From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from 93-97-173-237.zone5.bethere.co.uk ([93.97.173.237] helo=tim.rpsys.net) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1QNC96-0008HQ-OS for openembedded-core@lists.openembedded.org; Fri, 20 May 2011 00:56:45 +0200 Received: from localhost (localhost [127.0.0.1]) by tim.rpsys.net (8.13.6/8.13.8) with ESMTP id p4JMrnDX014441 for ; Thu, 19 May 2011 23:53:49 +0100 Received: from tim.rpsys.net ([127.0.0.1]) by localhost (tim.rpsys.net [127.0.0.1]) (amavisd-new, port 10024) with LMTP id 14260-03 for ; Thu, 19 May 2011 23:53:45 +0100 (BST) Received: from [192.168.3.10] ([192.168.3.10]) (authenticated bits=0) by tim.rpsys.net (8.13.6/8.13.8) with ESMTP id p4JMrf8t014435 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for ; Thu, 19 May 2011 23:53:41 +0100 From: Richard Purdie To: Patches and discussions about the oe-core layer In-Reply-To: <4DD42244.60806@gravedo.de> References: <43EE53C2-2A83-4F54-B8CF-F0614AB390B0@dominion.thruhere.net> <1303126125.5518.118.camel@rex> <0E9C83CD-3E86-46B3-B749-6FAFB94E4BAF@dominion.thruhere.net> <4DAC2BC5.4020107@mlbassoc.com> <4DD42244.60806@gravedo.de> Date: Thu, 19 May 2011 23:53:37 +0100 Message-ID: <1305845617.3424.570.camel@rex> Mime-Version: 1.0 X-Mailer: Evolution 2.32.2 X-Virus-Scanned: amavisd-new at rpsys.net Subject: Re: Is BBCLASSEXTEND in .bbappend supposed to work? X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list Reply-To: Patches and discussions about the oe-core layer List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 19 May 2011 22:56:45 -0000 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit On Wed, 2011-05-18 at 21:47 +0200, Simon Busch wrote: > On 18.04.2011 14:17, Gary Thomas wrote: > > On 04/18/2011 05:34 AM, Koen Kooi wrote: > >> > >> Op 18 apr 2011, om 13:28 heeft Richard Purdie het volgende geschreven: > >> > >>> On Mon, 2011-04-18 at 12:55 +0200, Koen Kooi wrote: > >>>> Martin did some Xorg cleanup in meta-oe today and we're seeing some > >>>> strange behaviour. The recipe in question: > >>>> > >>>> koen@dominion:/OE/tentacle/sources/meta-openembedded/meta-oe$ cat > >>>> recipes-graphics/xorg-lib/libxrender_0.9.6.bbappend > >>>> BBCLASSEXTEND = "native nativesdk" > >>>> koen@dominion:/OE/tentacle/sources/meta-openembedded/meta-oe$ > >>>> > >>>> Bitbake-layers tells me: > >>>> > >>>> INFO: libxrender_0.9.6.bb: > >>>> INFO: > >>>> /OE/tentacle/sources/meta-openembedded/meta-oe/recipes-graphics/xorg-lib/libxrender_0.9.6.bbappend > >>>> > >>>> > >>>> So the append is getting picked up, but bitbake gives me: > >>>> > >>>> ERROR: Nothing PROVIDES 'libxrender-native' > >>>> NOTE: Runtime target 'libnss-mdns' is unbuildable, removing... > >>>> Missing or unbuildable dependency chain was: ['libnss-mdns', > >>>> 'avahi', 'gtk+', 'gtk+-native', 'libxrender-native'] > >>>> > >>>> This is with HEAD of bitbake master. > >>> > >>> I did some experimentation here and this looks like bitbake-layers is > >>> seeing the file but bitbake is not. If I touch the libxrender .bb file > >>> and then run bitbake -DDD, I see the file mentioned in the logs: > >>> > >>> $ touch recipes-graphics/xorg-lib/libxrender_0.9.6.bb > >>> $ bitbake libxrender-native -DDD | grep append > >>> DEBUG: Appending .bbappend file > >>> /rphome/poky/meta/recipes-graphics/xorg-lib/libxrender_0.9.6.bbappend > >>> to /rphome/poky/meta/recipes-graphics/xorg-lib/libxrender_0.9.6.bb > >>> DEBUG: BB > >>> /rphome/poky/meta/recipes-graphics/xorg-lib/libxrender_0.9.6.bbappend: handle(data, > >>> include) > >>> DEBUG: LOAD > >>> /rphome/poky/meta/recipes-graphics/xorg-lib/libxrender_0.9.6.bbappend > >>> > >>> Can you confirm you're seeing that? Admittedly I just hacked a local > >>> checkout and I'm not using meta-oe. > >> > >> I get: > >> > >> koen@dominion:/OE/tentacle/build$ touch > >> ../sources/openembedded-core/meta/recipes-graphics/xorg-lib/libxrender_0.9.6.bb > >> > >> koen@dominion:/OE/tentacle/build$ bitbake libxrender-native -DDD | > >> grep append > >> DEBUG: Appending .bbappend file > >> /OE/tentacle/sources/meta-openembedded/meta-oe/recipes-graphics/xorg-lib/libxrender_0.9.6.bbappend > >> to > >> /OE/tentacle/sources/openembedded-core/meta/recipes-graphics/xorg-lib/libxrender_0.9.6.bb > >> > >> DEBUG: BB > >> /OE/tentacle/sources/meta-openembedded/meta-oe/recipes-graphics/xorg-lib/libxrender_0.9.6.bbappend: > >> handle(data, include) > >> DEBUG: LOAD > >> /OE/tentacle/sources/meta-openembedded/meta-oe/recipes-graphics/xorg-lib/libxrender_0.9.6.bbappend > >> > >> > >> It then chokes on renderproto-native, which gets fixes by touching the > >> .bb. So it looks like bitbake has some caching issues. > > > > I have seen exactly this behaviour. Rebuilding the cache makes it > > work. Is there an easy way > > to force the cache to rebuild? (I just touch a .conf file...) > > > >>> > >>> Do you need to add *.bbappend to BBFILES in layer.conf ? > >> > >> I have that already :) > > I discoverd the same behaviour today. Is there already something fix to > get bitbake to evaluate the bbappend file after creation? Someone should file a bug about this so we don't forget about it. I think bbappend files currently sometimes fall outside the cache logic but they obviously shouldn't and we've found a corner case... Cheers, Richard