Quantcast

Building Sakoman's GNOME-R13 image

classic Classic list List threaded Threaded
36 messages Options
12
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Building Sakoman's GNOME-R13 image

JamesAng
Dear all & especially to Steve,

I'm trying to bitbake steve sakoman's gnome-r13 image using his repository and OE tree at his site for my Overo FE and encounter some issues with it.

I've checked out the 'gnome-r13' branch of his OE tree and tried 'bitbake virtual/kernel' hoping to build v3.0 or above of the kernel but bitbake keep falling back to v2.6.39-r102 instead.

auto.conf defined MACHINE = "overo"
local.conf defined DISTRO = "angstrom-2010.x", DISTRO_TYPE = "debug", ANGSTROM_URI = "http://feeds.sakoman.com", ANGSTROM_FEED_NAME = "gnome-r13"

All are done on a fresh cycle on Ubuntu 11.10 x64.


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

JamesAng
I couldn't figure out why bitbake cannot build v3.0 of the kernel.

Any help and tips on this is greatly appreciated.

James.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

Steve Sakoman
On Sun, Nov 13, 2011 at 10:41 PM, JamesAng <[hidden email]> wrote:
>
> I couldn't figure out why bitbake cannot build v3.0 of the kernel.

bitbake linux-sakoman-pm

Steve

------------------------------------------------------------------------------
RSA(R) Conference 2012
Save $700 by Nov 18
Register now
http://p.sf.net/sfu/rsa-sfdev2dev1
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

JamesAng
Hi Steve,

sakoman wrote
On Sun, Nov 13, 2011 at 10:41 PM, JamesAng <angweiyang@gmail.com> wrote:
>
> I couldn't figure out why bitbake cannot build v3.0 of the kernel.

bitbake linux-sakoman-pm

Steve
I found out that I have to modify auto.conf's MACHINE to "omap3-multi" instead of using "overo" as both linux-sakoman and linux-sakoman-pm recipes does not have "overo" in its COMPATIBLE_MACHINE parameter.

What an oversight by me..

BTW comparing to GS's linux-omap3_git.bb recipe, there is alway an additional section

do_configure_prepend() {

        if [ "${MUSB_MODE}" = "host" ]; then
            sed -i 's:CONFIG_USB_GADGET=y:# CONFIG_USB_GADGET is not set:g' ${WORKDIR}/defconfig
            sed -i 's:# CONFIG_USB_MUSB_HOST is not set:CONFIG_USB_MUSB_HOST=y:g' ${WORKDIR}/defconfig
            sed -i 's:CONFIG_USB_MUSB_PERIPHERAL=y:# CONFIG_USB_MUSB_PERIPHERAL is not set:g' ${WORKDIR}/defconfig
            sed -i 's:CONFIG_USB_MUSB_OTG=y:# CONFIG_USB_MUSB_OTG is not set:g' ${WORKDIR}/defconfig
            sed -i 's:# CONFIG_USB_MUSB_HDRC_HCD is not set:CONFIG_USB_MUSB_HDRC_HCD=y:g' ${WORKDIR}/defconfig
            sed -i 's:CONFIG_USB_GADGET_MUSB_HDRC=y:# CONFIG_USB_GADGET_MUSB_HDRC is not set:g' ${WORKDIR}/defconfig
        fi

        if [ "${MUSB_MODE}" = "peripheral" ]; then
            sed -i 's:# CONFIG_USB_GADGET is not set:CONFIG_USB_GADGET=y:g' ${WORKDIR}/defconfig
            sed -i 's:CONFIG_USB_MUSB_HOST=y:# CONFIG_USB_MUSB_HOST is not set:g' ${WORKDIR}/defconfig
            sed -i 's:# CONFIG_USB_MUSB_PERIPHERAL is not set:CONFIG_USB_MUSB_PERIPHERAL=y:g' ${WORKDIR}/defconfig
            sed -i 's:CONFIG_USB_MUSB_OTG=y:# CONFIG_USB_MUSB_OTG is not set:g' ${WORKDIR}/defconfig
            sed -i 's:CONFIG_USB_MUSB_HDRC_HCD=y:# CONFIG_USB_MUSB_HDRC_HCD is not set:g' ${WORKDIR}/defconfig
            sed -i 's:# CONFIG_USB_GADGET_MUSB_HDRC is not set:CONFIG_USB_GADGET_MUSB_HDRC=y:g' ${WORKDIR}/defconfig
        fi

