[OE-core] [PATCH 2/2] dbus: move libdbus-1.so* to base_libdir

jhuang0 jackie.huang at windriver.com
Thu Jun 27 03:36:27 UTC 2013



On 6/21/2013 8:36 PM, Richard Purdie wrote:
> On Fri, 2013-06-21 at 10:02 +0800, jhuang0 wrote:
>>
>> On 6/20/2013 7:46 PM, Saul Wold wrote:
>>> On 06/20/2013 04:14 AM, Richard Purdie wrote:
>>>> On Thu, 2013-06-20 at 11:39 +0100, Burton, Ross wrote:
>>>>> On 20 June 2013 11:36,  <jackie.huang at windriver.com> wrote:
>>>>>> From: Jackie Huang <jackie.huang at windriver.com>
>>>>>>
>>>>>> move libdbus-1.so* to base_libdir to kill a warning:
>>>>>>
>>>>>>       WARNING: QA Issue: lib32-consolekit: /lib/security/
>>>>>>       pam_ck_connector.so, installed in the base_prefix,
>>>>>>       requires a shared library under exec_prefix (/usr):
>>>>>>       libdbus-1.so.3 => /usr/lib/libdbus-1.so.3 (0xdead3000)
>>>>>>
>>>>>> Signed-off-by: Roy.Li <rongqing.li at windriver.com>
>>>>>> Signed-off-by: Jackie Huang <jackie.huang at windriver.com>
>>>>>
>>>>> More and more and more libraries move.  Do we *really* need to support
>>>>> split-usr?  Have you audited all of the udev helpers yet?
>>>>
>>>> I have to admit I'm not happy with the piece by piece approach this is
>>>> taking. We're going to end up with nearly everything moving to
>>>> base_libdir at this rate. If we're going to do this, I want someone to
>>>> come up with a definitive list of what needs to move before any more of
>>>> these merge.
>>>>
>>> And this should be checked with various DISTRO_FEATURES such as pam
>>> (enabled/disabled) and systemd (enabled/disabled), not just one
>>> variation, please be sure to list what your testing matrix is.
>>
>> I tested with DISTRO_FEATURES 'pam' (enabled/disabled) and 'systemd'
>> (enabled/disabled) and WARN_QA = "unsafe-references-in-binaries
>> unsafe-references-in-scripts", and I have sent a v3. If you are agree to
>> merge, please merge the v3 one, thanks!
>
> As I said, I'm not taking any more of these moves until there is a well
> thought out analysis/plan. Sorry.

Hi Richard/Saul,

We sent these since we used bbappend approach to fix issues in our layer 
and now we want to minimize the usage of bbappend files, I admit that 
the piece by piece approach is not good so I undertand if it is not 
taken, we will analyze the issue and come out a plan sometime later.

Thanks,
Jackie

>
> Cheers,
>
> Richard
>
>

-- 
Jackie Huang
WIND RIVER | China Development Center
MSN:jackielily at hotmail.com
Tel: +86 8477 8594
Mobile: +86 138 1027 4745



More information about the Openembedded-core mailing list