[RFT][PATCH 0/2] Systemd 209

[RFT][PATCH 0/2] Systemd 209

[Khem Raj]

This patchset upgrades systemd to 209 as can be seen some enahancements
to uclibc were neeed. As of now. It builds for me on both uclibc/eglibc
Please test it out in your setups and report back any issues

The following changes since commit 9a98b8bfe13b94d89a9e39cca821f26fdeaeffc3:

  image_types.bbclass: Fix tar IMAGE_CMD to not change directories (2014-02-18 08:37:03 +0000)

are available in the git repository at:

  git://git.openembedded.org/openembedded-core-contrib kraj/systemd-209
  http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=kraj/systemd-209

Khem Raj (2):
  uclibc: Add new functionality needed for systemd 209
  systemd: Upgrade 208 -> 209

 ...-Use-bin-mkdir-instead-of-host-mkdir-path.patch |  31 --
 ...1-uClibc-doesn-t-implement-pwritev-preadv.patch |  15 +-
 ...ournal-Add-missing-byte-order-conversions.patch |  60 ----
 .../journal-file-protect-against-alloca-0.patch    |  23 --
 ...n-appending-to-journal-file-allocate-larg.patch |  83 -----
 .../systemd/journald-add-missing-error-check.patch |  16 -
 .../systemd/systemd/systemd-pam-fix-execvpe.patch  |  22 +-
 .../systemd/systemd-pam-fix-fallocate.patch        |  24 +-
 .../systemd/systemd-pam-fix-getty-unit.patch       |  20 +-
 .../systemd/systemd/systemd-pam-fix-mkostemp.patch | 166 +--------
 .../systemd/systemd/systemd-pam-fix-msformat.patch | 115 ++++---
 .../systemd/systemd/uclibc-sysinfo_h.patch         |  19 ++
 .../systemd/{systemd_208.bb => systemd_209.bb}     |  14 +-
 meta/recipes-core/uclibc/uclibc-git.inc            |   3 +
 .../0001-Add-eventfd_read-and-eventfd_write.patch  | 120 +++++++
 .../uclibc-git/0002-wire-setns-syscall.patch       |  68 ++++
 ...tl.h-Define-F_SETPIPE_SZ-and-F_GETPIPE_SZ.patch | 377 +++++++++++++++++++++
 17 files changed, 702 insertions(+), 474 deletions(-)
 delete mode 100644 meta/recipes-core/systemd/systemd/0001-Use-bin-mkdir-instead-of-host-mkdir-path.patch
 delete mode 100644 meta/recipes-core/systemd/systemd/journal-Add-missing-byte-order-conversions.patch
 delete mode 100644 meta/recipes-core/systemd/systemd/journal-file-protect-against-alloca-0.patch
 delete mode 100644 meta/recipes-core/systemd/systemd/journal-when-appending-to-journal-file-allocate-larg.patch
 delete mode 100644 meta/recipes-core/systemd/systemd/journald-add-missing-error-check.patch
 create mode 100644 meta/recipes-core/systemd/systemd/uclibc-sysinfo_h.patch
 rename meta/recipes-core/systemd/{systemd_208.bb => systemd_209.bb} (96%)
 create mode 100644 meta/recipes-core/uclibc/uclibc-git/0001-Add-eventfd_read-and-eventfd_write.patch
 create mode 100644 meta/recipes-core/uclibc/uclibc-git/0002-wire-setns-syscall.patch
 create mode 100644 meta/recipes-core/uclibc/uclibc-git/0003-fcntl.h-Define-F_SETPIPE_SZ-and-F_GETPIPE_SZ.patch

-- 
1.9.0

Re: [RFT][PATCH 0/2] Systemd 209

[Saul Wold]

On 02/20/2014 02:18 AM, Khem Raj wrote:
> This patchset upgrades systemd to 209 as can be seen some enahancements
> to uclibc were neeed. As of now. It builds for me on both uclibc/eglibc
> Please test it out in your setups and report back any issues
>
> The following changes since commit 9a98b8bfe13b94d89a9e39cca821f26fdeaeffc3:
>
>    image_types.bbclass: Fix tar IMAGE_CMD to not change directories (2014-02-18 08:37:03 +0000)
>
> are available in the git repository at:
>
>    git://git.openembedded.org/openembedded-core-contrib kraj/systemd-209
>    http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=kraj/systemd-209
>
> Khem Raj (2):
>    uclibc: Add new functionality needed for systemd 209
>    systemd: Upgrade 208 -> 209
>
>   ...-Use-bin-mkdir-instead-of-host-mkdir-path.patch |  31 --
>   ...1-uClibc-doesn-t-implement-pwritev-preadv.patch |  15 +-
>   ...ournal-Add-missing-byte-order-conversions.patch |  60 ----
>   .../journal-file-protect-against-alloca-0.patch    |  23 --
>   ...n-appending-to-journal-file-allocate-larg.patch |  83 -----
>   .../systemd/journald-add-missing-error-check.patch |  16 -
>   .../systemd/systemd/systemd-pam-fix-execvpe.patch  |  22 +-
>   .../systemd/systemd-pam-fix-fallocate.patch        |  24 +-
>   .../systemd/systemd-pam-fix-getty-unit.patch       |  20 +-
>   .../systemd/systemd/systemd-pam-fix-mkostemp.patch | 166 +--------
>   .../systemd/systemd/systemd-pam-fix-msformat.patch | 115 ++++---
>   .../systemd/systemd/uclibc-sysinfo_h.patch         |  19 ++
>   .../systemd/{systemd_208.bb => systemd_209.bb}     |  14 +-
>   meta/recipes-core/uclibc/uclibc-git.inc            |   3 +
>   .../0001-Add-eventfd_read-and-eventfd_write.patch  | 120 +++++++
>   .../uclibc-git/0002-wire-setns-syscall.patch       |  68 ++++
>   ...tl.h-Define-F_SETPIPE_SZ-and-F_GETPIPE_SZ.patch | 377 +++++++++++++++++++++
>   17 files changed, 702 insertions(+), 474 deletions(-)
>   delete mode 100644 meta/recipes-core/systemd/systemd/0001-Use-bin-mkdir-instead-of-host-mkdir-path.patch
>   delete mode 100644 meta/recipes-core/systemd/systemd/journal-Add-missing-byte-order-conversions.patch
>   delete mode 100644 meta/recipes-core/systemd/systemd/journal-file-protect-against-alloca-0.patch
>   delete mode 100644 meta/recipes-core/systemd/systemd/journal-when-appending-to-journal-file-allocate-larg.patch
>   delete mode 100644 meta/recipes-core/systemd/systemd/journald-add-missing-error-check.patch
>   create mode 100644 meta/recipes-core/systemd/systemd/uclibc-sysinfo_h.patch
>   rename meta/recipes-core/systemd/{systemd_208.bb => systemd_209.bb} (96%)
>   create mode 100644 meta/recipes-core/uclibc/uclibc-git/0001-Add-eventfd_read-and-eventfd_write.patch
>   create mode 100644 meta/recipes-core/uclibc/uclibc-git/0002-wire-setns-syscall.patch
>   create mode 100644 meta/recipes-core/uclibc/uclibc-git/0003-fcntl.h-Define-F_SETPIPE_SZ-and-F_GETPIPE_SZ.patch
>