        if [ "${MUSB_MODE}" = "otg" ]; then
            sed -i 's:# CONFIG_USB_GADGET is not set:CONFIG_USB_GADGET=y:g' ${WORKDIR}/defconfig
            sed -i 's:CONFIG_USB_MUSB_HOST=y:# CONFIG_USB_MUSB_HOST is not set:g' ${WORKDIR}/defconfig
            sed -i 's:CONFIG_USB_MUSB_PERIPHERAL=y:# CONFIG_USB_MUSB_PERIPHERAL is not set:g' ${WORKDIR}/defconfig
            sed -i 's:# CONFIG_USB_MUSB_OTG is not set:CONFIG_USB_MUSB_OTG=y:g' ${WORKDIR}/defconfig
            sed -i 's:CONFIG_USB_MUSB_HDRC_HCD=y:# CONFIG_USB_MUSB_HDRC_HCD is not set:g' ${WORKDIR}/defconfig
            sed -i 's:# CONFIG_USB_GADGET_MUSB_HDRC is not set:CONFIG_USB_GADGET_MUSB_HDRC=y:g' ${WORKDIR}/defconfig
        fi
}


is this a must? I presume it is a fail-safe to ensure that the defconfig is correctly modified.

Thanks.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

JamesAng
In reply to this post by JamesAng
After changing MACHINE parameter, I'm able to build a v3.0 kernel.

auto.conf now has MACHINE = omap3_multi

My host is Ubuntu 11.10 x64.
Thus, I've also modified angstrom-2010-preferred-versions.inc to workaround openssl v1.0.0d failure to v0.9.8m as suggested in other posts.

But when building sakoman-console-image and sakoman-gnome-image, the following errors appears and I'm stuck again.

Please advise..

Many, many thanks in adv.

James.

-- Output --
"bitbake sakoman-console-image"

-- #1 --
ERROR: 'virtual:native:/home/gumstix/overo-oe/sk.openembedded/recipes/openjade/openjade_1.3.2.bb' failed
openjade-1.3.2-r5.log.do_compile
Please help.

"bitbake sakoman-gnome-image" has additional errors on top of the above

-- #2 --
ERROR: '/home/gumstix/overo-oe/sk.openembedded/recipes/python/python-pyqt_4.8.4.bb' failed
python-pyqt-4.8.4-r0.log.do_compile
Source URL not available. Downloaded from http://pkgs.fedoraproject.org/repo/pkgs/. But do_compile() still failed.

-- #3 --
ERROR: '/home/gumstix/overo-oe/sk.openembedded/recipes/gs/gs-tools-native_9.01.bb' failed
gs-tools-native-9.01-r0.log.do_compile


-- #4 --
ERROR: '/home/gumstix/overo-oe/sk.openembedded/recipes/icedtea/icedtea6-native_1.7.10.bb' failed
icedtea6-native-1.7.10-r4.0.log.do_compile

-- #5 --
ERROR: '/home/gumstix/overo-oe/sk.openembedded/recipes/mozilla/firefox_3.6.8.bb' failed
firefox-3.6.8-r5.log.do_compile
Temporary removed it from angstrom-task-gnome. Don't know why it wouldn't compile. Any idea?

-- #6 --
ERROR: '/home/gumstix/overo-oe/sk.openembedded/recipes/ti/ti-msp430-chronos_1.05.00.00.bb' failed.
Removed from sakoman-gnome-image since I guess it is not part of Overo.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

JamesAng
A following update of my journey..

On Thu, Nov 17, 2011 at 5:54 PM, JamesAng <[hidden email]> wrote:
>
> After changing MACHINE parameter, I'm able to build a v3.0 kernel.
>
> auto.conf now has MACHINE = omap3_multi
>
> My host is Ubuntu 11.10 x64.
> Thus, I've also modified angstrom-2010-preferred-versions.inc to workaround
> openssl v1.0.0d failure to v0.9.8m as suggested in other posts.
>
> But when building sakoman-console-image and sakoman-gnome-image, the
> following errors appears and I'm stuck again.
>
> -- Output --
> "bitbake sakoman-console-image"
>
> -- #1 --
> ERROR:
> 'virtual:native:/home/gumstix/overo-oe/sk.openembedded/recipes/openjade/openjade_1.3.2.bb'
> failed
> http://old.nabble.com/file/p32860740/openjade-1.3.2-r5.log.do_compile
> openjade-1.3.2-r5.log.do_compile
> Please help.

