From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga11.intel.com (mga11.intel.com [192.55.52.93]) by mail.openembedded.org (Postfix) with ESMTP id B0FE960D74 for ; Wed, 16 Oct 2013 16:48:50 +0000 (UTC) Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by fmsmga102.fm.intel.com with ESMTP; 16 Oct 2013 09:48:30 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.93,508,1378882800"; d="scan'208";a="412031968" Received: from unknown (HELO [10.255.15.253]) ([10.255.15.253]) by fmsmga001.fm.intel.com with ESMTP; 16 Oct 2013 09:48:01 -0700 Message-ID: <525EC341.9080407@linux.intel.com> Date: Wed, 16 Oct 2013 09:48:01 -0700 From: Saul Wold User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130923 Thunderbird/17.0.9 MIME-Version: 1.0 To: Peter Seebach References: In-Reply-To: Cc: OE-core Subject: Re: [PATCH 0/2] Use alternatives for all of binutils instead of -symlinks X-BeenThere: openembedded-core@lists.openembedded.org X-Mailman-Version: 2.1.12 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: Wed, 16 Oct 2013 16:48:51 -0000 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit During MUT builds, we saw a failure with the toolchain | Computing transaction...error: Can't install binutils-cross-canadian-arm-2.23.2-r4@i686_nativesdk: no package provides update-alternatives-cworth | Can be seen all architectures. Sau! On 10/10/2013 01:06 PM, Peter Seebach wrote: > The original complaint that got me started on this was that someone > was using binutils on a target, and didn't have an ld command, so they > asked for it. We added binutils-symlinks, and that seemed to work, but > eventually we ran into a problem because "ar" and "strings" weren't > showing up. > > After some asking about as to why ar and strings might be omitted from > binutils, we concluded that it was probably because they were using > the alternatives mechanism. In a local branch, we experimented for a while > with just using that instead of the -symlinks package, and found that > it was overall better-behaved; in particular, it produces the desireable > result that you don't have to know about or add a "-symlinks" package > to use the utilities in the common case where you just install a package > and expect its binaries to show up. It also eliminates the odd special > case difference between ar/strings (which can be provided by busybox) > and the other utilities. > > This is implemented as two patches. The first switches to using the > alternatives mechanism, but leaves the alternatives in the -symlinks > package rather than in the base package. The second moves them into > the base package, and adds an RPROVIDES for the -symlinks name in > case anyone is using it. > > One caveat: This can produce warnings because the "embedspu" and "ld.gold" > binaries don't always exist, but they sometimes exist. I'm not sure whether > there's a good way to fix that... > > The following changes since commit 1149b1fef8912f77d971242dfec151fff5a3aa51: > > build-appliance: Update SRCREV for release (2013-10-08 16:33:25 +0100) > > are available in the git repository at: > git://git.yoctoproject.org/poky-contrib seebs/binutils-alternatives > http://git.yoctoproject.org/cgit.cgi/poky-contrib/log/?h=seebs/binutils-alternatives > > Peter Seebach (2): > Use alternatives for the binutils-symlinks package. > Drop binutils-symlinks > > meta/recipes-devtools/binutils/binutils.inc | 68 ++++++++++++++++----------- > 1 files changed, 40 insertions(+), 28 deletions(-) > > _______________________________________________ > Openembedded-core mailing list > Openembedded-core@lists.openembedded.org > http://lists.openembedded.org/mailman/listinfo/openembedded-core > >