[OE-core] [PATCH] gcc-4.6: Let G++ relocate gxx-include-dir when using --sysroot option

Eric Bénard eric at eukrea.com
Wed Dec 21 14:23:45 UTC 2011


Hi Khem,

Le 20/12/2011 20:59, Khem Raj a écrit :
> On Mon, Dec 19, 2011 at 10:36 PM, Eric Bénard<eric at eukrea.com>  wrote:
>> Hi, Khem,
>>
>> Le 14/12/2011 19:38, Khem Raj a écrit :
>>
>>> On Wed, Dec 14, 2011 at 10:08 AM, Ulf Samuelsson
>>> <openembedded-core at emagii.com>    wrote:
>>>>>
>>>>>
>>>>>
>>>> Not familiar with gcc internals, but just want to reiterate the
>>>> point that the problem occured (the one I saw) only when I had
>>>> BB_NUMBER_THREADS at a high value, so WHEN things are done seems to
>>>> be important.
>>>
>>>
>> Seems more complicated than this. A customeer reproduced it with
>> BB_THREAD=2 and PARALLEL=j3 ... (I never reproduced in with BB_THREAD<
>> 5)
>>
>
> Can you see when gcc-runtime was finished building and staging w.r.t to the
> failing app
>
rm_work on gcc-runtime was done far before do_configure on the failing app
(mysql in the present case).

Eric




More information about the Openembedded-core mailing list