I modified iputils recipe to remove the dependency for docbook and sgml so that it would not need openjade. This was a known long, old trick that is haunting lot of newbies again and again which I found out after searching deep into the archives.

Although this avoid the OpenJade error, now a new error appears with do_rootfs() of sakoman-console-image.bb.

See the log.do_rootfs file.

Mainly the errors are
-- Cut here --
| Installing task-base-extended (1.0-r97.9) to root...
| Downloading file:/mnt/oetemp/tmp12/deploy/eglibc/ipk/omap3-multi/task-base-extended_1.0-r97.9_omap3-multi.ipk.
| busybox: unsatisfied recommendation for libgcc1

...
| (lots of unsatisfied recommendation for XXXX)
| Collected errors:
|  * satisfy_dependencies_for: Cannot satisfy the following dependencies for task-base-extended:
|  *     libgcc1 (>= 4.5) *
|  * opkg_install_cmd: Cannot install package task-base-extended.
|  * resolve_conffiles: Existing conffile /mnt/oetemp/tmp12/rootfs/sakoman-console-image/etc/device_table is different from the conffile in the new package. The new conffile will be placed at /mnt/oetemp/tmp12/rootfs/sakoman-console-image/etc/device_table-opkg.

Please advise on this..

>
> "bitbake sakoman-gnome-image" has additional errors on top of the above
>
> -- #2 --
> ERROR:
> '/home/gumstix/overo-oe/sk.openembedded/recipes/python/python-pyqt_4.8.4.bb'
> failed
> http://old.nabble.com/file/p32860740/python-pyqt-4.8.4-r0.log.do_compile
> python-pyqt-4.8.4-r0.log.do_compile
> Source URL not available. Downloaded from
> http://pkgs.fedoraproject.org/repo/pkgs/. But do_compile() still failed.
>

To remove this dependency, I also remove 'hplip' from angstrom-task-gnome.bb and it goes away. I don't know if it will break other stuffs that depends on it as it is part of task-gnome-cups package.

> -- #3 --
> ERROR:
> '/home/gumstix/overo-oe/sk.openembedded/recipes/gs/gs-tools-native_9.01.bb'
> failed
> http://old.nabble.com/file/p32860740/gs-tools-native-9.01-r0.log.do_compile
> gs-tools-native-9.01-r0.log.do_compile
>

For this, removed 'evince' initially from angstrom-task-gnome but it still get called and compiled by some others and Scott's oe-deptool shows that it might not be avoidable as it's needed by python-gnome-desktop. Can python-gnome-desktop be removed with out affect the basic gnome desktop?

>
> -- #4 --
> ERROR:
> '/home/gumstix/overo-oe/sk.openembedded/recipes/icedtea/icedtea6-native_1.7.10.bb'
> failed
> http://old.nabble.com/file/p32860740/icedtea6-native-1.7.10-r4.0.log.do_compile
> icedtea6-native-1.7.10-r4.0.log.do_compile
>

Clueless about this.. needed by OpenJDK-6-JRE but oe-deptool does not show any link directly to both images..

> -- #5 --
> ERROR:
> '/home/gumstix/overo-oe/sk.openembedded/recipes/mozilla/firefox_3.6.8.bb'
> failed
> http://old.nabble.com/file/p32860740/firefox-3.6.8-r5.log.do_compile
> firefox-3.6.8-r5.log.do_compile
> Temporary removed it from angstrom-task-gnome. Don't know why it wouldn't
> compile. Any idea?
>
> -- #6 --
> ERROR:
> '/home/gumstix/overo-oe/sk.openembedded/recipes/ti/ti-msp430-chronos_1.05.00.00.bb'
> failed.
> Removed from sakoman-gnome-image since I guess it is not part of Overo.
>
> --
> View this message in context: http://old.nabble.com/Building-Sakoman%27s-GNOME-R13-image-tp32838470p32860740.html
> Sent from the Gumstix mailing list archive at Nabble.com.

Is it because using "2010.X" results in the use of GCC 4.5 which breaks the stuffs above or that Ubuntu 11.10 cause the faults?

--
Regards,
James


------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

sakoman-console-image.log.do_rootfs (77K) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

Steve Sakoman
On Thu, Nov 17, 2011 at 11:04 PM, James <[hidden email]> wrote:

> Is it because using "2010.X" results in the use of GCC 4.5 which breaks the
> stuffs above or that Ubuntu 11.10 cause the faults?

