Thank you Kevin,
I run the script the error seems to have gone away. I have hit another
error, which I will rasie separately.
thanks and regards
Anoop
Best regards
A.B
P Please think about the environment before printing this e-mail
*7 Switch off as you go | qAlways recycle | P Save a tree...please don't
print this e-mail unless you really need to.*
On Thu, Aug 7, 2014 at 8:52 PM, Kévin THIERRY <
[email protected]> wrote:
> Your recipe seems outdated, please update meta-tizen (you can use the
> script we provide to do that as explained in the wiki).
>
> Kévin
>
>
> On 07/08/2014 17:03, Anoop Babu wrote:
>
>> Dear Kevin
>>
>> I have updated rpm-extraconf.inc with changes as you have mentioned. But
>> unfortunately it has not resolved the issue.
>> I'm hitting the same issue. I have attached the log files as well as
>> the.inc file for your reference.
>> In the main log i can see the machine configuration as
>> TARGET_SYS = "arm-poky-linux-gnueabi"
>> MACHINE = "imx6qsabreauto"
>>
>> In the run file I see two suspicious steps mainly
>> unset MACHINE (line no 83) and unset DISTRO( line no 89)
>>
>> regards
>> Anoop
>>
>>
>> Best regards
>> A.B
>> PPlease think about the environment before printing this e-mail
>>
>> /
>>
>> *7*Switch off as you go |*q*Always recycle |*P*Save a tree...please don't
>> print this e-mail unless you really need to.
>>
>> /
>>
>>
>> On Thu, Aug 7, 2014 at 5:55 PM, Kévin THIERRY <kevin.thierry@open.
>> eurogiciel.org <mailto:[email protected]>> wrote:
>>
>>
>> On 07/08/2014 14:07, Kévin THIERRY wrote:
>>
>> commit 65bd0041da560c7ba9ede5e53c06b3f65d7906f2 is already
>> present in your branch, what I meant was that maybe this
>> change isn't correct and creates a new issue.
>>
>> Looking at the log you provided I see things that surprise me:
>>
>> * BUILDTARGET="--host=x86_64-linux-gnu "
>> This an error in the recipe, this value shouldn't be
>> hard-coded but set to "--host=${HOST_SYS} " and exported so
>> replacing this line with:
>> export BUILDTARGET="--host=${HOST_SYS} " should be better.
>>
>> Ignore that and try to add:
>>
>> export BUILDTARGET="--target=${TARGET_SYS} "
>>
>> in the do_compile_prepend() section of the rpm-extraconf.inc file
>> and replace:
>>
>> ./configure --build=${TARGET_SYS} --host=x86_64-linux-gnu
>>
>> with:
>>
>> ./configure --build=${TARGET_SYS} --host=${HOST_SYS}
>>
>> in rpm-extraconf.inc. (Not tested.)
>>
>>
>> The rpm recipe is a mess, I will try to clean it when I have time.
>>
>> Kevin
>>
>>
>>
>> * Are you sure you are building against imx6 (ARM) because I
>> see references to i686 everywhere in your log ? Did you set
>> the "MACHINE" variable in your conf/local.conf correctly and
>> add the required meta in your conf/bblayer.conf ?
>>
>> Kevin
>>
>>
>> On 07/08/2014 12:31, Anoop Babu wrote:
>>
>> Dear Kevin,
>>
>> Thank you for the inputs. For you reference I have
>> attached run.do_compile file.
>>
>> I will lookout for solution time being. On a next level I
>> will cherrypick commit
>> 65bd0041da560c7ba9ede5e53c06b3f65d7906f2.
>>
>> Will let you know of any relevant things w.r.t this issue.
>>
>> Best regards
>> Anoop
>>
>>
>>
>> On Thu, Aug 7, 2014 at 12:50 PM, Kévin THIERRY
>> <[email protected]
>> <mailto:[email protected]>
>> <mailto:[email protected]
>> <mailto:[email protected]>>> wrote:
>>
>> Hi Anoop,
>>
>> I can't reproduce this error (but I'm not building for
>> imx6). I've
>> seen similar error message regarding libcap but can't
>> remember
>> what we did to correct it. Anyway in your configure
>> command line
>> you have a $BUILDTARGET which is not expanded, that
>> looks suspicious.
>>
>> Another good file to provide when you have an issue is the
>> run.do_compile file which contains all the
>> instructions executed
>> for the compile section.
>>
>> I think looking at the $BUILDTARGET is a good start
>> toinvestigate
>> this bug. Also, a fix has been made on libcap in
>> meta-tizen which
>> removes configure options (commit
>> 65bd0041da560c7ba9ede5e53c06b3f65d7906f2), maybe a
>> useful option
>> has been removed in the process.
>>
>> Let me know if you find anything relevant to this issue.
>>
>> Kevin
>>
>>
>> On 06/08/2014 17:00, Anoop Babu wrote:
>>
>> Dear Kevin,
>>
>> Thank you for your inputs.
>> I had added codes to do_compile_prepend()
>>
>> export HOST_SYS=${HOST_SYS}
>> export BUILD_SYS=${BUILD_SYS}
>> export STAGING_LIBDIR=${STAGING_LIBDIR}
>> export STAGING_INCDIR=${STAGING_INCDIR}
>>
>> Seems I have the same error
>> Checking for cap_set_file in -lcap... no
>> configure: error: --with-msm given, but
>> cap_set_file not found
>> in libcap
>>
>> Attached is the log file please have a look.
