[oe] [PATCH] systemd: update to v26

Koen Kooi koen at dominion.thruhere.net
Fri May 6 09:30:08 UTC 2011


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 06-05-11 11:19, Paul Menzel wrote:
> Dear Koen,
> 
> 
> Am Montag, den 02.05.2011, 14:56 +0200 schrieb Paul Menzel:
>> Am Montag, den 02.05.2011, 13:50 +0200 schrieb Paul Menzel:
>>> Am Montag, den 02.05.2011, 10:22 +0200 schrieb Koen Kooi:
>>>> Signed-off-by: Koen Kooi <koen at openembedded.org>
>>>> ---
>>>>  recipes/systemd/files/disable_xml_generation.patch |   36 --------
>>>>  recipes/systemd/files/execute.patch                |   17 ----
>>>>  recipes/systemd/files/getty-serial at .service        |   19 ----
>>>>  recipes/systemd/files/replace_accpet4.patch        |   26 ------
>>>>  recipes/systemd/files/systemadm.patch              |   13 ---
>>>>  ...-disable-xml-file-stuff-and-introspection.patch |   93 ++++++++++++++++++++
>>>>  recipes/systemd/systemd-v26/execute.patch          |   17 ++++
>>>>  recipes/systemd/systemd-v26/getty-serial at .service  |   19 ++++
>>>>  recipes/systemd/systemd-v26/replace_accpet4.patch  |   26 ++++++
>>>>  recipes/systemd/systemd-v26/systemadm.patch        |   13 +++
>>>>  recipes/systemd/systemd_git.bb                     |   26 ++++--
>>>>  11 files changed, 185 insertions(+), 120 deletions(-)
>>>>  delete mode 100644 recipes/systemd/files/disable_xml_generation.patch
>>>>  delete mode 100644 recipes/systemd/files/execute.patch
>>>>  delete mode 100644 recipes/systemd/files/getty-serial at .service
>>>>  delete mode 100644 recipes/systemd/files/replace_accpet4.patch
>>>>  delete mode 100644 recipes/systemd/files/systemadm.patch
>>>>  create mode 100644 recipes/systemd/systemd-v26/0001-systemd-disable-xml-file-stuff-and-introspection.patch
>>>>  create mode 100644 recipes/systemd/systemd-v26/execute.patch
>>>>  create mode 100644 recipes/systemd/systemd-v26/getty-serial at .service
>>>>  create mode 100644 recipes/systemd/systemd-v26/replace_accpet4.patch
>>>>  create mode 100644 recipes/systemd/systemd-v26/systemadm.patch
>>>
>>> `git log --stat -M -C -C` makes the diff lock much more daunting.
>>>
>>> […]
>>>
>>>> --- /dev/null
>>>> +++ b/recipes/systemd/systemd-v26/systemadm.patch
>>>> @@ -0,0 +1,13 @@
>>>> +Index: git/src/systemadm.vala
>>>> +===================================================================
>>>> +--- git.orig/src/systemadm.vala	2010-06-15 11:02:35.863509632 +0200
>>>> ++++ git/src/systemadm.vala	2010-06-15 11:06:16.305986249 +0200
>>>> +@@ -297,7 +297,7 @@
>>>> + 
>>>> +                 bbox.pack_start(cancel_button, false, true, 0);
>>>> + 
>>>> +-                bus = Bus.get(session ? BusType.SESSION : BusType.SYSTEM);
>>>> ++                bus = DBus.Bus.get(session ? DBus.BusType.SESSION : DBus.BusType.SYSTEM);
>>>> + 
>>>> +                 manager = bus.get_object(
>>>> +                                 "org.freedesktop.systemd1",
>>>> diff --git a/recipes/systemd/systemd_git.bb b/recipes/systemd/systemd_git.bb
>>>> index 5f98e54..7634b22 100644
>>>> --- a/recipes/systemd/systemd_git.bb
>>>> +++ b/recipes/systemd/systemd_git.bb
>>
>> Thinking about it, should not this be named `system_26.bb` to avoid
>> confusion?
>>
>>>> @@ -1,48 +1,56 @@
>>>>  DESCRIPTION = "Systemd a init replacement"
>>>>  HOMEPAGE = "http://www.freedesktop.org/wiki/Software/systemd"
>>>> -LICENSE = "GPL"
>>>> +LICENSE = "GPLv2+
>>>> +LIC_FILES_CHKSUM = "file://LICENSE;md5=751419260aa954499f7abaabaa882bbe"
>>>> +
>>>>  DEPENDS = "readline udev dbus libcap2 libcgroup gtk+"
>>
>> `gtk+` should be removed. Actually I would prefer a separate patch
>> including disabling it.
>>
>>>>  PRIORITY = "optional"
>>>>  SECTION = "base/shell"
>>>>  
>>>> -PV = "0.0"
>>>> -PR_append = "+${SRCPV}"
>>>> +PV = "v26"
>>>> +#PR_append = "+${SRCPV}"
>>>
>>> Remove the commented line.
>>>  
>>>>  inherit autotools vala
>>>>  
>>>> -TAG = "1ebdf2d8793181f31b766b8342391aa1978f9917"