I suspect it is 64 bit Ubuntu 11.10 that is causing the issues.  I use
mainly 32 bit Ubuntu 10.04 and do not have these issues.

And since I don't really care for the Ubuntu 11.10 experience I am not
likely to switch to it anytime soon!

Steve

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

JamesAng
sakoman wrote
On Thu, Nov 17, 2011 at 11:04 PM, James <angweiyang@gmail.com> wrote:

> Is it because using "2010.X" results in the use of GCC 4.5 which breaks the
> stuffs above or that Ubuntu 11.10 cause the faults?

I suspect it is 64 bit Ubuntu 11.10 that is causing the issues.  I use
mainly 32 bit Ubuntu 10.04 and do not have these issues.

And since I don't really care for the Ubuntu 11.10 experience I am not
likely to switch to it anytime soon!

Steve
I really don't understand why 11.10 can cause so much issue since bitbake will build the toolchain for the necessary cross-compilation..

Darn..
Now, I've to find a way to downgrade back to 11.04 without having to re-install everything from scratch..

James.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

error booting from MMC on reset

Keane, Ben (STRX)
Hello,

I was wondering if anyone had any experience with the following error:

<error>
U-Boot 2011.06-00374-gfa82f87-dirty (Nov 16 2011 - 15:15:28)

OMAP3630/3730-GP ES2.1, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
Gumstix Overo board + LPDDR/NAND

<cut><cut><cut>

reading uImage
mmc_send_cmd: timedout waiting for cmddis!
mmc fail to send stop cmd
mmc_send_cmd: timedout waiting for cmddis!
mmc_send_cmd: timedout waiting for cmddis!

3152980 bytes read
## Booting kernel from Legacy Image at 82000000 ...
   Image Name:   Linux-2.6.32
   Image Type:   ARM Linux Kernel Image (uncompressed)
   Data Size:    3152916 Bytes = 3 MiB
   Load Address: 80008000
   Entry Point:  80008000
   Verifying Checksum ... Bad Data CRC
ERROR: can't get kernel image!
</error>

This happens when I reboot the WaterStorm com with the Reset button but boots normally when I remove the power plug and plug back in again. Pressing and holding the Reset button any different amounts of time does not seem to help.

Any help with this is appreciated.

Kind Regards,


Ben Keane


PROPRIETARY: This e-mail contains proprietary information some or all of which may be legally privileged.  It is intended for the recipient only.  If an addressing or transmission error has misdirected this e-mail, please notify the authority by replying to this e-mail.  If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this e-mail.

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: error booting from MMC on reset

Alex Gibson-3
On 21/11/2011 5:09 PM, Keane, Ben (STRX) wrote:

> Hello,
>
> I was wondering if anyone had any experience with the following error:
>
> <error>
> U-Boot 2011.06-00374-gfa82f87-dirty (Nov 16 2011 - 15:15:28)
>
> OMAP3630/3730-GP ES2.1, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
> Gumstix Overo board + LPDDR/NAND
>
> <cut><cut><cut>
>
> reading uImage
> mmc_send_cmd: timedout waiting for cmddis!
> mmc fail to send stop cmd
> mmc_send_cmd: timedout waiting for cmddis!
> mmc_send_cmd: timedout waiting for cmddis!
>
> 3152980 bytes read
> ## Booting kernel from Legacy Image at 82000000 ...
>     Image Name:   Linux-2.6.32
>     Image Type:   ARM Linux Kernel Image (uncompressed)
>     Data Size:    3152916 Bytes = 3 MiB
>     Load Address: 80008000
>     Entry Point:  80008000
>     Verifying Checksum ... Bad Data CRC
> ERROR: can't get kernel image!
> </error>
>
> This happens when I reboot the WaterStorm com with the Reset button but boots normally when I remove the power plug and plug back in again. Pressing and holding the Reset button any different amounts of time does not seem to help.
>
> Any help with this is appreciated.
>
> Kind Regards,
>
>
> Ben Keane


Don't know the waterstorm
(when are gumstix going to release them for general purchase ???)
but after updaing the image/uboot did  you clear the uboot environmental
variables ?

check this next line to make sure its correct for waterstorm
(its for overo)

#nand erase 240000 20000
#reset


------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: error booting from MMC on reset

Keane, Ben (STRX)