Seems like there might be an issue with ptest when I built this on the 
AB, it might be build vs src dir split?


> NOTE: Running task 5039 of 5679 (ID: 3843, /home/pokybuild/yocto-autobuilder-dev/yocto-slave/nightly-qa-systemd/build/meta/recipes-support/libcroco/libcroco_0.6.8.bb, do_package_write_ipk)
> ERROR: Function failed: do_install_ptest_base (log file is located at /home/pokybuild/yocto-autobuilder-dev/yocto-slave/nightly-qa-systemd/build/build/tmp/work/core2-64-poky-linux/systemd/1_209-r0/temp/log.do_install_ptest_base.24377)
> ERROR: Logfile of failure stored in: /home/pokybuild/yocto-autobuilder-dev/yocto-slave/nightly-qa-systemd/build/build/tmp/work/core2-64-poky-linux/systemd/1_209-r0/temp/log.do_install_ptest_base.24377
> Log data follows:
> | DEBUG: SITE files ['endian-little', 'common-linux', 'common-glibc', 'bit-64', 'x86_64-linux', 'common']
> | DEBUG: Executing shell function do_install_ptest_base
> | cp: cannot stat '/home/pokybuild/yocto-autobuilder-dev/yocto-slave/nightly-qa-systemd/build/build/tmp/work/core2-64-poky-linux/systemd/1_209-r0/build/rules': No such file or directory
> | WARNING: /home/pokybuild/yocto-autobuilder-dev/yocto-slave/nightly-qa-systemd/build/build/tmp/work/core2-64-poky-linux/systemd/1_209-r0/temp/run.do_install_ptest_base.24377:1 exit 1 from
> |   cp -rf /home/pokybuild/yocto-autobuilder-dev/yocto-slave/nightly-qa-systemd/build/build/tmp/work/core2-64-poky-linux/systemd/1_209-r0/build/rules /home/pokybuild/yocto-autobuilder-dev/yocto-slave/nightly-qa-systemd/build/build/tmp/work/core2-64-poky-linux/systemd/1_209-r0/image/usr/lib/systemd/ptest/
> | ERROR: Function failed: do_install_ptest_base (log file is located at /home/pokybuild/yocto-autobuilder-dev/yocto-slave/nightly-qa-systemd/build/build/tmp/work/core2-64-poky-linux/systemd/1_209-r0/temp/log.do_install_ptest_base.24377)
> NOTE: recipe systemd-1_209-r0: task do_install_ptest_base: Failed

Sau!

Re: [RFT][PATCH 0/2] Systemd 209

[Khem Raj]

On Thu, Feb 20, 2014 at 7:48 PM, Saul Wold <sgw@linux.intel.com> wrote:
> Seems like there might be an issue with ptest when I built this on the AB,
> it might be build vs src dir split?

hmmm, could be. but there are more rutime issues as this point, I am
waiting for 210 release which should happen soon

Re: [RFT][PATCH 0/2] Systemd 209

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 3705 bytes --]

On Thu, Feb 20, 2014 at 02:18:56AM -0800, Khem Raj wrote:
> This patchset upgrades systemd to 209 as can be seen some enahancements
> to uclibc were neeed. As of now. It builds for me on both uclibc/eglibc
> Please test it out in your setups and report back any issues
> 
> The following changes since commit 9a98b8bfe13b94d89a9e39cca821f26fdeaeffc3:
> 
>   image_types.bbclass: Fix tar IMAGE_CMD to not change directories (2014-02-18 08:37:03 +0000)
> 
> are available in the git repository at:
> 
>   git://git.openembedded.org/openembedded-core-contrib kraj/systemd-209
>   http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=kraj/systemd-209
> 
> Khem Raj (2):
>   uclibc: Add new functionality needed for systemd 209
>   systemd: Upgrade 208 -> 209

Thanks for upgrade

I've noticed pulseaudio failing to build when it tries to detect old
libsystemd-login lib:

| to the PKG_CONFIG_PATH environment variable
| No package 'libsystemd-login' found
| configure:26263: $? = 1
| configure:26277: result: no
| No package 'libsystemd-login' found
| configure:26310: error: *** Needed systemd support not found

do we want to enable compat libs or update pulseaudio to use new
libsystemd?

