Debian Bug report logs -
#525144
libfakeroot segfaults when building kernels in parallel
Reply or subscribe to this bug.
Toggle useless messages
Report forwarded
to debian-bugs-dist@lists.debian.org, Manoj Srivastava <srivasta@debian.org>:
Bug#525144; Package kernel-package.
(Wed, 22 Apr 2009 14:21:05 GMT) (full text, mbox, link).
Acknowledgement sent
to Christoph Anton Mitterer <christoph.anton.mitterer@physik.uni-muenchen.de>:
New Bug report received and forwarded. Copy sent to Manoj Srivastava <srivasta@debian.org>.
(Wed, 22 Apr 2009 14:21:05 GMT) (full text, mbox, link).
Message #5 received at submit@bugs.debian.org (full text, mbox, reply):
Package: kernel-package
Version: 12.009
Justification: renders package unusable
Severity: grave
Hi.
make-kpkg --revision 0.1 --append-to-version '-heisenberg' --initrd
buildpackage fails with:
...
...
LD [M] sound/drivers/snd-serial-u16550.ko
LD [M] sound/drivers/snd-virmidi.ko
LD [M] sound/usb/snd-usb-lib.ko
LD [M] sound/usb/snd-usb-audio.ko
make[3]: Leaving directory `/root/test/linux-source-2.6.29'
COLUMNS=150 dpkg -l 'gcc*' perl dpkg 'libc6*' binutils make dpkg-dev |\
awk '$1 ~ /[hi]i/ { printf("%s-%s\n", $2, $3) }'> debian/buildinfo
uname -a >> debian/buildinfo
echo using the compiler: >> debian/buildinfo
grep LINUX_COMPILER include/linux/compile.h | \
sed -e 's/.*LINUX_COMPILER "//' -e 's/"$//' >> debian/buildinfo
echo kernel source package used: >> debian/buildinfo
echo linux-source-2.6.29 3 >> debian/buildinfo
echo done > debian/stamp/build/kernel
====== making target debian/stamp/BUILD/linux-image-2.6.29-heisenberg
[new prereqs: do-pre-build-arch kernel]======
====== making target debian/stamp/dep-build-arch [new prereqs:
do-pre-build-arch linux-headers-2.6.29-heisenberg
linux-image-2.6.29-heisenberg linux-image-2.6.29-heisenberg-dbg
linux-uml-2.6.29-heisenberg linux-xen0-2.6.29-heisenberg
linux-xenu-2.6.29-heisenberg]======
make[2]: Leaving directory `/root/test/linux-source-2.6.29'
====== making target build [new prereqs: do-build-arch do-build-indep]======
make[1]: Leaving directory `/root/test/linux-source-2.6.29'
debian/rules binary
make[1]: Entering directory `/root/test/linux-source-2.6.29'
====== making target debian/stamp/do-post-build-arch [new prereqs:
do-build-arch]======
====== making target debian/stamp/do-post-build-indep [new prereqs:
do-build-indep]======
/usr/bin/make -f debian/rules debian/stamp/post-build-arch
/usr/bin/make -f debian/rules debian/stamp/post-build-indep
make[2]: Entering directory `/root/test/linux-source-2.6.29'
make[2]: warning: -jN forced in submake: disabling jobserver mode.
====== making target debian/stamp/post-build-arch [new prereqs:
do-build-arch]======
make[2]: Leaving directory `/root/test/linux-source-2.6.29'
====== making target debian/stamp/do-pre-inst-arch [new prereqs:
do-post-build-arch]======
make[2]: Entering directory `/root/test/linux-source-2.6.29'
make[2]: warning: -jN forced in submake: disabling jobserver mode.
====== making target debian/stamp/post-build-indep [new prereqs:
do-build-indep]======
make[2]: Leaving directory `/root/test/linux-source-2.6.29'
====== making target debian/stamp/do-pre-inst-indep [new prereqs:
do-post-build-indep]======
make[2]: Entering directory `/root/test/linux-source-2.6.29'
make[2]: warning: -jN forced in submake: disabling jobserver mode.
====== making target debian/stamp/pre-inst-common [new prereqs: ]======
make[2]: Leaving directory `/root/test/linux-source-2.6.29'
/usr/bin/make -f debian/rules debian/stamp/pre-inst-indep
make[2]: Entering directory `/root/test/linux-source-2.6.29'
make[2]: warning: -jN forced in submake: disabling jobserver mode.
make[2]: `debian/stamp/pre-inst-common' is up to date.
make[2]: Leaving directory `/root/test/linux-source-2.6.29'
/usr/bin/make -f debian/rules debian/stamp/pre-inst-arch
make[2]: Entering directory `/root/test/linux-source-2.6.29'
make[2]: warning: -jN forced in submake: disabling jobserver mode.