Hi Alex,

> -----Original Message-----
> From: Alex Gibson [mailto:[hidden email]]
> Sent: Tuesday, 22 November 2011 10:55 AM
> To: General mailing list for gumstix users.
> Subject: Re: [Gumstix-users] error booting from MMC on reset
>
> On 21/11/2011 5:09 PM, Keane, Ben (STRX) wrote:
> > Hello,
> >
> > I was wondering if anyone had any experience with the following error:
> >
> > <error>
> > U-Boot 2011.06-00374-gfa82f87-dirty (Nov 16 2011 - 15:15:28)
> >
> > OMAP3630/3730-GP ES2.1, CPU-OPP2, L3-165MHz, Max CPU Clock 1 Ghz
> > Gumstix Overo board + LPDDR/NAND
> >
> > <cut><cut><cut>
> >
> > reading uImage
> > mmc_send_cmd: timedout waiting for cmddis!
> > mmc fail to send stop cmd
> > mmc_send_cmd: timedout waiting for cmddis!
> > mmc_send_cmd: timedout waiting for cmddis!
> >
> > 3152980 bytes read
> > ## Booting kernel from Legacy Image at 82000000 ...
> >     Image Name:   Linux-2.6.32
> >     Image Type:   ARM Linux Kernel Image (uncompressed)
> >     Data Size:    3152916 Bytes = 3 MiB
> >     Load Address: 80008000
> >     Entry Point:  80008000
> >     Verifying Checksum ... Bad Data CRC
> > ERROR: can't get kernel image!
> > </error>
> >
> > This happens when I reboot the WaterStorm com with the Reset button but
> boots normally when I remove the power plug and plug back in again.
> Pressing and holding the Reset button any different amounts of time does
> not seem to help.
> >
> > Any help with this is appreciated.
> >
> > Kind Regards,
> >
> >
> > Ben Keane
>
>
> Don't know the waterstorm
> (when are gumstix going to release them for general purchase ???)
> but after updaing the image/uboot did  you clear the uboot environmental
> variables ?
>
> check this next line to make sure its correct for waterstorm
> (its for overo)
>
> #nand erase 240000 20000
> #reset
>

I just tested this with a Tobi board instead of the Chestnut43 - It worked fine - I think it's my 1/2 broken Chestnut43 board.  I will wait for my new Chestnut43 board on backorder and confirm.

Thanks for the help,

Ben

PROPRIETARY: This e-mail contains proprietary information some or all of which may be legally privileged.  It is intended for the recipient only.  If an addressing or transmission error has misdirected this e-mail, please notify the authority by replying to this e-mail.  If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this e-mail.

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

JamesAng
In reply to this post by Steve Sakoman
Hi Steve,
sakoman wrote
On Thu, Nov 17, 2011 at 11:04 PM, James <angweiyang@gmail.com> wrote:

> Is it because using "2010.X" results in the use of GCC 4.5 which breaks the
> stuffs above or that Ubuntu 11.10 cause the faults?

I suspect it is 64 bit Ubuntu 11.10 that is causing the issues.  I use
mainly 32 bit Ubuntu 10.04 and do not have these issues.

And since I don't really care for the Ubuntu 11.10 experience I am not
likely to switch to it anytime soon!

Steve
I downloaded the latest files from your server for gnome-r13 and encountered errors during first boot and when loading mt9v032 driver.