>  ...-Use-bin-mkdir-instead-of-host-mkdir-path.patch |  31 --
>  ...1-uClibc-doesn-t-implement-pwritev-preadv.patch |  15 +-
>  ...ournal-Add-missing-byte-order-conversions.patch |  60 ----
>  .../journal-file-protect-against-alloca-0.patch    |  23 --
>  ...n-appending-to-journal-file-allocate-larg.patch |  83 -----
>  .../systemd/journald-add-missing-error-check.patch |  16 -
>  .../systemd/systemd/systemd-pam-fix-execvpe.patch  |  22 +-
>  .../systemd/systemd-pam-fix-fallocate.patch        |  24 +-
>  .../systemd/systemd-pam-fix-getty-unit.patch       |  20 +-
>  .../systemd/systemd/systemd-pam-fix-mkostemp.patch | 166 +--------
>  .../systemd/systemd/systemd-pam-fix-msformat.patch | 115 ++++---
>  .../systemd/systemd/uclibc-sysinfo_h.patch         |  19 ++
>  .../systemd/{systemd_208.bb => systemd_209.bb}     |  14 +-
>  meta/recipes-core/uclibc/uclibc-git.inc            |   3 +
>  .../0001-Add-eventfd_read-and-eventfd_write.patch  | 120 +++++++
>  .../uclibc-git/0002-wire-setns-syscall.patch       |  68 ++++
>  ...tl.h-Define-F_SETPIPE_SZ-and-F_GETPIPE_SZ.patch | 377 +++++++++++++++++++++
>  17 files changed, 702 insertions(+), 474 deletions(-)
>  delete mode 100644 meta/recipes-core/systemd/systemd/0001-Use-bin-mkdir-instead-of-host-mkdir-path.patch
>  delete mode 100644 meta/recipes-core/systemd/systemd/journal-Add-missing-byte-order-conversions.patch
>  delete mode 100644 meta/recipes-core/systemd/systemd/journal-file-protect-against-alloca-0.patch
>  delete mode 100644 meta/recipes-core/systemd/systemd/journal-when-appending-to-journal-file-allocate-larg.patch
>  delete mode 100644 meta/recipes-core/systemd/systemd/journald-add-missing-error-check.patch
>  create mode 100644 meta/recipes-core/systemd/systemd/uclibc-sysinfo_h.patch
>  rename meta/recipes-core/systemd/{systemd_208.bb => systemd_209.bb} (96%)
>  create mode 100644 meta/recipes-core/uclibc/uclibc-git/0001-Add-eventfd_read-and-eventfd_write.patch
>  create mode 100644 meta/recipes-core/uclibc/uclibc-git/0002-wire-setns-syscall.patch
>  create mode 100644 meta/recipes-core/uclibc/uclibc-git/0003-fcntl.h-Define-F_SETPIPE_SZ-and-F_GETPIPE_SZ.patch
> 
> -- 
> 1.9.0
> 
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

Re: [RFT][PATCH 0/2] Systemd 209

[Richard Purdie]

On Fri, 2014-02-21 at 16:16 +0100, Martin Jansa wrote:
> On Thu, Feb 20, 2014 at 02:18:56AM -0800, Khem Raj wrote:
> > This patchset upgrades systemd to 209 as can be seen some enahancements
> > to uclibc were neeed. As of now. It builds for me on both uclibc/eglibc
> > Please test it out in your setups and report back any issues
> > 
> > The following changes since commit 9a98b8bfe13b94d89a9e39cca821f26fdeaeffc3:
> > 
> >   image_types.bbclass: Fix tar IMAGE_CMD to not change directories (2014-02-18 08:37:03 +0000)
> > 
> > are available in the git repository at:
> > 
> >   git://git.openembedded.org/openembedded-core-contrib kraj/systemd-209
> >   http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=kraj/systemd-209
> > 
> > Khem Raj (2):
> >   uclibc: Add new functionality needed for systemd 209
> >   systemd: Upgrade 208 -> 209
> 
> Thanks for upgrade
> 
> I've noticed pulseaudio failing to build when it tries to detect old
> libsystemd-login lib:
> 
> | to the PKG_CONFIG_PATH environment variable
> | No package 'libsystemd-login' found
> | configure:26263: $? = 1
> | configure:26277: result: no
> | No package 'libsystemd-login' found
> | configure:26310: error: *** Needed systemd support not found
> 
> do we want to enable compat libs or update pulseaudio to use new
> libsystemd?

FWIW the git recipe I was using enabled compat-libs...

Cheers,

Richard

Re: [RFT][PATCH 0/2] Systemd 209

[Khem Raj]

On Fri, Feb 21, 2014 at 7:16 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
>
> Thanks for upgrade
>
> I've noticed pulseaudio failing to build when it tries to detect old
> libsystemd-login lib:
>
> | to the PKG_CONFIG_PATH environment variable
> | No package 'libsystemd-login' found
> | configure:26263: $? = 1
> | configure:26277: result: no
> | No package 'libsystemd-login' found
> | configure:26310: error: *** Needed systemd support not found
>
> do we want to enable compat libs or update pulseaudio to use new
> libsystemd?

I think we dont have such a big legacy with systemd yet may be if its
few packages then lets fix the packages instead of enabling legacy
support if we can. in my local tree ( not yet pushed ) I was
explicitly enabling it and disabling kdbus support to see if I can fix
the logind not finding dbus issue
it seems to happen for all architectures

Re: [RFT][PATCH 0/2] Systemd 209

[Khem Raj]

On Fri, Feb 21, 2014 at 7:21 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
>
> FWIW the git recipe I was using enabled compat-libs...

I have tested both combinations. I would suggest that we drop compat-libs
if there are few packages needing it lests fix the packages instead

Re: [RFT][PATCH 0/2] Systemd 209

[Koen Kooi]

Op 21 feb. 2014, om 16:21 heeft Richard Purdie <richard.purdie@linuxfoundation.org> het volgende geschreven:

> On Fri, 2014-02-21 at 16:16 +0100, Martin Jansa wrote:
>> On Thu, Feb 20, 2014 at 02:18:56AM -0800, Khem Raj wrote:
>>> This patchset upgrades systemd to 209 as can be seen some enahancements
>>> to uclibc were neeed. As of now. It builds for me on both uclibc/eglibc
>>> Please test it out in your setups and report back any issues
>>> 
>>> The following changes since commit 9a98b8bfe13b94d89a9e39cca821f26fdeaeffc3:
>>> 
>>>  image_types.bbclass: Fix tar IMAGE_CMD to not change directories (2014-02-18 08:37:03 +0000)
>>> 
>>> are available in the git repository at:
>>> 
>>>  git://git.openembedded.org/openembedded-core-contrib kraj/systemd-209
>>>  http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=kraj/systemd-209
>>> 
>>> Khem Raj (2):
>>>  uclibc: Add new functionality needed for systemd 209
>>>  systemd: Upgrade 208 -> 209
>> 
>> Thanks for upgrade
>> 
>> I've noticed pulseaudio failing to build when it tries to detect old
>> libsystemd-login lib:
>> 
>> | to the PKG_CONFIG_PATH environment variable
>> | No package 'libsystemd-login' found
>> | configure:26263: $? = 1
>> | configure:26277: result: no
>> | No package 'libsystemd-login' found
>> | configure:26310: error: *** Needed systemd support not found
>> 
>> do we want to enable compat libs or update pulseaudio to use new
>> libsystemd?
> 
> FWIW the git recipe I was using enabled compat-libs...

