} .endif .endif --> Remove them and just add WITH_CCACHE_BUILD to /etc/make.conf: <-- WITH_CCACHE_BUILD=YES --> Regarding to SVN:290526 Bryans work should have been fixed all known issues with buildworld and ccache. After modifying my /etc/make.conf I was able to rebuild world and install it as usual.
Something is still broken. Here is my situation. Base system is 10-STABLE dated Sep 26 2016 New src code is RELENG 11.0 P9 ccache is installed on the system I first couldnt even get buildworld to complete, very odd error rejecting cputype of nehalem, this error appeared even with no cputype set in make.conf. I proceeded to uncomment the CC= lines that set as below .if (!empty(.CURDIR:M/usr/src*) || !empty(.CURDIR:M/usr/obj*)) .if !defined(NOCCACHE) && exists(/usr/local/libexec/ccache/world/cc) CC:=${CC:C,^cc,/usr/local/libexec/ccache/world/cc,1} CXX:=${CXX:C,^c\+\+,/usr/local/libexec/ccache/world/c++,1} .endif .endif After I did this the buildworld proceeded but very slow, I discovered ccache was operating, as the misses were tallying up and I had disk thrashing from the cache objects been written. So I then commented out WITH_CCACHE_BUILD=YES in make.conf, the line is not present in src.conf at all. This time buildworld completed and at normal speed. But now I get the error on make installworld. So the question is if none of the cache lines are active in either make.conf or src.conf and one gets this error. Is it a bug or operator error? bmake[2]: "/usr/src/110/share/mk/bsd.compiler.mk" line 132: warning: "/usr/local/bin/ccache cc --version || echo 0.0.0" returned non-zero status bmake[2]: "/usr/src/110/share/mk/bsd.compiler.mk" line 145: Unable to determine compiler type for CC=/usr/local/bin/ccache cc . Consider setting COMPILER_TYPE. So here is 2 observations. 1 - The compiler used during buildworld was not "/usr/local/bin/ccache cc"it was just "cc" 2 - Why is it showing "/usr/local/bin/ccache cc" when I have nothing enabling ccache active in either src.conf or make.conf, is simply having the ccache port installed enough to make the world tools try to use it?
Here is error bmake[2]: "/usr/src/110/share/mk/bsd.compiler.mk" line 132: warning: "cc --version || echo 0.0.0" returned non-zero status bmake[2]: "/usr/src/110/share/mk/bsd.compiler.mk" line 145: Unable to determine compiler type for CC=cc . Consider setting COMPILER_TYPE. *** Error code 1 The above is when I tried to add to src.conf following the advise in earlier reply.
So I added CC=clang to src.conf Now more weirdness bmake[2]: "/usr/src/110/share/mk/bsd.compiler.mk" line 132: warning: "clang --version || echo 0.0.0" returned non-zero status bmake[2]: "/usr/src/110/share/mk/bsd.compiler.mk" line 149: warning: "echo "" | awk -F. '{print $1 * 10000 + $2 * 100 + $3;}'" returned non-zero status bmake[2]: "/usr/src/110/share/mk/bsd.compiler.mk" line 154: warning: "{ echo "__FreeBSD_cc_version" | clang -E - 2>/dev/null || echo __FreeBSD_cc_version; } | sed -n '$p'" returned non-zero status bmake[2]: "/usr/src/110/Makefile.inc1" line 385: warning: "MK_AUTO_OBJ=no MAKEFLAGS= CPUTYPE=nehalem /usr/obj/usr/src/110/make.amd64/bmake -f /dev/null -m /usr/src/110/share/mk -V CPUTYPE" returned non-zero status bmake[2]: "/usr/src/110/Makefile.inc1" line 387: CPUTYPE global should be set with ?=. *** Error code 1 The weirdness is the fact CPUTYPE?= is actually set.
I am surprised to see in June 2020 : perfectly fine # make -j4 kernel . . . . kldxref /boot/kernel -------------------------------------------------------------- >>> Installing kernel GENERIC completed on Wed Jun 17 03:47:13 UTC 2020 -------------------------------------------------------------- You have new mail. root@inferno:/usr/src # make installworld make[1]: "/usr/src/Makefile.inc1" line 100: A build is required first. You may have the wrong MAKEOBJDIRPREFIX set. *** Error code 1 Stop. make: stopped in /usr/src root@inferno:/usr/src # OKay so now what ?