From boot message,

omap3_isp: disagrees about version of symbol v4l2_fh_del
omap3_isp: Unknown symbol v4l2_fh_del (err -22)
omap3_isp: disagrees about version of symbol v4l2_i2c_new_subdev_board
omap3_isp: Unknown symbol v4l2_i2c_new_subdev_board (err -22)
omap3_isp: disagrees about version of symbol v4l2_fh_add
omap3_isp: Unknown symbol v4l2_fh_add (err -22)
omap3_isp: disagrees about version of symbol v4l2_device_register_subdev_nodes
omap3_isp: Unknown symbol v4l2_device_register_subdev_nodes (err -22)
omap3_isp: disagrees about version of symbol v4l2_subdev_init
omap3_isp: Unknown symbol v4l2_subdev_init (err -22)
omap3_isp: disagrees about version of symbol v4l2_device_register_subdev
omap3_isp: Unknown symbol v4l2_device_register_subdev (err -22)
omap3_isp: disagrees about version of symbol video_devdata
omap3_isp: Unknown symbol video_devdata (err -22)
omap3_isp: disagrees about version of symbol v4l2_device_unregister_subdev
omap3_isp: Unknown symbol v4l2_device_unregister_subdev (err -22)
omap3_isp: disagrees about version of symbol v4l2_fh_init
omap3_isp: Unknown symbol v4l2_fh_init (err -22)
omap3_isp: disagrees about version of symbol video_unregister_device
omap3_isp: Unknown symbol video_unregister_device (err -22)
omap3_isp: disagrees about version of symbol v4l2_event_subscribe
omap3_isp: Unknown symbol v4l2_event_subscribe (err -22)
omap3_isp: disagrees about version of symbol v4l2_device_register
omap3_isp: Unknown symbol v4l2_device_register (err -22)
omap3_isp: disagrees about version of symbol __video_register_device
omap3_isp: Unknown symbol __video_register_device (err -22)
omap3_isp: disagrees about version of symbol v4l2_device_unregister
omap3_isp: Unknown symbol v4l2_device_unregister (err -22)
omap3_isp: disagrees about version of symbol v4l2_event_queue
omap3_isp: Unknown symbol v4l2_event_queue (err -22)
omap3_isp: disagrees about version of symbol video_device_release_empty
omap3_isp: Unknown symbol video_device_release_empty (err -22)
omap3_isp: disagrees about version of symbol v4l2_event_unsubscribe
omap3_isp: Unknown symbol v4l2_event_unsubscribe (err -22)


After login,

The Angstrom Distribution omap3-multi ttyO2

Angstrom v2011.11 omap3-multi ttyO2

omap3-multi login: root
Password:
root@omap3-multi:~# lsmod
Module                  Size  Used by
bufferclass_ti          4976  0
omaplfb                 8025  0
pvrsrvkm              146868  2 bufferclass_ti,omaplfb
ipv6                  226224  20
libertas_sdio          14871  0
libertas               92472  1 libertas_sdio
cfg80211              157222  1 libertas
media                  11885  0
v4l2_common             8543  0
videodev               77043  1 v4l2_common
lib80211                5291  1 libertas
firmware_class          6269  2 libertas_sdio,libertas
ads7846                10331  0
root@omap3-multi:~# modprobe mt9v032
mt9v032: disagrees about version of symbol v4l2_i2c_subdev_init
mt9v032: Unknown symbol v4l2_i2c_subdev_init (err -22)
mt9v032: disagrees about version of symbol v4l2_device_unregister_subdev
mt9v032: Unknown symbol v4l2_device_unregister_subdev (err -22)
FATAL: Error inserting mt9v032 (/lib/modules/3.0.0/kernel/drivers/media/video/mt9v032.ko): Invalid argument
root@omap3-multi:~# modinfo mt9v032
filename:       /lib/modules/3.0.0/kernel/drivers/media/video/mt9v032.ko
license:        GPL
author:         Laurent Pinchart <laurent.pinchart@ideasonboard.com>
description:    Aptina MT9V032 Camera driver
srcversion:     6B44A05428DC1C76E7BD200
alias:          i2c:mt9v032
depends:        media,v4l2-common,videodev
vermagic:       3.0.0 mod_unload modversions ARMv7
root@omap3-multi:~# media-ctl -p
Opening media device /dev/media0
media_open: Can't open media device /dev/media0
root@omap3-multi:~# uname -a
Linux omap3-multi 3.0.0 #1 Mon Nov 21 08:46:01 PST 2011 armv7l GNU/Linux
root@omap3-multi:~#


The version that I loaded is from the server directory "201111210816"

Can you advise if that version is broken?

Thanks.

James.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

Steve Sakoman
On Mon, Nov 21, 2011 at 11:54 PM, JamesAng <[hidden email]> wrote:

> I downloaded the latest files from your server for gnome-r13 and encountered
> errors during first boot and when loading mt9v032 driver.
>
> >From boot message,
>
> omap3_isp: disagrees about version of symbol v4l2_fh_del

I just used the script to prepare a bootable GNOME r13 image and do
not see that issue, the omap_isp driver loads just fine:

Linux media interface: v0.10
Linux video capture interface: v2.00
omap3isp omap3isp: Revision 2.0 found
omap-iommu omap-iommu.0: isp: version 1.1
mt9v032 3-005c: Probing MT9V032 at address 0x5c
overo: setting xclk to 25000000 hz

Not sure what you did, but seems to work for me!

