From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail1.windriver.com (mail1.windriver.com [147.11.146.13]) by mail.openembedded.org (Postfix) with ESMTP id BFF3673190 for ; Thu, 21 Jul 2016 06:54:51 +0000 (UTC) Received: from ALA-HCA.corp.ad.wrs.com (ala-hca.corp.ad.wrs.com [147.11.189.40]) by mail1.windriver.com (8.15.2/8.15.1) with ESMTPS id u6L6spUZ026715 (version=TLSv1 cipher=AES128-SHA bits=128 verify=FAIL); Wed, 20 Jul 2016 23:54:51 -0700 (PDT) Received: from [128.224.162.240] (128.224.162.240) by ALA-HCA.corp.ad.wrs.com (147.11.189.40) with Microsoft SMTP Server id 14.3.248.2; Wed, 20 Jul 2016 23:54:50 -0700 To: "Burton, Ross" References: <6dc817964ba46c0263b3b07cccf84a9917b72121.1468994042.git.liezhi.yang@windriver.com> <579029B1.8030102@windriver.com> From: Robert Yang Message-ID: <579071B9.1010705@windriver.com> Date: Thu, 21 Jul 2016 14:54:49 +0800 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.8.0 MIME-Version: 1.0 In-Reply-To: <579029B1.8030102@windriver.com> Cc: OE-core Subject: Re: [PATCH 1/1] btrfs-tools: fix installed-vs-shipped 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: Thu, 21 Jul 2016 06:54:53 -0000 Content-Type: text/plain; charset="windows-1252"; format=flowed Content-Transfer-Encoding: 7bit On 07/21/2016 09:47 AM, Robert Yang wrote: > > > On 07/20/2016 07:59 PM, Burton, Ross wrote: >> >> On 20 July 2016 at 06:54, Robert Yang > > wrote: >> >> /lib64/udev/rules.d/64-btrfs-dm.rules >> >> >> This says that the btrfstools recipe is broken, not the packaging. udev looks >> in /lib no matter what $libdir is. >> >> udev reads UDEVLIBEXECDIR/rules.d, where udevlibexecdir=$(rootprefix)/lib/udev >> and we pass rootprefix=$baseprefix (which defaults to /). > > Thanks, I was misleaded by eudev, whose udevlibexecdir is ${rootlibexecdir} > which is ${libdir}/udev, so it is /lib64/udev. It supports /lib/udev and > /lib64/udev as the code in eudev-3.2/src/udev/udev-rules.c shows: > > static const char* const rules_dirs[] = { > UDEV_CONF_DIR "/rules.d", > UDEV_RULES_DIR, > UDEV_ROOT_RUN "/udev/rules.d", > UDEV_LIBEXEC_DIR "/rules.d", > #ifdef HAVE_SPLIT_USR > "/lib/udev/rules.d", > "/usr/lib/udev/rules.d", > #endif > NULL}; > > > But for systemd, it hardcoded to $(rootprefix)/lib/udev. > > I looked into tmp/deploy/rpms, the following packages doesn't install > to /lib/udev, but /lib64/udev or /usr/lib64/udev: > > tmp/deploy/rpm/core2_64/pcmciautils-018-r1.0.core2_64.rpm > tmp/deploy/rpm/core2_64/pulseaudio-server-8.0-r0.0.core2_64.rpm > tmp/deploy/rpm/core2_64/alsa-utils-alsactl-1.1.1-r0.0.core2_64.rpm > tmp/deploy/rpm/core2_64/mdadm-3.4-r0.0.core2_64.rpm > tmp/deploy/rpm/core2_64/libinput-1.3.0-r0.0.core2_64.rpm > tmp/deploy/rpm/core2_64/bluez5-5.40-r0.0.core2_64.rpm Hi Ross, I just found that when build with eudev, the files will be installed to /lib64/udev, and when build with systemd, they will be inistalled to /lib/udev, so I think that they are OK. They use "pkg-config --variable=udevdir udev" to find udev dir, while systemd is /lib/udev and eudev is /lib64/eudev. The one which has problem is libinput which installs /usr/lib/udev or /usr/lib64/udev, this in incorrect. Another problem is that systemd doesn't install udev.pc. I think that the fix for btrfs-tools is correct. I will send a V2 with another 2 fixes. // Robert > > I will fix all of them. > > // Robert > >> >> Ross