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 1TuMyd-00057e-ON for openembedded-core@lists.openembedded.org; Sun, 13 Jan 2013 13:48:20 +0100 Received: from localhost (localhost [127.0.0.1]) by tim.rpsys.net (8.13.6/8.13.8) with ESMTP id r0DCWSPH003117; Sun, 13 Jan 2013 12:32:28 GMT 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 18272-06; Sun, 13 Jan 2013 12:32:23 +0000 (GMT) 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 r0DCWFOQ003110 (version=TLSv1/SSLv3 cipher=AES256-SHA bits=256 verify=NO); Sun, 13 Jan 2013 12:32:18 GMT Message-ID: <1358080338.23113.10.camel@ted> From: Richard Purdie To: Chris Larson Date: Sun, 13 Jan 2013 12:32:18 +0000 In-Reply-To: References: <1357995930.4072.134.camel@ted> X-Mailer: Evolution 3.2.3-0ubuntu6 Mime-Version: 1.0 X-Virus-Scanned: amavisd-new at rpsys.net Cc: OE Core mailing list Subject: Re: image_types.bbclass: what's with "elf" and "cpio.gz"? X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list List-Id: Patches and discussions about the oe-core layer List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 13 Jan 2013 12:48:20 -0000 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit On Sat, 2013-01-12 at 12:15 -0700, Chris Larson wrote: > Order is important. The elf image depends on the cpio.gz image > already > having been created so a) it must be created and b) it must > happen > before elf. > > I'd think we could implement a data-driven method for this, rather > than hardcoding this knowledge. Either declare the deps via a flag, or > have one directly call the other. Calling one from the other would cause issues like knowing when the image type had already run (or cause it to run multiple times) but the deps idea in flags would be better I agree. Its just the small matter of implementing it and the complexities that entails. With pseudo now able to safely store state, there is probably no excuse not to split the rootfs task into multiple pieces and have the image creation parts as separate tasks. This would mean we can use the normal dependency logic and not have to reinvent the wheel. It would also allow parallelism too, all with existing code. The more I think about this, the more I like the idea... Cheers, Richard