[OE-core] Reproducible build problem with BB_NUMBER_THREADS=8

[Daniel Lazzari]

>Date: Tue, 15 Nov 2011 15:32:36 +0100
>From: Eric B?nard <eric at eukrea.com>
>Subject: Re: [OE-core] Reproducible build problem with
>	BB_NUMBER_THREADS=8
>To: Patches and discussions about the oe-core layer
>	<openembedded-core at lists.openembedded.org>
>Message-ID: <4EC27804.7060902 at eukrea.com>
>Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
>Le 15/11/2011 12:03, Eric B?nard a ?crit :
>> - log2.txt (plain oe-core + angstrom setup to remove the doubt of my
>overlay
>> being the source of the problem) :
>> directfb fails with
>> mkdgifft.cpp:64:16: fatal error: list: No such file or directory
>> the file exists :
>> build/tmp-angstrom_2010_x-
>eglibc/sysroots/beagleboard/usr/include/c++/list
>> here again, it seems that include/c++ is not in the header search path
>> => I reduced BB_NUMBER_THREADS to 1 and launched again bitbake (build
>actually
>> in progress).
>>
>this build also crashed a few hours later on gnutls with :
>| ./includes/gnutls/gnutlsxx.h:4:21: fatal error: exception: No such
>file or
>directory
>
>Here again the only solution is to remove tmp and start again with a
>reduced
>BB_NUMBER_THREADS.
>
>Eric
>

I actually just ran into this very problem late last week. I have narrowed it down to a problem with one of the gcc recipes (gcc-cross, gcc-cross-initial, gcc-cross-intermediate, gcc-runtime) being built. I seem to get the problem when my PARALLEL_MAKE > 2. If I run into the problem, dial back PARALLEL_MAKE to 2, cleanall the above 4 recipes, and then continue, the problem seems to go away. I don't really have any extra cycles to continue my investigation, but hopefully that helps put someone on the right track.