====== making target debian/stamp/pre-inst-indep [new prereqs:
do-post-build-indep]======
make[2]: Leaving directory `/root/test/linux-source-2.6.29'
====== making target debian/stamp/do-install-indep [new prereqs:
do-pre-inst-indep]======
/usr/bin/make -f debian/rules debian/stamp/dep-install-indep
make[2]: Entering directory `/root/test/linux-source-2.6.29'
make[2]: warning: -jN forced in submake: disabling jobserver mode.
====== making target debian/stamp/pre-inst-arch [new prereqs:
do-post-build-arch]======
make[2]: Leaving directory `/root/test/linux-source-2.6.29'
====== making target debian/stamp/do-install-arch [new prereqs:
do-pre-inst-arch]======
/usr/bin/make -f debian/rules debian/stamp/dep-install-arch
make[2]: Entering directory `/root/test/linux-source-2.6.29'
make[2]: warning: -jN forced in submake: disabling jobserver mode.
make[2]: Entering directory `/root/test/linux-source-2.6.29'
make[2]: warning: -jN forced in submake: disabling jobserver mode.
====== making target
debian/stamp/install/linux-source-2.6.29-heisenberg [new prereqs:
]======
This is kernel package version 12.009.
make[1]: *** [debian/stamp/do-install-indep] Segmentation fault
make[1]: *** Waiting for unfinished jobs....
====== making target
debian/stamp/install/linux-headers-2.6.29-heisenberg [new prereqs:
]======
This is kernel package version 12.009.
rm -rf /root/test/linux-source-2.6.29/debian/linux-headers-2.6.29-heisenberg
install -p -d -o root -g root -m 755
/root/test/linux-source-2.6.29/debian/linux-headers-2.6.29-heisenberg/usr/src/linux-headers-2.6.29-heisenberg
install -p -d -o root -g root -m 755
/root/test/linux-source-2.6.29/debian/linux-headers-2.6.29-heisenberg/usr/share/doc/linux-headers-2.6.29-heisenberg/examples
install -p -d -o root -g root -m 755
/root/test/linux-source-2.6.29/debian/linux-headers-2.6.29-heisenberg/etc/kernel/header_postinst.d
install -p -d -o root -g root -m 755
/root/test/linux-source-2.6.29/debian/linux-headers-2.6.29-heisenberg/etc/kernel/header_preinst.d
install -p -d -o root -g root -m 755
/root/test/linux-source-2.6.29/debian/linux-headers-2.6.29-heisenberg/etc/kernel/header_postrm.d
install -p -d -o root -g root -m 755
/root/test/linux-source-2.6.29/debian/linux-headers-2.6.29-heisenberg/etc/kernel/header_prerm.d
install -p -d -o root -g root -m 755
/root/test/linux-source-2.6.29/debian/linux-headers-2.6.29-heisenberg/usr/src/linux-headers-2.6.29-heisenberg/arch/x86
install -p -d -o root -g root -m 755
/root/test/linux-source-2.6.29/debian/linux-headers-2.6.29-heisenberg/usr/src/linux-headers-2.6.29-heisenberg/arch/x86/kernel/
make[1]: *** [debian/stamp/do-install-arch] Segmentation fault
make[1]: Leaving directory `/root/test/linux-source-2.6.29'
dpkg-buildpackage: failure: debian/rules binary gave error exit status 2
make: *** [debian/stamp/build/buildpackage] Error 2
Any ideas?
Regards,
Chris.
-- System Information:
Debian Release: squeeze/sid
APT prefers unstable
APT policy: (500, 'unstable')
Architecture: amd64 (x86_64)
Kernel: Linux 2.6.28-heisenberg (SMP w/2 CPU cores; PREEMPT)
Locale: LANG=en_DE.UTF-8, LC_CTYPE=en_DE.UTF-8 (charmap=UTF-8)
Shell: /bin/sh linked to /bin/dash
Versions of packages kernel-package depends on:
ii binutils 2.19.1-1 The GNU assembler, linker
and bina
ii build-essential 11.4 Informational list of
build-essent
ii debianutils 3.0.1 Miscellaneous utilities
specific t
ii file 5.00-1 Determines file type
using "magic"
ii gettext 0.17-6 GNU Internationalization
utilities
ii make 3.81-5 The GNU version of the
"make" util
ii module-init-tools 3.7-pre9-1 tools for managing Linux
kernel mo
ii po-debconf 1.0.16 tool for managing
templates file t
ii util-linux 2.13.1.1-1 Miscellaneous system utilities
kernel-package recommends no packages.