>>
>> Best regards
>> Anoop
>>
>>
>> On Wed, Aug 6, 2014 at 6:16 PM, Kévin THIERRY
>> <[email protected]
>> <mailto:[email protected]>
>> <mailto:[email protected]
>> <mailto:[email protected]>>
>> <mailto:[email protected]
>> <mailto:[email protected]>
>> <mailto:[email protected]
>> <mailto:[email protected]>>>> wrote:
>>
>> As I said the error looks very similar to the
>> one I get with
>> ecryptfs-utils. I got it working by exporting
>> the apparently
>> missing variables. To do that, just add a
>> 'do_compile_prepend()"
>> section in the rpm-extraconf.inc file like
>> this one:
>>
>> do_compile_prepend() {
>> export HOST_SYS=${HOST_SYS}
>> export BUILD_SYS=${BUILD_SYS}
>> }
>>
>> If you still have unknown variables, just add
>> them the
>> same way (I
>> also had to add "STAGING_LIBDIR" and
>> "STAGING_INCDIR").
>>
>> I really don't know where this comes from
>> neither why you have
>> this issue with rpm and I don't. I think it is
>> strange
>> that such
>> variables are not available. Anyway, it seems
>> to work with
>> this
>> correction...
>>
>> Anoop, please tell me if it works for you.
>>
>> Also I'd like to know if anyone have the same
>> issue with
>> rpm. If
>> it's the case it will probably be necessary to
>> add this
>> correction
>> in the recipe.
>>
>> Regards,
>> Kevin
>>
>>
>>
>> On 06/08/2014 14:18, Anoop Babu wrote:
>>
>> Dear Kevin,
>>
>> I did a retry as suggested by you
>> bitbake rpm -c cleanall
>> bitbake rpm
>>
>> Unfortunately I have hit the same issue.
>> Attached is
>> the log
>> file, please have a look.
>>
>> Regards
>> Anoop
>>
>> Best regards
>> A.B
>> PPlease think about the environment before
>> printing
>> this e-mail
>>
>> /
>>
>> *7*Switch off as you go |*q*Always recycle
>> |*P*Save a
>> tree...please don't print this e-mail
>> unless you
>> really need to.
>>
>> /
>>
>>
>>
>> On Wed, Aug 6, 2014 at 12:57 PM, Kévin THIERRY
>> <[email protected]
>> <mailto:[email protected]>
>> <mailto:[email protected]
>> <mailto:[email protected]>>
>> <mailto:[email protected]
>> <mailto:[email protected]>
>> <mailto:[email protected]
>> <mailto:[email protected]>>>
>> <mailto:[email protected]
>> <mailto:[email protected]>
>> <mailto:[email protected]
>> <mailto:[email protected]>>
>>
>> <mailto:[email protected]
>> <mailto:[email protected]>
>> <mailto:[email protected]
>> <mailto:[email protected]>>>>> wrote:
>>
>> Hi Anoop,
>>
>>
>> On 05/08/2014 18:46, Anoop Babu wrote:
>>
>> Building tizen using yocto
>>
>> Host OS and architecture -
>> -Ubuntu SMP i686/ i386
>> GNU/Linux
>> Target architecture/BSP - iMX6 Sabre /
>> meta-fsl-arm
>> The image you're trying to build -
>> tizen-common-core-image-minimal-dev
>>
>> Build procedure followed is from
>> https://wiki.tizen.org/wiki/Build_Tizen_with_Yocto
>> but for imx6
>>
>> Fetch latest for all repos today.
>>
>> Also updated rpm code based for
>> bashism issue
>> Replaced "export CCFLAGS+=" -fPIC
>> "" with
>> "export CCFLAGS="${CCFLAGS} -fPIC "".
>>
>> attached is the rpm .inc files
>>
>> Looking quickly at the logs you
>> provided I suspect the
>> error to be
>> there:
>>
>> checking for a Python interpreter
>> with version
>> >= 2.6...
>> python
>> checking for python...
>> /home/sfm/YOCTO/build/tmp/sysroots/i686-linux/usr/bin/
>> python-native/python
>> checking for python version...
>> 2.7
>> checking for python platform...
>> linux2
>>
>> checking for python script
>> directory...
>> Traceback (most recent call last):
>>
>> File "<string>", line 22, in
>> <module>
>>
>> File
>> "/home/sfm/YOCTO/build/tmp/sysroots/i686-linux/usr/lib/
>> python2.7/distutils/sysconfig.py",
>> line 22, in <module>
>>
>> PREFIX =
>> os.path.normpath(sys.prefix).replace(
>> os.getenv("BUILD_SYS"),
>> os.getenv("HOST_SYS") )
>>
>> TypeError
>> :
>> expected a character buffer object
>>
>>
>> ${libdir}/python2.7/site-packages
>> checking for python extension
>> module directory...
>> Traceback (most recent call last):
>> File "<string>", line 22, in
>> <module>
>>
>> File
>> "/home/sfm/YOCTO/build/tmp/sysroots/i686-linux/usr/lib/
>> python2.7/distutils/sysconfig.py",
>> line 22, in <module>
>> PREFIX =
>> os.path.normpath(sys.prefix).replace(
>> os.getenv("BUILD_SYS"),
>> os.getenv("HOST_SYS") )
>> TypeError: expected a character
>> buffer object
>>
>> ${libdir}/python2.7/site-packages
>>
>> That's weird because yesterday I
>> encountered a similar
>> issue but
>> with ecryptfs-utils not rpm but didn't
>> have time to
>> investigate
>> it. I just rebuilt rpm without issue.
>> Anoop, can you
>> please clean
>> retry a clean rpm build ?
>>
>> bitbake rpm -c cleanall
>> bitbake rpm
>>
>> Please, let us know if you don't have
>> this issue
>> anymore.
>>
>> Regards,
>> Kevin
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>
_______________________________________________
Dev mailing list
[email protected]
https://lists.tizen.org/listinfo/dev