And that needs IFUNC which only x86 toolchains support :(

Re: [RFT][PATCH 0/2] Systemd 209

[Richard Purdie]

On Sat, 2014-02-22 at 08:57 +0100, Koen Kooi wrote:
> Op 21 feb. 2014, om 16:21 heeft Richard Purdie <richard.purdie@linuxfoundation.org> het volgende geschreven:
> 
> > On Fri, 2014-02-21 at 16:16 +0100, Martin Jansa wrote:
> >> On Thu, Feb 20, 2014 at 02:18:56AM -0800, Khem Raj wrote:
> >>> This patchset upgrades systemd to 209 as can be seen some enahancements
> >>> to uclibc were neeed. As of now. It builds for me on both uclibc/eglibc
> >>> Please test it out in your setups and report back any issues
> >>> 
> >>> The following changes since commit 9a98b8bfe13b94d89a9e39cca821f26fdeaeffc3:
> >>> 
> >>>  image_types.bbclass: Fix tar IMAGE_CMD to not change directories (2014-02-18 08:37:03 +0000)
> >>> 
> >>> are available in the git repository at:
> >>> 
> >>>  git://git.openembedded.org/openembedded-core-contrib kraj/systemd-209
> >>>  http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=kraj/systemd-209
> >>> 
> >>> Khem Raj (2):
> >>>  uclibc: Add new functionality needed for systemd 209
> >>>  systemd: Upgrade 208 -> 209
> >> 
> >> Thanks for upgrade
> >> 
> >> I've noticed pulseaudio failing to build when it tries to detect old
> >> libsystemd-login lib:
> >> 
> >> | to the PKG_CONFIG_PATH environment variable
> >> | No package 'libsystemd-login' found
> >> | configure:26263: $? = 1
> >> | configure:26277: result: no
> >> | No package 'libsystemd-login' found
> >> | configure:26310: error: *** Needed systemd support not found
> >> 
> >> do we want to enable compat libs or update pulseaudio to use new
> >> libsystemd?
> > 
> > FWIW the git recipe I was using enabled compat-libs...
> 
> And that needs IFUNC which only x86 toolchains support :(

I did run test builds of the git version on the autobuilders and didn't
see any failures? 

Cheers,

Richard

Re: [RFT][PATCH 0/2] Systemd 209

[Khem Raj]

[-- Attachment #1: Type: text/plain, Size: 2447 bytes --]


On Feb 22, 2014, at 12:49 AM, Richard Purdie <richard.purdie@linuxfoundation.org> wrote:

> On Sat, 2014-02-22 at 08:57 +0100, Koen Kooi wrote:
>> Op 21 feb. 2014, om 16:21 heeft Richard Purdie <richard.purdie@linuxfoundation.org> het volgende geschreven:
>> 
>>> On Fri, 2014-02-21 at 16:16 +0100, Martin Jansa wrote:
>>>> On Thu, Feb 20, 2014 at 02:18:56AM -0800, Khem Raj wrote:
>>>>> This patchset upgrades systemd to 209 as can be seen some enahancements
>>>>> to uclibc were neeed. As of now. It builds for me on both uclibc/eglibc
>>>>> Please test it out in your setups and report back any issues
>>>>> 
>>>>> The following changes since commit 9a98b8bfe13b94d89a9e39cca821f26fdeaeffc3:
>>>>> 
>>>>> image_types.bbclass: Fix tar IMAGE_CMD to not change directories (2014-02-18 08:37:03 +0000)
>>>>> 
>>>>> are available in the git repository at:
>>>>> 
>>>>> git://git.openembedded.org/openembedded-core-contrib kraj/systemd-209
>>>>> http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=kraj/systemd-209
>>>>> 
>>>>> Khem Raj (2):
>>>>> uclibc: Add new functionality needed for systemd 209
>>>>> systemd: Upgrade 208 -> 209
>>>> 
>>>> Thanks for upgrade
>>>> 
>>>> I've noticed pulseaudio failing to build when it tries to detect old
>>>> libsystemd-login lib:
>>>> 
>>>> | to the PKG_CONFIG_PATH environment variable
>>>> | No package 'libsystemd-login' found
>>>> | configure:26263: $? = 1
>>>> | configure:26277: result: no
>>>> | No package 'libsystemd-login' found
>>>> | configure:26310: error: *** Needed systemd support not found
>>>> 
>>>> do we want to enable compat libs or update pulseaudio to use new
>>>> libsystemd?
>>> 
>>> FWIW the git recipe I was using enabled compat-libs...
>> 
>> And that needs IFUNC which only x86 toolchains support :(
> 
> I did run test builds of the git version on the autobuilders and didn't
> see any failures? 


builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
across all architectures. I have pushed my latest rework to pull branch


> 
> Cheers,
> 
> Richard
> 
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core


[-- Attachment #2: Message signed with OpenPGP using GPGMail --]
[-- Type: application/pgp-signature, Size: 211 bytes --]

Re: [RFT][PATCH 0/2] Systemd 209

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1597 bytes --]

On Fri, Feb 21, 2014 at 08:32:49AM -0800, Khem Raj wrote:
> On Fri, Feb 21, 2014 at 7:16 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> >
> > Thanks for upgrade
> >
> > I've noticed pulseaudio failing to build when it tries to detect old
> > libsystemd-login lib:
> >
> > | to the PKG_CONFIG_PATH environment variable
> > | No package 'libsystemd-login' found
> > | configure:26263: $? = 1
> > | configure:26277: result: no
> > | No package 'libsystemd-login' found
> > | configure:26310: error: *** Needed systemd support not found
> >
> > do we want to enable compat libs or update pulseaudio to use new
> > libsystemd?
> 
> I think we dont have such a big legacy with systemd yet may be if its
> few packages then lets fix the packages instead of enabling legacy
> support if we can. in my local tree ( not yet pushed ) I was
> explicitly enabling it and disabling kdbus support to see if I can fix
> the logind not finding dbus issue
> it seems to happen for all architectures

Agreed,

I've found another one: polkit:

| checking for LIBSYSTEMD_LOGIN... no
| configure: error: libsystemd-login support requested but libsystemd-login library not found
| Configure failed. The contents of all config.log files follows to aid debugging
| /home/jenkins/oe/shr-core-branches/shr-core/tmp-eglibc/work/armv5te-oe-linux-gnueabi/polkit/0.111-r0/polkit-0.111/config.log
| This file contains any messages produced by compilers while
| running configure, to aid debugging if configure makes a mistake.

