[meta-intel] [PATCH 1/5] canterbury-corpus: don't skip all QA tests

Kamble, Nitin A nitin.a.kamble at intel.com
Tue May 20 09:45:51 PDT 2014


On 5/20/2014 9:17 AM, Ross Burton wrote:
> Instead of setting ERROR_QA to "" (disabling all tests, instead of just the
> problematic ones), adding nothing to WARN_QA (cruft from previous revisions),
> and also marking do_package_qa as noexec (which doesn't work), just set
> INSANE_SKIP to skip the specific tests that fail with this package.
>
> Signed-off-by: Ross Burton <ross.burton at intel.com>
> ---
>   common/recipes-corpus/canterbury-corpus/canterbury-corpus.bb |   10 ++--------
>   1 file changed, 2 insertions(+), 8 deletions(-)
>
> diff --git a/common/recipes-corpus/canterbury-corpus/canterbury-corpus.bb b/common/recipes-corpus/canterbury-corpus/canterbury-corpus.bb
> index 4dcdea1..d15ad15 100644
> --- a/common/recipes-corpus/canterbury-corpus/canterbury-corpus.bb
> +++ b/common/recipes-corpus/canterbury-corpus/canterbury-corpus.bb
> @@ -17,14 +17,6 @@ SRC_URI = "http://corpus.canterbury.ac.nz/resources/cantrbry.tar.gz"
>   SRC_URI[md5sum] = "442e56cfffdf460d25b0b91650a55908"
>   SRC_URI[sha256sum] = "f140e8a5b73d3f53198555a63bfb827889394a42f20825df33c810c3d5e3f8fb"
>   
> -# Disable architecture QA check for this package since it contains
> -# pre-compiled executable "sum" for SPARC. The package is used
> -# for compression benchmarking only.
> -WARN_QA += ""
> -ERROR_QA = ""
> -
> -do_package_qa[noexec] = "1"
> -
>   do_unpack () {
>   	mkdir -p ${S}
>   	tar -xf ${DL_DIR}/cantrbry.tar.gz -C ${S}
> @@ -40,3 +32,5 @@ do_install () {
>   	install -d ${D}${base_libdir}/firmware
>   	install -m 644 ${S}/* ${D}${base_libdir}/firmware
>   }
> +
> +INSANE_SKIP_${PN} = "arch ldflags"
Hi Ross,
  These are good code cleanups. As there are some special handling in 
these recipes, it demands some explanation in comments why is it done 
this way. That way if someone looks at recipes in future can easily 
understand why these lines are here.

Thanks,
Nitin




More information about the meta-intel mailing list