From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Korsgaard Date: Thu, 06 Jan 2011 14:19:47 +0100 Subject: [Buildroot] Problem building avahi In-Reply-To: <20110106135549.7f6ef437@surf> (Thomas Petazzoni's message of "Thu, 6 Jan 2011 13:55:49 +0100") References: <20110105171027.1e64ffe6@surf> <20110106123238.1c227155@surf> <87ei8qgqju.fsf@macbook.be.48ers.dk> <20110106135549.7f6ef437@surf> Message-ID: <87aajegoek.fsf@macbook.be.48ers.dk> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net >>>>> "Thomas" == Thomas Petazzoni writes: Hi, Thomas> Ok, good. At least this has exhibited an interesting bug of Thomas> Buildroot that we'll have to fix. >> >> Yeah. Thomas> Something like "grep -v" the lines that have already been changed to Thomas> start with $(STAGING_DIR)/usr ? We could simply skip lines that already contain STAGING_DIR/usr - E.G. $(SED) -e '/$(subst /,\/,$(STAGING_DIR)/usr)/b' -e "s:\(['= ]\)/usr:.. >> For such a special situation, I would prefer to we tell users to do it >> with BR2_ROOTFS_POST_BUILD_SCRIPT instead. Thomas> On x86, it's fairly usual to have the kernel inside the root Thomas> filesystem, and embedded x86 systems are quite common. So this Thomas> feature could be seen as sufficiently useful to be integrated Thomas> into BR. But u-boot on x86 isn't common, right? Thomas> Something like: Thomas> config BR2_LINUX_KERNEL_INSTALL_TO_ROOTFS Thomas> bool "Install kernel image in rootfs" Thomas> default y if BR2_i386 || BR2_x86_64 Thomas> help Thomas> Copy the generated kernel image to the /boot directory Thomas> in the root filesystem Ahh, for non-uImage stuff - Yes, that's fine. -- Bye, Peter Korsgaard