[OE-core] [PATCH 2/2] lsbinitscripts: Add a linking for functions.lsbinitscripts

Saul Wold sgw at linux.intel.com
Sat Jul 7 00:43:10 UTC 2012


On 06/28/2012 02:35 AM, Kang Kai wrote:
> From: Xiaofeng Yan <xiaofeng.yan at windriver.com>
>
> Increase ALTERNATIVE_PRIORITY of functions to be higher than the
> value of package initscript. When lsbinitscripts installed, file
> functions provided by lsbinitscripts will be used.
>
> [YOCTO #2133]
>
> Signed-off-by: Xiaofeng Yan <xiaofeng.yan at windriver.com>
> Signed-off-by: Kang Kai <kai.kang at windriver.com>
> ---
>   meta/recipes-extended/lsb/lsbinitscripts_9.03.bb |    6 +++---
>   1 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/meta/recipes-extended/lsb/lsbinitscripts_9.03.bb b/meta/recipes-extended/lsb/lsbinitscripts_9.03.bb
> index dd92a92..f324994 100644
> --- a/meta/recipes-extended/lsb/lsbinitscripts_9.03.bb
> +++ b/meta/recipes-extended/lsb/lsbinitscripts_9.03.bb
> @@ -2,7 +2,7 @@ DESCRIPTION = "SysV init scripts which only is used in an LSB image"
>   SECTION = "base"
>   LICENSE = "GPLv2"
>   DEPENDS = "popt"
> -PR = "r0"
> +PR = "r1"
>
>   LIC_FILES_CHKSUM = "file://COPYING;md5=ebf4e8b49780ab187d51bd26aaa022c6"
>
> @@ -15,7 +15,7 @@ SRC_URI[md5sum] = "668fa2762b57ef75436303857847bba3"
>   SRC_URI[sha256sum] = "d56547a68ce223a7413b2676650b042125f047c8d6d139c5b970e118b3dc958a"
>   inherit update-alternatives
>
> -ALTERNATIVE_PRIORITY = "10"
> +ALTERNATIVE_PRIORITY = "100"
>   ALTERNATIVE_${PN} = "functions"
>   ALTERNATIVE_LINK_NAME[functions] = "${sysconfdir}/init.d/functions"
>
> @@ -23,5 +23,5 @@ do_configure[noexec] = "1"
>
>   do_install(){
>   	install -d ${D}/etc/init.d/
> -	install -m 0755 ${S}/rc.d/init.d/functions ${D}/etc/init.d/functions
> +	install -m 0755 ${S}/rc.d/init.d/functions ${D}/etc/init.d/functions.${BPN}
>   }
>
As Richard already pointed out, this rename during install is not 
needed, it happens automagically via the update-alternatives bbclass, 
this is true with the initscripts recipe also, please fix these two 
issues and we can move forward with this patch.

Thanks
	Sau!





More information about the Openembedded-core mailing list