>>>> +TAG = "${PV}"
>>>>  
>>>>  SRC_URI = "git://anongit.freedesktop.org/systemd;protocol=git;tag=${TAG} \
>>>>             file://execute.patch \
>>>> -           file://systemadm.patch \
>>>> -           file://disable_xml_generation.patch \
>>>> +#           file://systemadm.patch \
>>>
>>> Maybe a comment in the recipe or the commit message would be nice, why
>>> this is commented. Could it be needed in the future?
>>>
>>>> +           file://0001-systemd-disable-xml-file-stuff-and-introspection.patch \
>>>>             file://replace_accpet4.patch \
>>>>             file://getty-serial@.service \
>>>>            "
>>>>  
>>>>  S = "${WORKDIR}/git"
>>>>  
>>>> +# The gtk+ tools should get built as a seperate recipe e.g. systemd-tools
>>>
>>> s/seperate/separate/
>>>
>>> … to not force people to build all gtk+ packages/recipes.
>>>
>>>>  EXTRA_OECONF = " --with-distro=debian \
>>>>                   --with-rootdir=${base_prefix} \
>>>> +                 --disable-gtk \
>>>>                 "
>>>>  
>>>> -PACKAGES = "${PN} ${PN}-gui ${PN}-dbg ${PN}-doc"
>>>> +PACKAGES =+ "${PN}-gui"
>>>>  
>>>>  FILES_${PN}-gui = "${bindir}/systemadm"
>>>>  
>>>>  FILES_${PN} = " ${base_bindir}/* \
>>>>                  ${datadir}/dbus-1/services \
>>>>                  ${datadir}/dbus-1/system-services \
>>>> +                ${datadir}/polkit-1 \
>>>>                  ${datadir}/${PN} \
>>>>                  ${sysconfdir} \
>>>>                  ${base_libdir}/systemd/* \
>>>>                  ${base_libdir}/systemd/system/* \
>>>>                  ${base_libdir}/udev/rules.d \
>>>> +                ${base_libdir}/security/*.so \
>>>>                  /cgroup \
>>>> -                ${bindir}/systemd-install \
>>>> +                ${bindir}/systemd* \
>>>> +                ${libdir}/tmpfiles.d/*.conf \
>>>> +                ${libdir}/systemd \
>>>>                 "
>>>>  
>>>> -FILES_${PN}-dbg += " /lib/systemd/.debug "
>>>> +FILES_${PN}-dbg += "${base_libdir}/systemd/.debug ${base_libdir}/systemd/*/.debug"
>>>>  
>>>>  def get_baudrate(bb, d):
>>>>      return bb.data.getVar('SERIAL_CONSOLE', d, 1).split()[0]
>>>
>>> A split in two commits to fix/improve the recipe and to update to v26
>>> would have been nice.
>>>
>>> Anyway, using `angstrom-2010.x` for `MACHINE = "beagleboard"` fails with
>>> the following error message.
>>>
>>>         […]
>>>           CCLD   systemd-getty-generator
>>>           GEN    man/systemd.1
>>>         I/O error : Attempt to load network entity http://docbook.sourceforge.net/release/xsl/current/manpages/docbook.xsl
>>>         warning: failed to load external entity "http://docbook.sourceforge.net/release/xsl/current/manpages/docbook.xsl"
>>>         cannot parse http://docbook.sourceforge.net/release/xsl/current/manpages/docbook.xsl
>>>         make[1]: *** [man/systemd.1] Error 4
>>>         make[1]: Leaving directory `/oe/build-angstrom-next/angstrom-dev/work/armv7a-angstrom-linux-gnueabi/systemd-v26-r0/git'
>>>         make: *** [all] Error 2
>>>         + die 'oe_runmake failed'
>>>         + oefatal 'oe_runmake failed'
>>>         + echo FATAL: 'oe_runmake failed'
>>>         FATAL: oe_runmake failed
>>>         + exit 1
>>>
>>> Please find `log.do_compile` attached.
> 
> why did you ignore my comments and pushed anyway in 1f14659d [1]?

You might have noticed that the patch I pushed is quite different from
the one I sent to this list.

Anyway, Henning OK'd the patch:

10:00 < koen> woglinde_: OK to commit the patch so you can fix up epoll
later?
10:01 < woglinde_> koen its booting
10:01 < woglinde_> so commit it

And since he's the maintainer I pushed it.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)

iD8DBQFNw7+gMkyGM64RGpERAnAwAJ9Hkk3agonPzHzp06YrZzPD2sbBPQCghkEU
g0+AycJHMgLgbbj+mR0oIEI=
=9eoo
-----END PGP SIGNATURE-----





More information about the Openembedded-devel mailing list