From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from seketeli.net ([91.121.166.71] helo=ms.seketeli.net) by linuxtogo.org with esmtp (Exim 4.72) (envelope-from ) id 1UPJil-0001dl-G5 for openembedded-devel@lists.openembedded.org; Mon, 08 Apr 2013 23:35:23 +0200 Received: from localhost (176-26-190-109.dsl.ovh.fr [109.190.26.176]) by ms.seketeli.net (Postfix) with ESMTP id 23425EA042 for ; Mon, 8 Apr 2013 23:25:35 +0200 (CEST) Received: by localhost (Postfix, from userid 1000) id 8D0F9A40685; Mon, 8 Apr 2013 23:18:31 +0200 (CEST) From: Apelete Seketeli To: openembedded-devel@lists.openembedded.org Date: Mon, 8 Apr 2013 23:18:21 +0200 Message-Id: X-Mailer: git-send-email 1.7.10.4 Subject: [meta-handheld][PATCH v3 0/1] Pull request 20130408 X-BeenThere: openembedded-devel@lists.openembedded.org X-Mailman-Version: 2.1.11 Precedence: list Reply-To: openembedded-devel@lists.openembedded.org List-Id: Using the OpenEmbedded metadata to build Distributions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 08 Apr 2013 21:35:26 -0000 Hello, On Sunday 07 April 2013, Apelete Seketeli wrote: > On Friday 05 April 2013, Apelete Seketeli wrote: > > Following the patches I sent to the meta-handheld layer a few days ago > > to add machine support for the Ben Nanonote, I found out that the > > corresponding kernel recipe fails to build on oe-core (builds fine on > > oe-classic). > > A few settings were missing and, more importantly, the version of gcc > > in oe-core seems to enable the -Wunused-but-set-variable switch, which > > makes it error on many variables that were indeed 'unused but > > set'. After trying to solve this issue the easy way, by setting > > 'CFLAGS=-Wno-error=unused-but-set-variable -Wno-error=unused-but-set-parameter' > > to no avail, I chose to annotate the aforementioned variables in the > > kernel code with __attribute__((__unused__)). > > > > The patches that come as a follow-up of this message are what it's all > > about; sorry for the inconvenience in the first place. > > I took the time to fix the patch for the GCC 4.6 build issue. Instead > of just fixing the issue on my own, I looked at how it was done in the > mainline kernel and backported the fixes through a new patch named > fix-gcc-4.6.0-build-errors.patch; this approach seems more logical to > me. > While I was at it, also removed "[PATCH 2/2] ben-nanonote: add > DEFAULTTUNE to machine configuration" from the patchset, following > Paul's advice. Added a header to the "fix-gcc-4.6.0-build-errors.patch" kernel patch, describing the changes and stating the Upstream-Status. Tested by building linux-jlime-ben-nanonote from scratch on top of master branch. Please pull into meta-handheld. The following changes since commit 913735dce388f660dfbbe288ed736e12ed48c8d8: linux-yocto-tiny-kexecboot_3.2: removed unused, non-working MTD-RW patch (2013-04-06 00:25:08 +0100) are available in the git repository at: git://seketeli.fr/~apelete/meta-handheld.git fix-nn-kernel-build Apelete Seketeli (1): linux-jlime-ben-nanonote: fix GCC-4.6 build errors .../fix-gcc-4.6.0-build-errors.patch | 124 ++++++++++++++++++++ .../linux/linux-jlime-ben-nanonote_2.6.36.bb | 5 +- 2 files changed, 128 insertions(+), 1 deletion(-) create mode 100644 recipes-kernel/linux/linux-jlime-ben-nanonote-2.6.36/fix-gcc-4.6.0-build-errors.patch -- 1.7.10.4