Versions of packages kernel-package suggests:
ii bzip2 1.0.5-1 high-quality
block-sorting file co
pn docbook-utils <none> (no description available)
ii e2fsprogs 1.41.4-2 ext2/ext3/ext4 file
system utiliti
ii initramfs-tools [linux-in 0.93.2 tools for generating an initramfs
pn libdb3-dev <none> (no description available)
ii libncurses5-dev [libncurs 5.7+20090404-1 developer's libraries and
docs for
ii linux-source-2.6.29 [linu 2.6.29-3 Linux kernel source for
version 2.
pn xmlto <none> (no description available)
-- no debconf information
----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
Information forwarded
to debian-bugs-dist@lists.debian.org, Manoj Srivastava <srivasta@debian.org>:
Bug#525144; Package kernel-package.
(Wed, 22 Apr 2009 19:57:04 GMT) (full text, mbox, link).
Acknowledgement sent
to Manoj Srivastava <srivasta@acm.org>:
Extra info received and forwarded to list. Copy sent to Manoj Srivastava <srivasta@debian.org>.
(Wed, 22 Apr 2009 19:57:04 GMT) (full text, mbox, link).
Message #10 received at 525144@bugs.debian.org (full text, mbox, reply):
tags 525144 +unreproducible, moreinfo
severity 525144 normal
thanks
Hi,
I can not reproduce this. Also, I think there is definitely
something going on not reported int he bug report, since the command
line in the bug report creates a ./debian directory, and then reports
-- nothing to be done.
So, the command line was probably different, or there was a
./debian directory left over from before.
There is also the issue of make warning that -jN is being forced
on submakes, don't know what that is from.
I note that man make-kpkg says:
,----
| WARNING: Do NOT set the -j option in MAKEFLAGS directly, this
| shall cause the build to fail. Use CONCURRENCY_LEVEL as specified
| below.
`----
though I think the warning ought to be obsolete. If this is a make -j
bug, it should be investigated.
So, I ran
make-kpkg --revision 0.1 --append-to-version '-heisenberg' --initrd buildpackage
and it worked.
manoj
[anzu-2.6.29.1]__> make-kpkg --revision 0.1 --append-to-version '-heisenberg' --initrd
unlink debian/docs/README.modules
<SNIP>
rmdir debian
exec make kpkg_version=12.009 -f /usr/share/kernel-package/ruleset/minimal.mk debian DEBIAN_REVISION=0.1 APPEND_TO_VERSION=-heisenberg INITRD=YES
====== making target debian/stamp/conf/minimal_debian [new prereqs: ]======
This is kernel package version 12.009.
test -d debian || mkdir debian
test ! -e stamp-building || rm -f stamp-building
install -p -m 755 /usr/share/kernel-package/rules debian/rules
for file in ChangeLog Control Control.bin86 config templates.in rules; do \
cp -f /usr/share/kernel-package/$file ./debian/; \
done
for dir in Config docs examples ruleset scripts pkg po; do \
cp -af /usr/share/kernel-package/$dir ./debian/; \
done
test -f debian/control || sed -e 's/=V/2.6.29.1-heisenberg/g' \
-e 's/=D/0.1/g' -e 's/=A/amd64/g' \
-e 's/=SA//g' \
-e 's/=I//g' \
-e 's/=CV/2.6/g' \
-e 's/=M/Unknown Kernel Package Maintainer <unknown@unconfigured.in.etc.kernel-pkg.conf>/g' \
-e 's/=ST/linux/g' -e 's/=B/x86_64/g' \
/usr/share/kernel-package/Control > debian/control
test -f debian/changelog || sed -e 's/=V/2.6.29.1-heisenberg/g' \
-e 's/=D/0.1/g' -e 's/=A/amd64/g' \
-e 's/=ST/linux/g' -e 's/=B/x86_64/g' \
-e 's/=M/Unknown Kernel Package Maintainer <unknown@unconfigured.in.etc.kernel-pkg.conf>/g' \
/usr/share/kernel-package/changelog > debian/changelog
