[OE-core] [PATCH] mtd-utils: do not stage headers in sysroot

[Saul Wold]

On 06/24/2012 02:52 PM, Andrea Adami wrote:
> On Sun, Jun 24, 2012 at 7:35 PM, Saul Wold<sgw at linux.intel.com>  wrote:
>> On 06/21/2012 03:57 PM, Andrea Adami wrote:
>>>
>>> * Headers are included in the package for compatibility
>>> * but have not yet been synched with linux 3.0
>>> * The actual issue was that ubi-user.h in sysroot
>>> * was overwritten by the older version.
>>> * Unfortunately one ioctl was renamed:
>>> * http://lists.infradead.org/pipermail/linux-mtd/
>>> * 2011-March/034419.html
>>> * Note: the recipe will still use its own older header,
>>> * following upstream.
>>>
>>> Signed-off-by: Andrea Adami<andrea.adami at gmail.com>
>>> ---
>>>   meta/recipes-devtools/mtd/mtd-utils_1.5.0.bb |    6 +-----
>>>   1 files changed, 1 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/meta/recipes-devtools/mtd/mtd-utils_1.5.0.bb
>>> b/meta/recipes-devtools/mtd/mtd-utils_1.5.0.bb
>>> index daa3554..1a9d4d3 100644
>>> --- a/meta/recipes-devtools/mtd/mtd-utils_1.5.0.bb
>>> +++ b/meta/recipes-devtools/mtd/mtd-utils_1.5.0.bb
>>> @@ -11,16 +11,12 @@ SRC_URI =
>>> "git://git.infradead.org/mtd-utils.git;protocol=git;tag=ca39eb1d98e736
>>>
>>>   S = "${WORKDIR}/git/"
>>>
>>> -PR = "r0"
>>> +PR = "r1"
>>>
>>>   EXTRA_OEMAKE = "'CC=${CC}' 'RANLIB=${RANLIB}' 'AR=${AR}'
>>> 'CFLAGS=${CFLAGS} -I${S}/include -DWITHOUT_XATTR' 'BUILDDIR=${S}'"
>>>
>>>   do_install () {
>>>         oe_runmake install DESTDIR=${D} SBINDIR=${sbindir} MANDIR=${mandir}
>>> INCLUDEDIR=${includedir}
>>> -       install -d ${D}${includedir}/mtd/
>>> -       for f in ${S}/include/mtd/*.h; do
>>> -               install -m 0644 $f ${D}${includedir}/mtd/
>>> -       done
>>>   }
>>
>> Just curious how this was tested?
>>
>> I did a couple of different builds and on one clean rebuild, I found that
>> udev failed because it could not find mtd/mtd-user.h
>>
>> Thanks
>>         Sau!
>>
>>
>>>
>>>   PARALLEL_MAKE = ""
>
> Strange...
> I've done build from scratch of core-image-base to test and all was
> built just fine, included udev_164.
> It would be indeed bizarre if udev would depend on headers provided by
> mtd-utils, though.
>
> Finally, mtd-user.h belongs to linux-libc-headers(-dev) package.
>
If that's the case, then maybe there's a race condition as to or we need 
to also add linux-libc-headers to the udev package to ensure they are 
installed correctly before udev is built.

Sau!

> Regards
>
> Andrea
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core at lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
>