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

[meta-oe][PATCH] polkit: make it compatible with systemd-209

[Martin Jansa]

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
---
 ...c-Check-only-for-libsystemd-not-libsystem.patch | 39 ++++++++++++++++++++++
 meta-oe/recipes-extended/polkit/polkit_0.111.bb    |  1 +
 2 files changed, 40 insertions(+)
 create mode 100644 meta-oe/recipes-extended/polkit/polkit/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch

diff --git a/meta-oe/recipes-extended/polkit/polkit/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch b/meta-oe/recipes-extended/polkit/polkit/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch
new file mode 100644
index 0000000..b9334a8
--- /dev/null
+++ b/meta-oe/recipes-extended/polkit/polkit/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch
@@ -0,0 +1,39 @@
+From 56cadbc45c5ef98a7b775638d2109c4a09a58c06 Mon Sep 17 00:00:00 2001
+From: Martin Jansa <Martin.Jansa@gmail.com>
+Date: Sat, 22 Feb 2014 18:09:57 +0100
+Subject: [PATCH] configure.ac: Check only for libsystemd not libsystemd-login
+
+* they were merged into libsystemd in systemd-209
+
+Upstream-Status: Pending (it would need to be conditional on systemd version for upstream to accept this)
+
+Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
+---
+ configure.ac | 4 ++--
+ 1 file changed, 2 insertions(+), 2 deletions(-)
+
+diff --git a/configure.ac b/configure.ac
+index a7b0148..64d513d 100644
+--- a/configure.ac
++++ b/configure.ac
+@@ -196,7 +196,7 @@ AC_ARG_ENABLE([libsystemd-login],
+               [enable_libsystemd_login=auto])
+ if test "$enable_libsystemd_login" != "no"; then
+   PKG_CHECK_MODULES(LIBSYSTEMD_LOGIN,
+-                    [libsystemd-login],
++                    [libsystemd],
+                     have_libsystemd_login=yes,
+                     have_libsystemd_login=no)
+   if test "$have_libsystemd_login" = "yes"; then
+@@ -223,7 +223,7 @@ AS_IF([test "x$cross_compiling" != "xyes" ], [
+       AS_IF([test "$enable_libsystemd_login" = "no" ], [
+         AC_MSG_WARN([ConsoleKit requested but system appears to use systemd])
+       ], [
+-        AC_MSG_ERROR([ConsoleKit autoconfigured, but systemd is in use (missing libsystemd-login pkg-config?)])
++        AC_MSG_ERROR([ConsoleKit autoconfigured, but systemd is in use (missing libsystemd pkg-config?)])
+       ])
+     ])
+   ])
+-- 
+1.8.5.3
+
diff --git a/meta-oe/recipes-extended/polkit/polkit_0.111.bb b/meta-oe/recipes-extended/polkit/polkit_0.111.bb
index 33101cf..4614387 100644
--- a/meta-oe/recipes-extended/polkit/polkit_0.111.bb
+++ b/meta-oe/recipes-extended/polkit/polkit_0.111.bb
@@ -20,6 +20,7 @@ PACKAGECONFIG[consolekit] = ",,,consolekit"
 PAM_SRC_URI = "file://polkit-1_pam.patch"
 SRC_URI = "http://www.freedesktop.org/software/polkit/releases/polkit-${PV}.tar.gz \
            ${@base_contains('DISTRO_FEATURES', 'pam', '${PAM_SRC_URI}', '', d)} \
+           file://0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch \
 "
 
 SRC_URI[md5sum] = "81b116edf986d8e13502929a171f4e0d"
-- 
1.8.5.3

[PATCH] pulseaudio: Make it compatible with systemd-209

[Martin Jansa]

Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
---
 ...c-Check-only-for-libsystemd-not-libsystem.patch | 30 ++++++++++++++++++++++
 .../pulseaudio/pulseaudio_4.0.bb                   |  1 +
 2 files changed, 31 insertions(+)
 create mode 100644 meta/recipes-multimedia/pulseaudio/pulseaudio/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch

diff --git a/meta/recipes-multimedia/pulseaudio/pulseaudio/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch b/meta/recipes-multimedia/pulseaudio/pulseaudio/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch
new file mode 100644
index 0000000..467cd2c
--- /dev/null
+++ b/meta/recipes-multimedia/pulseaudio/pulseaudio/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch
@@ -0,0 +1,30 @@
+From 002b16f0f2176b4c685e210e335bf69c02563ede Mon Sep 17 00:00:00 2001
+From: Martin Jansa <Martin.Jansa@gmail.com>
+Date: Sat, 22 Feb 2014 18:03:10 +0100
+Subject: [PATCH] configure.ac: Check only for libsystemd not libsystemd-login
+
+* they were merged into libsystemd in systemd-209
+
+Upstream-Status: Pending (it would need to be conditional on systemd version for upstream to accept this)
+
+Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
+---
+ configure.ac | 2 +-
+ 1 file changed, 1 insertion(+), 1 deletion(-)
+
+diff --git a/configure.ac b/configure.ac
+index 388fae2..fff7a83 100644
+--- a/configure.ac
++++ b/configure.ac
+@@ -1160,7 +1160,7 @@ AC_ARG_ENABLE([systemd],
+     AS_HELP_STRING([--disable-systemd],[Disable optional systemd support]))
+ 
+ AS_IF([test "x$enable_systemd" != "xno"],
+-    [PKG_CHECK_MODULES(SYSTEMD, [ libsystemd-login ], HAVE_SYSTEMD=1, HAVE_SYSTEMD=0)],
++    [PKG_CHECK_MODULES(SYSTEMD, [ libsystemd ], HAVE_SYSTEMD=1, HAVE_SYSTEMD=0)],
+     HAVE_SYSTEMD=0)
+ 
+ AS_IF([test "x$enable_systemd" = "xyes" && test "x$HAVE_SYSTEMD" = "x0"],
+-- 
+1.8.5.3
+
diff --git a/meta/recipes-multimedia/pulseaudio/pulseaudio_4.0.bb b/meta/recipes-multimedia/pulseaudio/pulseaudio_4.0.bb
index b419c54..53dd494 100644
--- a/meta/recipes-multimedia/pulseaudio/pulseaudio_4.0.bb
+++ b/meta/recipes-multimedia/pulseaudio/pulseaudio_4.0.bb
@@ -1,6 +1,7 @@
 require pulseaudio.inc
 