chmod 0644 debian/control debian/changelog
test -d ./debian/stamp || mkdir debian/stamp
make -f debian/rules debian/stamp/conf/kernel-conf
make[1]: Entering directory `/usr/local/src/kernel/linux-2.6'
====== making target debian/stamp/conf/kernel-conf [new prereqs: ]======
make EXTRAVERSION=.1-heisenberg ARCH=x86_64 \
oldconfig;
make[2]: Entering directory `/usr/local/src/kernel/linux-2.6'
scripts/kconfig/conf -o arch/x86/Kconfig
#
# configuration written to .config
#
make[2]: Leaving directory `/usr/local/src/kernel/linux-2.6'
make EXTRAVERSION=.1-heisenberg ARCH=x86_64 prepare
make[2]: Entering directory `/usr/local/src/kernel/linux-2.6'
scripts/kconfig/conf -s arch/x86/Kconfig
make[2]: Leaving directory `/usr/local/src/kernel/linux-2.6'
make[2]: Entering directory `/usr/local/src/kernel/linux-2.6'
CHK include/linux/version.h
CHK include/linux/utsrelease.h
SYMLINK include/asm -> include/asm-x86
CALL scripts/checksyscalls.sh
make[2]: Leaving directory `/usr/local/src/kernel/linux-2.6'
echo done > debian/stamp/conf/kernel-conf
make[1]: Leaving directory `/usr/local/src/kernel/linux-2.6'
make -f debian/rules debian/stamp/conf/full-changelog
make[1]: Entering directory `/usr/local/src/kernel/linux-2.6'
====== making target debian/stamp/conf/full-changelog [new prereqs: ]======
for file in ChangeLog Control Control.bin86 config templates.in rules; do \
cp -f /usr/share/kernel-package/$file ./debian/; \
done
for dir in Config docs examples ruleset scripts pkg po; do \
cp -af /usr/share/kernel-package/$dir ./debian/; \
done
install -p -m 755 /usr/share/kernel-package/rules debian/rules
sed -e 's/=V/2.6.29.1-heisenberg/g' \
-e 's/=D/0.1/g' -e 's/=A/amd64/g' \
-e 's/=SA//g' \
-e 's/=I//g' \
-e 's/=CV/2.6/g' \
-e 's/=M/Unknown Kernel Package Maintainer <unknown@unconfigured.in.etc.kernel-pkg.conf>/g' \
-e 's/=ST/linux/g' -e 's/=B/x86_64/g' \
/usr/share/kernel-package/Control > debian/control
sed -e 's/=V/2.6.29.1-heisenberg/g' -e 's/=D/0.1/g' \
-e 's/=A/amd64/g' -e 's/=M/Unknown Kernel Package Maintainer <unknown@unconfigured.in.etc.kernel-pkg.conf>/g' \
-e 's/=ST/linux/g' -e 's/=B/x86_64/g' \
/usr/share/kernel-package/changelog > debian/changelog
chmod 0644 debian/control debian/changelog
make -f debian/rules debian/stamp/conf/kernel-conf
make[2]: Entering directory `/usr/local/src/kernel/linux-2.6'
make[2]: `debian/stamp/conf/kernel-conf' is up to date.
make[2]: Leaving directory `/usr/local/src/kernel/linux-2.6'
make[1]: Leaving directory `/usr/local/src/kernel/linux-2.6'
echo done > debian/stamp/conf/minimal_debian
exec debian/rules DEBIAN_REVISION=0.1 APPEND_TO_VERSION=-heisenberg INITRD=YES
nothing to be done.
--
He played the king as if afraid someone else would play the ace. John
Mason Brown, drama critic
Manoj Srivastava <srivasta@acm.org> <http://www.golden-gryphon.com/>
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Information forwarded
to debian-bugs-dist@lists.debian.org, Manoj Srivastava <srivasta@debian.org>:
Bug#525144; Package kernel-package.
(Wed, 22 Apr 2009 20:42:07 GMT) (full text, mbox, link).
Acknowledgement sent
to Manoj Srivastava <srivasta@acm.org>:
Extra info received and forwarded to list. Copy sent to Manoj Srivastava <srivasta@debian.org>.
(Wed, 22 Apr 2009 20:42:07 GMT) (full text, mbox, link).
Message #15 received at 525144@bugs.debian.org (full text, mbox, reply):
Hi,
More testing.
