Differences between revisions 22 and 23
Revision 22 as of 2015-01-06 18:54:47
Size: 4795
Editor: ?HelmutGrohne
Comment: more thanks
Revision 23 as of 2015-01-06 18:56:36
Size: 4850
Editor: ?HelmutGrohne
Comment: porter thanks
Deletions are marked like this. Additions are marked like this.
Line 35: Line 35:
   * mips64el: YunQiang Su
Line 37: Line 38:
   * x32: Daniel Schepler

rebootstrap

rebootstrap is a crude hack to test the bootstrappability of Debian architectures. In this context bootstrapping refers to building binaries for an architecture from source without using any pre-built binaries for that architecture. In the past 20 years about 20 architectures have been bootstrapped for Debian. At all times this has been a manual and non-repeatable process. rebootstrap is trying to address the very early bootstrap phase involving the gcc/eglibc dance.

The source can be obtained using git:

git clone git://anonscm.debian.org/users/helmutg/rebootstrap.git
# or
git clone git.debian.org:/git/users/helmutg/rebootstrap.git

One of rebootstraps major goals is to highlight bugs.

rebootstrap on jenkins.d.n

Contact Helmut Grohne or stop by #debian-bootstrap on OFTC if you have any questions.

Occasionally, rebootstrap works around limitations of packages by carrying patches locally. This raises the question of which patches are acceptable. Rules are:

  1. Patches must be intended for inclusion into the patched package and therefore should be filed as a bug against the relevant package.
  2. Patches beyond a couple of hundred lines are simply too long atm.

Thanks

Without the following (non-exhaustive list of) people rebootstrap would not be where it is today. Thanks!

  • Porters for providing patches
    • arm64: Wookey
    • mips64el: ?YunQiang Su

    • or1k: Christian Svensson
    • ppc64el: Breno Leitao, Mauricio Faria de Oliveira, Erwan Prioul, IBM
    • x32: Daniel Schepler
  • Package maintainers for applying patches, in particular Matthias Klose for being quick
  • Infrastructure:
    • Holger Levson for administrating jenkins.d.n
    • ProfitBricks for donating the hardware running jenkins.d.n

  • Advice and other kinds of help
    • Guillem Jover for telling how to do things correctly
    • Bernhard R. Link for explaining difficult error messages
    • Peter Pentchev for fixing bugs that rebootstrap discovers tomorrow
    • Johannes Schauer for making things work in theory
    • Wookey, for NMUing lots of packages
  • Many others! (Multiarch, Emdebian, etc.)

Issues

Below is a non-exhaustive list of unsolved problems. Patches welcome.

alpha

libc6.1-dev:alpha conflicts with libc6-dev:amd64. Thus eglibc stage1 is not installable.

dpkg: regarding libc6.1-dev_2.18-4_alpha.deb containing libc6.1-dev:alpha:
 libc6.1-dev conflicts with libc6-dev
  libc6-dev:amd64 (version 2.18-4) is present and installed.

Wookey: It's correct for libc6.1:alpha to conflict with libc6-dev:i386 (because they have the same linux loader path: /lib/ld-linux.so.2

Wookey: Has i386/amd64 got confised here? I can't see any reason why it should conflict with

Helmut: Does that mean, alpha should be cross-built from a i386 build machine rather than an amd64 build machine?

arm

OABI is dead. There is no support for arm in linux (no target binary-libc-dev_arm), gcc-4.9 (target not supported) or glibc anymore.

hppa

It is not possible to cross-build binutils for hppa64 (relevant code not present), but gcc needs binutils-hppa64 and it is not possible to compile linux without hppa64 support. 745116

ia64

Same as alpha.

Wookey: This makes no obvious sense as alpha had a uniqueloadepat ko

dpkg: regarding libc6.1-dev_2.18-4_ia64.deb containing libc6.1-dev:ia64:
 libc6.1-dev conflicts with libc6-dev
  libc6-dev:amd64 (version 2.18-4) is present and installed.

In addition Debian's glibc dropped support for ia64.

kfreebsd-*

kfreebsd works entirely different to Linux architectures. bootstrapping involves building clang rather than gcc and a completely different order. It is not clear whether it is feasible to bootstrap a kfreebsd system from a Linux system at all. Thus it is not feasible to cover it in rebootstrap for the time being.

or1k

gcc as found in Debian sid for version 4.8 and 4.9 do not support or1k as a target. Thus gcc stage1 fails to configure. It is expected that this issue vanishes with future gcc upstream releases.

s390

eglibc does not support this architecture:

dh_gencontrol -plibc6-dev
dpkg-gencontrol: error: current host architecture 's390' does not appear in package's architecture list (amd64 arm arm64 armel armhf hppa i386 m68k mips mipsel powerpc powerpcspe ppc64 ppc64el sparc sparc64 s390x sh4 x32)