 SRC_URI = "http://freedesktop.org/software/pulseaudio/releases/pulseaudio-${PV}.tar.xz \
+           file://0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch \
            file://volatiles.04_pulse"
 
 SRC_URI[md5sum] = "591f211db2790a7e4d222f2dc6858db3"
-- 
1.8.5.3

Re: [PATCH] pulseaudio: Make it compatible with systemd-209

[Khem Raj]

On Sat, Feb 22, 2014 at 9:29 AM, Martin Jansa <martin.jansa@gmail.com> wrote:
> Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
> ---
>  ...c-Check-only-for-libsystemd-not-libsystem.patch | 30 ++++++++++++++++++++++
>  .../pulseaudio/pulseaudio_4.0.bb                   |  1 +
>  2 files changed, 31 insertions(+)
>  create mode 100644 meta/recipes-multimedia/pulseaudio/pulseaudio/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch
>
> diff --git a/meta/recipes-multimedia/pulseaudio/pulseaudio/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch b/meta/recipes-multimedia/pulseaudio/pulseaudio/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch
> new file mode 100644
> index 0000000..467cd2c
> --- /dev/null
> +++ b/meta/recipes-multimedia/pulseaudio/pulseaudio/0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch
> @@ -0,0 +1,30 @@
> +From 002b16f0f2176b4c685e210e335bf69c02563ede Mon Sep 17 00:00:00 2001
> +From: Martin Jansa <Martin.Jansa@gmail.com>
> +Date: Sat, 22 Feb 2014 18:03:10 +0100
> +Subject: [PATCH] configure.ac: Check only for libsystemd not libsystemd-login
> +
> +* they were merged into libsystemd in systemd-209
> +
> +Upstream-Status: Pending (it would need to be conditional on systemd version for upstream to accept this)
> +
> +Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
> +---

while we fight to make it work in OE you can in parallel stage/get
reviews on these patches in respective upstream projects.


> + configure.ac | 2 +-
> + 1 file changed, 1 insertion(+), 1 deletion(-)
> +
> +diff --git a/configure.ac b/configure.ac
> +index 388fae2..fff7a83 100644
> +--- a/configure.ac
> ++++ b/configure.ac
> +@@ -1160,7 +1160,7 @@ AC_ARG_ENABLE([systemd],
> +     AS_HELP_STRING([--disable-systemd],[Disable optional systemd support]))
> +
> + AS_IF([test "x$enable_systemd" != "xno"],
> +-    [PKG_CHECK_MODULES(SYSTEMD, [ libsystemd-login ], HAVE_SYSTEMD=1, HAVE_SYSTEMD=0)],
> ++    [PKG_CHECK_MODULES(SYSTEMD, [ libsystemd ], HAVE_SYSTEMD=1, HAVE_SYSTEMD=0)],
> +     HAVE_SYSTEMD=0)
> +
> + AS_IF([test "x$enable_systemd" = "xyes" && test "x$HAVE_SYSTEMD" = "x0"],
> +--
> +1.8.5.3
> +
> diff --git a/meta/recipes-multimedia/pulseaudio/pulseaudio_4.0.bb b/meta/recipes-multimedia/pulseaudio/pulseaudio_4.0.bb
> index b419c54..53dd494 100644
> --- a/meta/recipes-multimedia/pulseaudio/pulseaudio_4.0.bb
> +++ b/meta/recipes-multimedia/pulseaudio/pulseaudio_4.0.bb
> @@ -1,6 +1,7 @@
>  require pulseaudio.inc
>
>  SRC_URI = "http://freedesktop.org/software/pulseaudio/releases/pulseaudio-${PV}.tar.xz \
> +           file://0001-configure.ac-Check-only-for-libsystemd-not-libsystem.patch \
>             file://volatiles.04_pulse"
>
>  SRC_URI[md5sum] = "591f211db2790a7e4d222f2dc6858db3"
> --
> 1.8.5.3
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core

Re: [RFT][PATCH 0/2] Systemd 209

[Phil Blundell]

On Sat, 2014-02-22 at 01:31 -0800, Khem Raj wrote:
> builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
> architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
> across all architectures. I have pushed my latest rework to pull branch

What is the problem with ifunc on non-x86?  As far as I know the
compiler, assembler, linker and ld.so for all reasonable architectures
support that nowadays (though prelink support is a bit patchy).  GCC
didn't have ifuncs enabled for some architectures until quite recently
but if that was the problem then you'd see compile-time failures.  So,
if it compiles then I can't think of any reason why it oughtn't to run.
Is this a uclibc thing?

p.

Re: [RFT][PATCH 0/2] Systemd 209

[Khem Raj]

[-- Attachment #1: Type: text/plain, Size: 1511 bytes --]


On Feb 22, 2014, at 10:12 AM, Phil Blundell <pb@pbcl.net> wrote:

> On Sat, 2014-02-22 at 01:31 -0800, Khem Raj wrote:
>> builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
>> architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
>> across all architectures. I have pushed my latest rework to pull branch
> 
> What is the problem with ifunc on non-x86?  

they are not enabled across all architectures, well specifically, its not enabled on arm and seems also ppc is missing it.