(In reply to Dennis Clarke from comment #18) `world` target doesn't include kernel, it's only userland bits. Before `installworld`, you may want `buildworld` first. Please refer to https://www.freebsd.org/doc/handbook/makeworld.html
I've observed the same error on h_raw attempting to build -CURRENT pkgbase packages, on a FreeBSD 13.0-CURRENT #1 r363183+3db797cc03ab-c269925(laptop): Tue Jul 14 12:45:00 EDT 2020 host
(In reply to Ed Maste from comment #20) My issue was a local problem, caused by an incomplete buildworld.
I'm experience the same (?) problem with 14-stable. #uname -a FreeBSD 14.1-STABLE amd64 src#git log -1 commit ae979c76443bd7c33ed9f22182b4421d2c64b619 (HEAD -> stable/14, origin/stable/14) Date: Thu Jun 20 12:27:43 2024 +0800 ... #cat /etc/make.conf CPUTYPE=nehalem MACHINE_ARCH=amd64 #cat /etc/src.conf #--Compiler/Linker-- WITHOUT_CROSS_COMPILER="YES" # == WITHOUT_TOOLCHAIN + WITH_LLDB WITHOUT_CLANG="YES" WITHOUT_CLANG_EXTRAS="YES" WITHOUT_CLANG_FULL="YES" WITHOUT_LLD="YES" WITHOUT_LLVM_COV="YES" ... === After successfull 'buildworld' 'distributeworld' fails with error: install -N /tank/bsdsrc/14s/src14s/etc -s -o root -g wheel -m 555 h_memset /tank/bsdsrc/14s/distworld/tests/usr/tests/lib/libc/ssp/h_memset install -N /tank/bsdsrc/14s/src14s/etc -o root -g wheel -m 444 h_memset.debug /tank/bsdsrc/14s/distworld/tests/usr/lib/debug/usr/tests/lib/libc/ssp/h_memset.debug (cd /tank/bsdsrc/14s/src14s/lib/libc/tests/ssp && DEPENDFILE=.depend.h_raw NO_SUBDIR=1 make -f /tank/bsdsrc/14s/src14s/lib/libc/tests/ssp/Makefile _RECURSING_PROGS=t PROG=h_raw install) install -N /tank/bsdsrc/14s/src14s/etc -s -o root -g wheel -m 555 h_raw /tank/bsdsrc/14s/distworld/tests/usr/tests/lib/libc/ssp/h_raw install: h_raw: No such file or directory *** Error code 71 Stop. make[9]: stopped in /tank/bsdsrc/14s/src14s/lib/libc/tests/ssp *** Error code 1 Stop. make[8]: stopped in /tank/bsdsrc/14s/src14s/lib/libc/tests/ssp *** Error code 1 Stop. make[7]: stopped in /tank/bsdsrc/14s/src14s/lib/libc/tests/ssp *** Error code 1 Stop. make[6]: stopped in /tank/bsdsrc/14s/src14s/lib/libc/tests *** Error code 1 Stop. make[5]: stopped in /tank/bsdsrc/14s/src14s/lib/libc *** Error code 1 Stop. make[4]: stopped in /tank/bsdsrc/14s/src14s/lib *** Error code 1 Stop. make[3]: stopped in /tank/bsdsrc/14s/src14s *** Error code 1 Stop. make[2]: stopped in /tank/bsdsrc/14s/src14s 94,39 real 53,77 user 34,58 sys *** Error code 1 Stop. make[1]: stopped in /tank/bsdsrc/14s/src14s *** Error code 1 Stop. make: stopped in /tank/bsdsrc/14s/src14s ==== It seems that h_raw conditionally included(excluded) in a build process but this condition differs in 'distributeworld` target: ./lib/libc/tests/ssp/Makefile: ... # XXX: the h_raw/h_read testcases don't cause a SIGABRT with in-tree gcc right # now on amd64 when it trips the stack bounds specified in t_ssp.sh . This # probably needs to be fixed as it's currently hardcoded. .if ${COMPILER_TYPE} == "clang" && !defined(_SKIP_BUILD) && \ (!defined(_RECURSING_PROGS) || ${PROG} == "h_raw") .include "${SRCTOP}/lib/libclang_rt/compiler-rt-vars.mk" _libclang_rt_ubsan= ${SYSROOT}${SANITIZER_LIBDIR}/libclang_rt.ubsan_standalone-${C .if exists(${_libclang_rt_ubsan}) PROGS+= h_raw LDADD.h_raw+= ${SANITIZER_LDFLAGS} .else .if make(all) .info Could not find runtime library ${_libclang_rt_ubsan}, skipping h_raw .endif .endif .endif ...