NetBSD-Bugs archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: port-zaurus/56238: Several ARM-Variants seem to miss gcov-iov.h

On Wed, 2021-06-09 11:35:19 +0200, Martin Husemann <> wrote:
> On Wed, Jun 09, 2021 at 10:47:37AM +0200, Jan-Benedict Glaw wrote:
> >   Building tools worked, but building the release broke as it broke
> > when building within the Linux (Debian) docker containers due to not
> > finding gcov-iov.h.  So I guess something broke in the toolchain?
> > 
> >   Logs can be found at:
> > 
> >
> >
> Well, it works for the official builds (which are done from cvs source
> trees), so it must be something git specific.
> Maybe a timestamp getting set differently, unveiling an unnoticed makefile
> bug?

Whee!  So I've not got a quite pristinely installed NetBSD 9.2
(installed via boot-com.iso), just a SSH key for root added. From the
Laminar CI job scheduler, the following script is called.
JOB=nnetbsd-evbarm-earmv5hfeb, so BUILD_MACHINE=evbarm and

#!/usr/bin/env bash

set -ex


BUILD_MACHINE="$(echo "${JOB}" | cut -f 2 -d -)"
BUILD_ARCH="$(   echo "${JOB}" | cut -f 3 -d -)"

function Scp() {
	scp -i "${laminar_jb_tool_dir}/netbsd-ssh-key" -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null "${@}"

function Ssh() {
	ssh -i "${laminar_jb_tool_dir}/netbsd-ssh-key" -o StrictHostKeyChecking=no -o UserKnownHostsFile=/dev/null "${@}"

	cd "${laminar_jb_tool_dir}"

# Copy CVS Tree.
( cd "${netbsd_src_tree}" && tar cf - .) | Ssh root@netbsd2 "cd /root && mkdir NetBSD-src-in-cvs && cd NetBSD-src-in-cvs && tar xf -"

set +e
	export PKG_PATH="$(uname -s)/$(uname -m)/$(uname -r)/All"
	export PATH="/sbin:/usr/sbin:/bin:/usr/bin:/usr/pkg/sbin:/usr/pkg/bin:/usr/X11R7/bin:/usr/local/sbin:/usr/local/bin:$PATH"
	pkg_add moreutils
	pkg_add bash
	cd /root/NetBSD-src-in-cvs
	ls -l
	cat > log_execute << 'LOG_EXECUTE'
#!/usr/bin/env bash

	NAME="$( echo "${1}" | sed -e 's/%/&%/g')"; shift
	TS_FORMAT="[${NAME} %Y-%m-%d %H:%M:%S]"

	"${@}" 2>&1 | ts "${TS_FORMAT}"
	exit "${PIPESTATUS[0]}"
	chmod +x log_execute
	./log_execute 'bld tls' ./ -P -U    -m "${BUILD_MACHINE}" -a "${BUILD_ARCH}" -E -D "/root/DEST-${BUILD_MACHINE}-${BUILD_ARCH}" -R "/root/REL-${BUILD_MACHINE}-${BUILD_ARCH}" tools     && \
	./log_execute 'bld rls' ./ -P -U -u -m "${BUILD_MACHINE}" -a "${BUILD_ARCH}" -E -D "/root/DEST-${BUILD_MACHINE}-${BUILD_ARCH}" -R "/root/REL-${BUILD_MACHINE}-${BUILD_ARCH}" release   && \
	./log_execute 'bld iso' ./ -P -U -u -m "${BUILD_MACHINE}" -a "${BUILD_ARCH}" -E -D "/root/DEST-${BUILD_MACHINE}-${BUILD_ARCH}" -R "/root/REL-${BUILD_MACHINE}-${BUILD_ARCH}" iso-image
	poweroff < /dev/null > /dev/null 2>&1 &
	exit $RET

So basically, the NetBSD VM is started (with a snapshot overlay to
keep it clean), a freshly checked-out CVS tree is tar'ed over and
built. (I only scheduled one variant up to now, this can be found at, the
log file is .

  That's as fresh as it gets, and fails:

[bld rls 2021-06-15 11:27:06] /root/NetBSD-src-in-cvs/obj/tooldir.NetBSD-9.2-amd64/bin/armeb--netbsdelf-eabihf-c++ -O2 -fdebug-prefix-map=\$DESTDIR= -fdebug-prefix-map=\$NETBSDSRCDIR=/usr/src -fdebug-prefix-map=\$X11SRCDIR=/usr/xsrc -fdebug-regex-map='/usr/src/(.*)/obj$=/usr/obj/\1' -fdebug-regex-map='/usr/src/(.*)/obj/(.*)=/usr/obj/\1/\2' -fdebug-regex-map='/usr/src/(.*)/obj\..*=/usr/obj/\1' -fdebug-regex-map='/usr/src/(.*)/obj\..*/(.*)=/usr/obj/\1/\2' -Wall -Wpointer-arith -Wno-sign-compare -Wa,--fatal-warnings -Werror   -fPIE -Wno-narrowing -Wno-unused -std=gnu++98 -fdebug-prefix-map=\$DESTDIR= -fdebug-prefix-map=\$NETBSDSRCDIR=/usr/src -fdebug-prefix-map=\$X11SRCDIR=/usr/xsrc -fdebug-regex-map='/usr/src/(.*)/obj$=/usr/obj/\1' -fdebug-regex-map='/usr/src/(.*)/obj/(.*)=/usr/obj/\1/\2' -fdebug-regex-map='/usr/src/(.*)/obj\..*=/usr/obj/\1' -fdebug-regex-map='/usr/src/(.*)/obj\..*/(.*)=/usr/obj/\1/\2' -Wno-stack-protector -fno-exceptions -fno-rtti -fasynchronous-unwind-tables   -I. -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/usr.bin/backend/../gcc/arch/earmhfeb -DIN_GCC -DHAVE_CONFIG_H -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc/. -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc/../include -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc/../libcpp/include -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc/../libdecnumber -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc/../libdecnumber/dpd -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc/../libbacktrace -DTARGET_NAME=\"armeb--netbsdelf-eabihf\" -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/libgcc -I/root/NetBSD-src-in-cvs/external/gpl3/gcc.old/usr.bin/backend/../../lib/libgcc/libgcov/arch/earmv5hfeb --sysroot=/root/DEST-evbarm-earmv5hfeb -DLOCALEDIR=\"/usr/share/locale\" -DNETBSD_NATIVE -I. -DENABLE_SHARED_LIBGCC -Wp,-iremap,/root/DEST-evbarm-earmv5hfeb: -Wp,-fno-canonical-system-headers -Wp,-iremap,/root/NetBSD-src-in-cvs:/usr/src -Wp,-iremap,/usr/xsrc:/usr/xsrc  -c    /root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc/auto-profile.c -o auto-profile.o
[bld rls 2021-06-15 11:27:06] In file included from /root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc/auto-profile.c:34:
[bld rls 2021-06-15 11:27:06] /root/NetBSD-src-in-cvs/external/gpl3/gcc.old/dist/gcc/gcov-io.h:215:10: fatal error: gcov-iov.h: No such file or directory
[bld rls 2021-06-15 11:27:06]   215 | #include "gcov-iov.h"
[bld rls 2021-06-15 11:27:06]       |          ^~~~~~~~~~~~
[bld rls 2021-06-15 11:27:06] compilation terminated.

I really think there's some leftover from a former successful run that
enables new builds...

PS: I'll attach the expect script that I used/created to install this
NetBSD instance.

Description: Bourne shell script

Attachment: signature.asc
Description: PGP signature

Home | Main Index | Thread Index | Old Index