> As far as I know the
> compiler, assembler, linker and ld.so for all reasonable architectures
> support that nowadays (though prelink support is a bit patchy).  GCC
> didn't have ifuncs enabled for some architectures until quite recently
> but if that was the problem then you'd see compile-time failures.  So,
> if it compiles then I can't think of any reason why it oughtn't to run.

systemd is using ifuncs to patch up the library migration it has done in 209
so it only shows up when we enable compat lib support in systemd, and if we patch the apps to forward port to systemd 209 then we don’t need to enable compat lib support and
it could be left alone.

if we want to enable compat libs/ifunc then we can think of adding --enable-gnu-indirect-function
to gcc configure but then it may not work on all architectures that we are interested in.


> Is this a uclibc thing?

no

> 
> p.
> 
> 


[-- Attachment #2: Message signed with OpenPGP using GPGMail --]
[-- Type: application/pgp-signature, Size: 211 bytes --]

Re: [RFT][PATCH 0/2] Systemd 209

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 1259 bytes --]

On Sat, Feb 22, 2014 at 06:12:10PM +0000, Phil Blundell wrote:
> On Sat, 2014-02-22 at 01:31 -0800, Khem Raj wrote:
> > builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
> > architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
> > across all architectures. I have pushed my latest rework to pull branch
> 
> What is the problem with ifunc on non-x86?  As far as I know the
> compiler, assembler, linker and ld.so for all reasonable architectures
> support that nowadays (though prelink support is a bit patchy).  GCC
> didn't have ifuncs enabled for some architectures until quite recently
> but if that was the problem then you'd see compile-time failures.  So,
> if it compiles then I can't think of any reason why it oughtn't to run.
> Is this a uclibc thing?

I'm seeing lots of compile errors like this:

| arm-oe-linux-gnueabi-libtool: link: arm-oe-linux-gnueabi-ranlib .libs/libsystemd-units.a
| libsystemd-id128.c:6:6: error: ifunc is not supported in this configuration
|  void sd_id128_to_string(void) __attribute__((ifunc("resolve_sd_id128_to_string")));
|       ^

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

Re: [RFT][PATCH 0/2] Systemd 209

[Khem Raj]

On Sat, Feb 22, 2014 at 12:47 PM, Martin Jansa <martin.jansa@gmail.com> wrote:
> On Sat, Feb 22, 2014 at 06:12:10PM +0000, Phil Blundell wrote:
>> On Sat, 2014-02-22 at 01:31 -0800, Khem Raj wrote:
>> > builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
>> > architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
>> > across all architectures. I have pushed my latest rework to pull branch
>>
>> What is the problem with ifunc on non-x86?  As far as I know the
>> compiler, assembler, linker and ld.so for all reasonable architectures
>> support that nowadays (though prelink support is a bit patchy).  GCC
>> didn't have ifuncs enabled for some architectures until quite recently
>> but if that was the problem then you'd see compile-time failures.  So,
>> if it compiles then I can't think of any reason why it oughtn't to run.
>> Is this a uclibc thing?
>
> I'm seeing lots of compile errors like this:
>
> | arm-oe-linux-gnueabi-libtool: link: arm-oe-linux-gnueabi-ranlib .libs/libsystemd-units.a
> | libsystemd-id128.c:6:6: error: ifunc is not supported in this configuration
> |  void sd_id128_to_string(void) __attribute__((ifunc("resolve_sd_id128_to_string")));
> |       ^
>

this is exact problem I mentioned in last email. Try to remove
--enable-compat-libs in systemd recipe.

> --
> Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

Re: [RFT][PATCH 0/2] Systemd 209

[Koen Kooi]

Op 22 feb. 2014, om 20:52 heeft Khem Raj <raj.khem@gmail.com> het volgende geschreven:

> 
> On Feb 22, 2014, at 10:12 AM, Phil Blundell <pb@pbcl.net> wrote:
> 
>> On Sat, 2014-02-22 at 01:31 -0800, Khem Raj wrote:
>>> builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
>>> architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
>>> across all architectures. I have pushed my latest rework to pull branch
>> 
>> What is the problem with ifunc on non-x86?  
> 
> they are not enabled across all architectures, well specifically, its not enabled on arm and seems also ppc is missing it.
> 
>> As far as I know the
>> compiler, assembler, linker and ld.so for all reasonable architectures
>> support that nowadays (though prelink support is a bit patchy).  GCC
>> didn't have ifuncs enabled for some architectures until quite recently
>> but if that was the problem then you'd see compile-time failures.  So,
>> if it compiles then I can't think of any reason why it oughtn't to run.
> 
> systemd is using ifuncs to patch up the library migration it has done in 209
> so it only shows up when we enable compat lib support in systemd, and if we patch the apps to forward port to systemd 209 then we don’t need to enable compat lib support and
> it could be left alone.

Or we can cherry-pick http://cgit.freedesktop.org/systemd/systemd/commit/?id=03e37dd767e52908f30783d9b4c09fb6a4e865c7 :)

> 
> if we want to enable compat libs/ifunc then we can think of adding --enable-gnu-indirect-function
> to gcc configure but then it may not work on all architectures that we are interested in.
> 
> 
>> Is this a uclibc thing?
> 
> no
> 
>> 
>> p.
>> 
>> 
> 

Re: [RFT][PATCH 0/2] Systemd 209

[Martin Jansa]

[-- Attachment #1: Type: text/plain, Size: 2296 bytes --]