DEB_BUILD_OPTIONS=parallel=3 make-kpkg --rootcmd=fakeroot \
--revision 0.1 --append-to-version '-heisenberg' --initrd\
--uc --us kernel-image
Works for me as well, tried twice in a row.
manoj
--
Now is the time for all good men to come to. Walt Kelly
Manoj Srivastava <srivasta@acm.org> <http://www.golden-gryphon.com/>
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Information forwarded
to debian-bugs-dist@lists.debian.org, Manoj Srivastava <srivasta@debian.org>:
Bug#525144; Package kernel-package.
(Wed, 22 Apr 2009 22:42:02 GMT) (full text, mbox, link).
Acknowledgement sent
to Christoph Anton Mitterer <christoph.anton.mitterer@physik.uni-muenchen.de>:
Extra info received and forwarded to list. Copy sent to Manoj Srivastava <srivasta@debian.org>.
(Wed, 22 Apr 2009 22:42:02 GMT) (full text, mbox, link).
Message #20 received at 525144@bugs.debian.org (full text, mbox, reply):
[Message part 1 (text/plain, inline)]
On Wed, 2009-04-22 at 14:38 -0500, Manoj Srivastava wrote:
> I note that man make-kpkg says:
> ,----
> | WARNING: Do NOT set the -j option in MAKEFLAGS directly, this
> | shall cause the build to fail. Use CONCURRENCY_LEVEL as specified
> | below.
> `----
> though I think the warning ought to be obsolete. If this is a make -j
> bug, it should be investigated.
I actually use CONCURRENCY_LEVEL and when removing it,... buildpackage
works.
Have you tried it with CONCURRENCY_LEVEL := 2 in ~/.kernel-pkg.conf?
btw: Perhaps I'm wrong, but I feel, that the
non-kernel/modules-compilation part of make-kpkg takes ages longer than
with the 11.x branch.
Is this possible?
Thanks,
Chris.
[smime.p7s (application/x-pkcs7-signature, attachment)]
Tags added: unreproducible, moreinfo
Request was from Manoj Srivastava <srivasta@golden-gryphon.com>
to control@bugs.debian.org.
(Thu, 23 Apr 2009 14:54:05 GMT) (full text, mbox, link).
Severity set to `normal' from `grave'
Request was from Manoj Srivastava <srivasta@golden-gryphon.com>
to control@bugs.debian.org.
(Thu, 23 Apr 2009 14:54:06 GMT) (full text, mbox, link).
Information forwarded
to debian-bugs-dist@lists.debian.org, Manoj Srivastava <srivasta@debian.org>:
Bug#525144; Package kernel-package.
(Thu, 23 Apr 2009 16:33:05 GMT) (full text, mbox, link).
Acknowledgement sent
to Sven Hartge <sven@svenhartge.de>:
Extra info received and forwarded to list. Copy sent to Manoj Srivastava <srivasta@debian.org>.
(Thu, 23 Apr 2009 16:33:05 GMT) (full text, mbox, link).
Message #29 received at 525144@bugs.debian.org (full text, mbox, reply):
Hi!
I am seeing this segfault as well and I am using
CONCURRENCY_LEVEL:=5
in /etc/kernel-pkg.conf. But commenting this line and using the following
command
DEB_BUILD_OPTIONS=parallel=5 fakeroot make-kpkg --revision=skuld.338
--append-to-version=-338 --arch=amd64 --cross-compile=- buildpackage
also causes make to segfault.
If I don't use any parallel build options, the package(s) build fine
without error.
So something inside kernel-packages makefiles triggers a make bug when
used in a parallel build situation.
Grüße,
Sven.
Information forwarded
to debian-bugs-dist@lists.debian.org, Manoj Srivastava <srivasta@debian.org>:
Bug#525144; Package kernel-package.
(Thu, 23 Apr 2009 20:27:05 GMT) (full text, mbox, link).
Acknowledgement sent
to Manoj Srivastava <srivasta@acm.org>:
Extra info received and forwarded to list. Copy sent to Manoj Srivastava <srivasta@debian.org>.
(Thu, 23 Apr 2009 20:27:05 GMT) (full text, mbox, link).
Message #34 received at 525144@bugs.debian.org (full text, mbox, reply):
severity 525144 important
retitle 525144 libfakeroot segfaults when building kernels in parallel
reassign 525144 fakeroot
thanks
Hi,
This is true for kernel-package version 12.009. I am putting in
a work around in kenrel-package 12.010.
