Re: G_RPI4 Buildkernel stops silently on RPi4
- In reply to: bob prohaska : "G_RPI4 Buildkernel stops silently on RPi4"
- Go to: [ bottom of page ] [ top of archives ] [ this month ]
Date: Thu, 17 Jun 2021 19:58:12 UTC
On 2021-Jun-17, at 11:06, bob prohaska <fbsd at www.zefox.net> wrote: > I'm not sure G_RPI4 is still relevant to anybody, but a silent > stoppage in buildkernel struck me as surprising. A retry using > GENERIC-MMCCAM worked without trouble and is running now. > > A recent attempt to update world and kernel G_RPI4 on a Pi4 running > 14-current stopped in buildworld, with no obvious error: > > root@nemesis:/usr/src # tail b*.log > ==> buildkernel.log <== > MAKEOBJDIRPREFIX='' > MAKESYSPATH='/usr/src/share/mk' > MAKE_VERSION='20210206' > PATH='/sbin:/bin:/usr/sbin:/usr/bin' > SRCTOP='/usr/src' > OBJTOP='/usr/obj/usr/src/arm64.aarch64' > .MAKE.MAKEFILES='/usr/src/share/mk/sys.mk /usr/src/share/mk/local.sys.env.mk /usr/src/share/mk/src.sys.env.mk /usr/src/share/mk/bsd.mkopt.mk /usr/src/share/mk/src.sys.obj.mk /usr/src/share/mk/auto.obj.mk /usr/src/share/mk/bsd.suffixes.mk /usr/src/share/mk/local.sys.mk /usr/src/share/mk/src.sys.mk /usr/src/Makefile.inc1 /usr/src/share/mk/src.tools.mk /usr/src/share/mk/bsd.compiler.mk /usr/src/share/mk/bsd.opts.mk /usr/src/share/mk/bsd.cpu.mk /usr/src/share/mk/src.opts.mk /usr/src/share/mk/bsd.own.mk /usr/src/share/mk/bsd.linker.mk' > .PATH='. /usr/src' > > make: stopped in /usr/src > > > There was nothing on the serial console, nor the controlling terminal. > The sources were updated a few minutes earlier, reporting > > root@nemesis:/usr/src # git pull > remote: Enumerating objects: 100, done. > remote: Counting objects: 100% (100/100), done. > remote: Compressing objects: 100% (48/48), done. > remote: Total 59 (delta 51), reused 18 (delta 10), pack-reused 0 > Unpacking objects: 100% (59/59), 7.13 KiB | 17.00 KiB/s, done. > From https://git.freebsd.org/src > 8fa5c577de..0247c33e89 main -> origin/main > Updating 8fa5c577de..0247c33e89 > Fast-forward > sys/amd64/amd64/efirt_machdep.c | 9 ++++ > sys/kern/subr_param.c | 6 +-- > sys/mips/conf/AR71XX_BASE | 1 - > sys/mips/conf/PB92 | 1 - > sys/mips/conf/QCA953X_BASE | 1 - > sys/mips/conf/std.AR5312 | 1 - > sys/mips/conf/std.AR5315 | 1 - > sys/mips/conf/std.AR724X | 1 - > sys/mips/conf/std.AR91XX | 1 - > sys/mips/conf/std.AR933X | 1 - > sys/mips/conf/std.AR934X | 1 - > sys/mips/conf/std.MALTA | 1 + > sys/mips/conf/std.QCA955X | 1 - > sys/mips/conf/std.XLP | 1 - > sys/tests/fib_lookup/fib_lookup.c | 140 +++++++++++++++++++++++++++++++++++++++++++++-------- > sys/vm/vm_page.c | 2 +- > sys/vm/vm_page.h | 1 + > usr.sbin/bsdconfig/include/messages.subr | 1 + > usr.sbin/bsdconfig/share/media/ftp.subr | 1 + > usr.sbin/bsdinstall/scripts/mirrorselect | 1 + > usr.sbin/bsnmpd/modules/snmp_pf/pf_snmp.c | 4 +- > 21 files changed, 137 insertions(+), 40 deletions(-) > > The build command was > make -j4 -DWITH_META_MODE buildworld > buildworld.log && make -j4 buildkernel -DWITH_META_MODE KERNCONF=G_RPI4 > buildkernel.log > which has worked well for some months. # ls -Tld /usr/fbsd/mm-src/sys/arm64/conf/* -rw-r--r-- 1 root wheel 293 Jan 26 13:11:51 2021 /usr/fbsd/mm-src/sys/arm64/conf/DEFAULTS -rw-r--r-- 1 root wheel 12838 Mar 3 15:49:42 2021 /usr/fbsd/mm-src/sys/arm64/conf/GENERIC -rw-r--r-- 1 root wheel 429 Jan 26 13:11:51 2021 /usr/fbsd/mm-src/sys/arm64/conf/GENERIC-MMCCAM -rw-r--r-- 1 root wheel 323 Jan 26 13:11:51 2021 /usr/fbsd/mm-src/sys/arm64/conf/GENERIC-MMCCAM-NODEBUG -rw-r--r-- 1 root wheel 1253 Jan 26 13:11:51 2021 /usr/fbsd/mm-src/sys/arm64/conf/GENERIC-NODEBUG -rw-r--r-- 1 root wheel 840 Jan 26 13:11:51 2021 /usr/fbsd/mm-src/sys/arm64/conf/GENERIC-UP -rw-r--r-- 1 root wheel 54 Jan 26 13:11:51 2021 /usr/fbsd/mm-src/sys/arm64/conf/LINT -rw-r--r-- 1 root wheel 6389 Jan 26 13:11:51 2021 /usr/fbsd/mm-src/sys/arm64/conf/NOTES But that is on system that has not been updated in some time: merge-base: 7381bbee29df959e88ec59866cf2878263e7f3b2 merge-base: CommitDate: 2021-03-12 20:29:42 +0000 7381bbee29df (HEAD -> main, pure-src) cam: Run all XPT_ASYNC ccbs in a dedicated thread n245444 (--first-parent --count for merge-base) G_RPI4 seems to not be from FreeBSD's source tree. You may have to give more context or show the contents of the file. > Just noticed the KERNCONF=G_RPI4, which I tried as an experiment that > seemed to work. That reads like you forgot the KERNCONF=G_RPI4 the first time and the build worked when it was retried with it supplied? (Ultimately: I'm confused.) > The original motive was a smaller kernel, but that no > longer seems to be the case; GENERIC-MMCCAM is ~30 kB smaller 8-) Was this a comparison with an older G_RPI4 build? A new rebuild that worked? === Mark Millard marklmi at yahoo.com ( dsl-only.net went away in early 2018-Mar)