On Sun, Feb 23, 2014 at 08:36:56AM +0100, Koen Kooi wrote:
> 
> Op 22 feb. 2014, om 20:52 heeft Khem Raj <raj.khem@gmail.com> het volgende geschreven:
> 
> > 
> > On Feb 22, 2014, at 10:12 AM, Phil Blundell <pb@pbcl.net> wrote:
> > 
> >> On Sat, 2014-02-22 at 01:31 -0800, Khem Raj wrote:
> >>> builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
> >>> architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
> >>> across all architectures. I have pushed my latest rework to pull branch
> >> 
> >> What is the problem with ifunc on non-x86?  
> > 
> > they are not enabled across all architectures, well specifically, its not enabled on arm and seems also ppc is missing it.
> > 
> >> As far as I know the
> >> compiler, assembler, linker and ld.so for all reasonable architectures
> >> support that nowadays (though prelink support is a bit patchy).  GCC
> >> didn't have ifuncs enabled for some architectures until quite recently
> >> but if that was the problem then you'd see compile-time failures.  So,
> >> if it compiles then I can't think of any reason why it oughtn't to run.
> > 
> > systemd is using ifuncs to patch up the library migration it has done in 209
> > so it only shows up when we enable compat lib support in systemd, and if we patch the apps to forward port to systemd 209 then we don’t need to enable compat lib support and
> > it could be left alone.
> 
> Or we can cherry-pick http://cgit.freedesktop.org/systemd/systemd/commit/?id=03e37dd767e52908f30783d9b4c09fb6a4e865c7 :)

And 210 is out :)

FWIW: My world builds revealed only networkmanager failing to detect
libsystemd.

> > if we want to enable compat libs/ifunc then we can think of adding --enable-gnu-indirect-function
> > to gcc configure but then it may not work on all architectures that we are interested in.
> > 
> > 
> >> Is this a uclibc thing?
> > 
> > no
> > 
> >> 
> >> p.
> >> 
> >> 
> > 
> 
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core

-- 
Martin 'JaMa' Jansa     jabber: Martin.Jansa@gmail.com

[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 205 bytes --]

Re: [RFT][PATCH 0/2] Systemd 209

[Khem Raj]

[-- Attachment #1: Type: text/plain, Size: 897 bytes --]

>> 
>> I did run test builds of the git version on the autobuilders and didn't
>> see any failures? 
> 
> 
> builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
> architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
> across all architectures. I have pushed my latest rework to pull branch

I have now refreshed my contrib tree

http://git.openembedded.org/openembedded-core-contrib/log/?h=kraj/systemd-210

I have documented additional change you might need in your distro if you are not
using OE-Core’s fs-perms.txt, systemd now needs /var/run to be symlinked to /run
or else it won’t be able to talk to d-bus and kdbus is too new to enable yet.

Please try this patch, it now builds and boots fine on x86 and arm here. I am also
testing out other architecture in meantime

-Khem

[-- Attachment #2: Message signed with OpenPGP using GPGMail --]
[-- Type: application/pgp-signature, Size: 211 bytes --]

Re: [RFT][PATCH 0/2] Systemd 209

[Saul Wold]

On 02/26/2014 09:14 AM, Khem Raj wrote:
>>>
>>> I did run test builds of the git version on the autobuilders and didn't
>>> see any failures?
>>
>>
>> builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
>> architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
>> across all architectures. I have pushed my latest rework to pull branch
>
> I have now refreshed my contrib tree
>
> http://git.openembedded.org/openembedded-core-contrib/log/?h=kraj/systemd-210
>
> I have documented additional change you might need in your distro if you are not
> using OE-Core’s fs-perms.txt, systemd now needs /var/run to be symlinked to /run
> or else it won’t be able to talk to d-bus and kdbus is too new to enable yet.
>
> Please try this patch, it now builds and boots fine on x86 and arm here. I am also
> testing out other architecture in meantime
>

Khem,

Did you test with ptest enabled in DISTRO_FEATURES?  I just built it 
locally and got the following error:

> | cp: cannot stat '/home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/build/rules': No such file or directory
> | WARNING: /home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/temp/run.do_install_ptest_base.9533:1 exit 1 from
> |   cp -rf /home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/build/rules /home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/image/usr/lib/systemd/ptest/
> | ERROR: Function failed: do_install_ptest_base (log file is located at /home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/temp/log.do_install_ptest_base.9533)


Sau!


> -Khem
>
>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.openembedded.org/mailman/listinfo/openembedded-core
>

Re: [RFT][PATCH 0/2] Systemd 209

[Khem Raj]

[-- Attachment #1: Type: text/plain, Size: 2226 bytes --]


On Feb 26, 2014, at 1:00 AM, Saul Wold <sgw@linux.intel.com> wrote:

> On 02/26/2014 09:14 AM, Khem Raj wrote:
>>>> 
>>>> I did run test builds of the git version on the autobuilders and didn't
>>>> see any failures?
>>> 
>>> 
>>> builds might work but they won’t run and yes so far IFUNC support is needed so it won’t work on non-x86
>>> architectures. One common runtime problem I am seeing is logind not being able to talk to dbus
>>> across all architectures. I have pushed my latest rework to pull branch
>> 
>> I have now refreshed my contrib tree
>> 
>> http://git.openembedded.org/openembedded-core-contrib/log/?h=kraj/systemd-210
>> 
>> I have documented additional change you might need in your distro if you are not
>> using OE-Core’s fs-perms.txt, systemd now needs /var/run to be symlinked to /run
>> or else it won’t be able to talk to d-bus and kdbus is too new to enable yet.
>> 
>> Please try this patch, it now builds and boots fine on x86 and arm here. I am also
>> testing out other architecture in meantime
>> 
> 
> Khem,
> 
> Did you test with ptest enabled in DISTRO_FEATURES?  

no


> I just built it locally and got the following error:
> 
>> | cp: cannot stat '/home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/build/rules': No such file or directory
>> | WARNING: /home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/temp/run.do_install_ptest_base.9533:1 exit 1 from
>> |   cp -rf /home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/build/rules /home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/image/usr/lib/systemd/ptest/
>> | ERROR: Function failed: do_install_ptest_base (log file is located at /home/sgw/yocto/builds/world/tmp/work/core2-64-poky-linux/systemd/1_210-r0/temp/log.do_install_ptest_base.9533)
> 


this is because you might have S != B, anyway I have fixed this and refreshed the branch

> 
> Sau!
> 
> 
>> -Khem
>> 
>> 
>> 
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.openembedded.org/mailman/listinfo/openembedded-core
>> 


[-- Attachment #2: Message signed with OpenPGP using GPGMail --]
[-- Type: application/pgp-signature, Size: 211 bytes --]