Either when DEB_BUILD_OPTIONS=parallel=3, or wehn
CONCURRENCY_LEVEL=3 is set in /etc/kernel-pkg.conf, calling
make-kpkg --rootcmd=fakeroot buildpackage
fails with a seg fault.
| | serial build | parallel build |
|---------------+--------------+----------------|
| fakeroot | Success | *Failure* |
| sudo | Success | Success |
| build as root | Success | Success |
|---------------+--------------+----------------|
It seems pretty clear that there is some interaction with
fakeroot and the buildpackage target.
kernel_image, kernel_headers, and most non-buildpackage targets have
this calling pattern:
make-kpkg -> ./debian/rules <target> -> fakeroot make <some target>
This works no matter what, serial or parallel.
With buildpackage, things get a little more complex.
make-kpkg --> ./debian/rules buildpackage -> dpkg-buildpackage -->
fakeroot ./debian/rules <some target>
This works when we are not building in parallel, fails when we
are.
Now, sudo works just fin in either case, so it is not a make
fault or a kernel-package fault that I can see, though I can just
disable parallel builds when buildpackage is the target.
manoj
--
The cart has no place where a fifth wheel could be used. Herbert von
Fritzlar
Manoj Srivastava <srivasta@acm.org> <http://www.golden-gryphon.com/>
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Severity set to `important' from `normal'
Request was from Manoj Srivastava <srivasta@acm.org>
to control@bugs.debian.org.
(Thu, 23 Apr 2009 20:27:12 GMT) (full text, mbox, link).
Changed Bug title to `libfakeroot segfaults when building kernels in parallel' from `make-kpgk fails to build kernel (segfaults)'.
Request was from Manoj Srivastava <srivasta@acm.org>
to control@bugs.debian.org.
(Thu, 23 Apr 2009 20:27:13 GMT) (full text, mbox, link).
Bug reassigned from package `kernel-package' to `fakeroot'.
Request was from Manoj Srivastava <srivasta@acm.org>
to control@bugs.debian.org.
(Thu, 23 Apr 2009 20:27:14 GMT) (full text, mbox, link).
Information forwarded
to debian-bugs-dist@lists.debian.org, Clint Adams <schizo@debian.org>:
Bug#525144; Package fakeroot.
(Thu, 23 Apr 2009 20:36:03 GMT) (full text, mbox, link).
Acknowledgement sent
to Manoj Srivastava <srivasta@acm.org>:
Extra info received and forwarded to list. Copy sent to Clint Adams <schizo@debian.org>.
(Thu, 23 Apr 2009 20:36:03 GMT) (full text, mbox, link).
Message #45 received at 525144@bugs.debian.org (full text, mbox, reply):
On Thu, Apr 23 2009, Sven Hartge wrote:
> I am seeing this segfault as well and I am using
> CONCURRENCY_LEVEL:=5
> in /etc/kernel-pkg.conf. But commenting this line and using the following
> command
> DEB_BUILD_OPTIONS=parallel=5 fakeroot make-kpkg --revision=skuld.338
> --append-to-version=-338 --arch=amd64 --cross-compile=- buildpackage
> also causes make to segfault.
It is not make which is crashing -- it is libfakeroot. Try the
same build command with --rootcmd=sudo, and the build works just
fine. I even tested with setting CONCURRENCY_LEVEL in
/etc/kernel-img.conf, in case sudo was stripping off the
DEB_BUILD_OPTION
So the problem is not in make or kernel-package, it is in fakeroot.
> If I don't use any parallel build options, the package(s) build fine
> without error.
> So something inside kernel-packages makefiles triggers a make bug when
> used in a parallel build situation.
Here is the scenario:
On Wed, Apr 22 2009, Christoph Anton Mitterer wrote:
> I actually use CONCURRENCY_LEVEL and when removing it,... buildpackage
> works.
> Have you tried it with CONCURRENCY_LEVEL := 2 in ~/.kernel-pkg.conf?
Yes.
> btw: Perhaps I'm wrong, but I feel, that the
> non-kernel/modules-compilation part of make-kpkg takes ages longer than
> with the 11.x branch.
> Is this possible?
Perhaps. 11.XXX kernel-package was very unfriendly to parallel
compilation, with the current 12.XX all targets except buildpackage can
be compiled in parallel. This has cause a little bit of a slowdown, as
the build assures synchronization.
manoj
--
Q: Why is it that Mexico isn't sending anyone to the '84 summer games?
A: Anyone in Mexico who can run, swim or jump is already in LA.
Manoj Srivastava <srivasta@acm.org> <http://www.golden-gryphon.com/>
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Information forwarded
to debian-bugs-dist@lists.debian.org:
Bug#525144; Package fakeroot.
(Thu, 23 Apr 2009 20:54:07 GMT) (full text, mbox, link).
Acknowledgement sent
to Clint Adams <schizo@debian.org>:
Extra info received and forwarded to list.
(Thu, 23 Apr 2009 20:54:07 GMT) (full text, mbox, link).
Message #50 received at 525144@bugs.debian.org (full text, mbox, reply):
On Thu, Apr 23, 2009 at 03:13:30PM -0500, Manoj Srivastava wrote:
> This works when we are not building in parallel, fails when we
> are.
Is fakeroot being invoked multiple times in parallel? Are environment
variables being shared between them? Are pthreads involved?
Does substituting fakeroot-tcp make things magically better?
Information forwarded
to debian-bugs-dist@lists.debian.org, Clint Adams <schizo@debian.org>:
Bug#525144; Package fakeroot.
(Thu, 23 Apr 2009 21:36:07 GMT) (full text, mbox, link).
Acknowledgement sent
to Sven Hartge <sven@svenhartge.de>:
Extra info received and forwarded to list. Copy sent to Clint Adams <schizo@debian.org>.
(Thu, 23 Apr 2009 21:36:07 GMT) (full text, mbox, link).
Message #55 received at 525144@bugs.debian.org (full text, mbox, reply):
Um 20:51 Uhr am 23.04.09 schrieb Clint Adams:
> On Thu, Apr 23, 2009 at 03:13:30PM -0500, Manoj Srivastava wrote:
>> This works when we are not building in parallel, fails when we
>> are.
> Is fakeroot being invoked multiple times in parallel? Are environment
> variables being shared between them? Are pthreads involved? Does
> substituting fakeroot-tcp make things magically better?
Difficult to test. With fakeroot-tcp I am not able to fully compile a
kernel, the process stops at different steps each attempt and leaves a
stray faked-tcp process.
Grüße,
Sven.
Information forwarded
to debian-bugs-dist@lists.debian.org, Clint Adams <schizo@debian.org>:
Bug#525144; Package fakeroot.
(Thu, 23 Apr 2009 22:15:03 GMT) (full text, mbox, link).
Acknowledgement sent
to Manoj Srivastava <srivasta@debian.org>:
Extra info received and forwarded to list. Copy sent to Clint Adams <schizo@debian.org>.
(Thu, 23 Apr 2009 22:15:03 GMT) (full text, mbox, link).
Message #60 received at 525144@bugs.debian.org (full text, mbox, reply):
On Thu, Apr 23 2009, Clint Adams wrote:
> On Thu, Apr 23, 2009 at 03:13:30PM -0500, Manoj Srivastava wrote:
>> This works when we are not building in parallel, fails when we
>> are.
>
> Is fakeroot being invoked multiple times in parallel?
This is likely, since dpkg-buildpackage might be trying to build
linux-headers. linux-image, linux-debug, linux-dc, and linux-source
packages in parallel, which is what happens when the binary target is
called.
> Are environment variables being shared between them?
Shared? well, probably, since the y inherit the same
environment, and make is called recursively.
> Are pthreads involved?
I do not think so. This is process based parallelism.
> Does substituting fakeroot-tcp make things magically better?
Not really.
manoj
--
There is nothing wrong with abstinence, in moderation.
Manoj Srivastava <srivasta@debian.org> <http://www.debian.org/~srivasta/>
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Information forwarded
to debian-bugs-dist@lists.debian.org:
Bug#525144; Package fakeroot.
(Thu, 23 Apr 2009 22:42:02 GMT) (full text, mbox, link).
Acknowledgement sent
to Clint Adams <schizo@debian.org>:
Extra info received and forwarded to list.
(Thu, 23 Apr 2009 22:42:03 GMT) (full text, mbox, link).
Message #65 received at 525144@bugs.debian.org (full text, mbox, reply):
On Thu, Apr 23, 2009 at 04:59:28PM -0500, Manoj Srivastava wrote:
> This is likely, since dpkg-buildpackage might be trying to build
> linux-headers. linux-image, linux-debug, linux-dc, and linux-source
> packages in parallel, which is what happens when the binary target is
> called.
Can we get a backtrace of the segfault?
Information forwarded
to debian-bugs-dist@lists.debian.org, Clint Adams <schizo@debian.org>:
Bug#525144; Package fakeroot.
(Thu, 23 Apr 2009 22:42:04 GMT) (full text, mbox, link).
Acknowledgement sent
to Christoph Anton Mitterer <christoph.anton.mitterer@physik.uni-muenchen.de>:
Extra info received and forwarded to list. Copy sent to Clint Adams <schizo@debian.org>.
(Thu, 23 Apr 2009 22:42:04 GMT) (full text, mbox, link).
Message #70 received at 525144@bugs.debian.org (full text, mbox, reply):
[Message part 1 (text/plain, inline)]
On Thu, 2009-04-23 at 15:13 -0500, Manoj Srivastava wrote:
> | | serial build | parallel build |
> |---------------+--------------+----------------|
> | fakeroot | Success | *Failure* |
> | sudo | Success | Success |
> | build as root | Success | Success |
> |---------------+--------------+----------------|
I used make-kpkg always directly as root.... and it still failed,...?!
Or does this somehow internally still use fakeroot?
Cheers,
Chris.
[smime.p7s (application/x-pkcs7-signature, attachment)]
Information forwarded
to debian-bugs-dist@lists.debian.org, Clint Adams <schizo@debian.org>:
Bug#525144; Package fakeroot.
(Thu, 23 Apr 2009 22:45:05 GMT) (full text, mbox, link).
Acknowledgement sent
to Christoph Anton Mitterer <christoph.anton.mitterer@physik.uni-muenchen.de>:
Extra info received and forwarded to list. Copy sent to Clint Adams <schizo@debian.org>.
(Thu, 23 Apr 2009 22:45:05 GMT) (full text, mbox, link).
Message #75 received at 525144@bugs.debian.org (full text, mbox, reply):
[Message part 1 (text/plain, inline)]
On Thu, 2009-04-23 at 15:18 -0500, Manoj Srivastava wrote:
> > btw: Perhaps I'm wrong, but I feel, that the
> > non-kernel/modules-compilation part of make-kpkg takes ages longer than
> > with the 11.x branch.
> > Is this possible?
>
> Perhaps. 11.XXX kernel-package was very unfriendly to parallel
> compilation, with the current 12.XX all targets except buildpackage can
> be compiled in parallel. This has cause a little bit of a slowdown, as
> the build assures synchronization.
Hmm,.. ok,... thanks,...
I've just felt that it is really _very much_ slower ;)
Thanks,
Chris.
[smime.p7s (application/x-pkcs7-signature, attachment)]
Information forwarded
to debian-bugs-dist@lists.debian.org, Clint Adams <schizo@debian.org>:
Bug#525144; Package fakeroot.
(Fri, 24 Apr 2009 00:36:02 GMT) (full text, mbox, link).
Acknowledgement sent
to Manoj Srivastava <srivasta@debian.org>:
Extra info received and forwarded to list. Copy sent to Clint Adams <schizo@debian.org>.
(Fri, 24 Apr 2009 00:36:02 GMT) (full text, mbox, link).
Message #80 received at 525144@bugs.debian.org (full text, mbox, reply):
On Thu, Apr 23 2009, Christoph Anton Mitterer wrote:
> On Thu, 2009-04-23 at 15:13 -0500, Manoj Srivastava wrote:
>> | | serial build | parallel build |
>> |---------------+--------------+----------------|
>> | fakeroot | Success | *Failure* |
>> | sudo | Success | Success |
>> | build as root | Success | Success |
>> |---------------+--------------+----------------|
> I used make-kpkg always directly as root.... and it still failed,...?!
> Or does this somehow internally still use fakeroot?
Full logs would be nice.
For the record, I just compiled in parallel as root, and
everything still worked.
manoj
--
The computer can't tell you the emotional story. It can give you the
exact mathematical design, but what's missing is the eyebrows.- Frank
Zappa
Manoj Srivastava <srivasta@debian.org> <http://www.debian.org/~srivasta/>
1024D/BF24424C print 4966 F272 D093 B493 410B 924B 21BA DABB BF24 424C
Send a report that this bug log contains spam.
Debian bug tracking system administrator <owner@bugs.debian.org>.
Last modified:
Tue Jan 9 18:42:38 2018;
Machine Name:
buxtehude
Debian Bug tracking system
Debbugs is free software and licensed under the terms of the GNU
Public License version 2. The current version can be obtained
from https://bugs.debian.org/debbugs-source/.
Copyright © 1999 Darren O. Benham,
1997,2003 nCipher Corporation Ltd,
1994-97 Ian Jackson,
2005-2017 Don Armstrong, and many other contributors.