Steve

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

JamesAng
Hi Steve,
sakoman wrote
On Mon, Nov 21, 2011 at 11:54 PM, JamesAng <angweiyang@gmail.com> wrote:

> I downloaded the latest files from your server for gnome-r13 and encountered
> errors during first boot and when loading mt9v032 driver.
>
> >From boot message,
>
> omap3_isp: disagrees about version of symbol v4l2_fh_del

I just used the script to prepare a bootable GNOME r13 image and do
not see that issue, the omap_isp driver loads just fine:

Linux media interface: v0.10
Linux video capture interface: v2.00
omap3isp omap3isp: Revision 2.0 found
omap-iommu omap-iommu.0: isp: version 1.1
mt9v032 3-005c: Probing MT9V032 at address 0x5c
overo: setting xclk to 25000000 hz

Not sure what you did, but seems to work for me!

Steve
My bad, Sorry. (^^,)"
I downloaded the files directly from the server and unpacked the modules-omap3-multi-XXX.tgz file together with the rootfs.
Files in modules-omap3-multi-XXX.tgz might be the cause of the errors I seen earlier. I found mt9v032.ko within the archive which shouldn't be as I believe mt9v032 must be compiled together during building and not as a module.

BTW, please advise what is the purpose of this modules-omap3-multi-XXX.tgz file as I understand that these are the extra loadable modules that are usually added to the rootfs but packed separately.

James.
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

Steve Sakoman
On Tue, Nov 22, 2011 at 5:17 PM, JamesAng <[hidden email]> wrote:

> My bad, Sorry. (^^,)"
> I downloaded the files directly from the server and unpacked the
> modules-omap3-multi-XXX.tgz file together with the rootfs.

I provide the script to avoid the work (and possible pilot error!) of
doing it by hand.

> Files in modules-omap3-multi-XXX.tgz might be the cause of the errors I seen
> earlier. I found mt9v032.ko within the archive which shouldn't be as I
> believe mt9v032 must be compiled together during building and not as a
> module.

No, the mt9v032 driver is built as a module, i.e. CONFIG_VIDEO_MT9V032=m

> BTW, please advise what is the purpose of this modules-omap3-multi-XXX.tgz
> file as I understand that these are the extra loadable modules that are
> usually added to the rootfs but packed separately.

Despite my advice above to use the script, I sometimes do experiments
and like to have a modules tarball available for manual installs :-)

So it is there for my own use and to allow others to get themselves in
trouble just like I do all too often!

Steve

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

JamesAng
Hi Steve,

On Wed, Nov 23, 2011 at 12:32 PM, Steve Sakoman <[hidden email]> wrote:
On Tue, Nov 22, 2011 at 5:17 PM, JamesAng <[hidden email]> wrote:

> My bad, Sorry. (^^,)"
> I downloaded the files directly from the server and unpacked the
> modules-omap3-multi-XXX.tgz file together with the rootfs.

I provide the script to avoid the work (and possible pilot error!) of
doing it by hand.

I like to learn manually as I goes along and these automated script do come in handy to learn for similar purpose. :)

 
> Files in modules-omap3-multi-XXX.tgz might be the cause of the errors I seen
> earlier. I found mt9v032.ko within the archive which shouldn't be as I
> believe mt9v032 must be compiled together during building and not as a
> module.

No, the mt9v032 driver is built as a module, i.e. CONFIG_VIDEO_MT9V032=m

Hmm.. So the rootfs already has a mt9v032 driver included or not?
It doesn't explain why that versioning error came about if both the modules tarball is the same as included in the rootfs tarball.

I though all modules are packed separately in these modules-XXX.tgz file and only unpacked separately on top of the rootfs.
 
> BTW, please advise what is the purpose of this modules-omap3-multi-XXX.tgz
> file as I understand that these are the extra loadable modules that are
> usually added to the rootfs but packed separately.

Despite my advice above to use the script, I sometimes do experiments
and like to have a modules tarball available for manual installs :-) 
So it is there for my own use and to allow others to get themselves in
trouble just like I do all too often!

Steve


Understood..
Getting into trouble is part of learning. :)
Forgive me if I do offended anyone..

Your image is good for developing natively on top of the machine and board and thus I like to be able to build yours from you repo. But I keep banging onto walls..

BTW, other than Laurent's yavta application to test mt9v032, do you have success in doing so with either mplayer or gst-launch to display streaming video on screen?

I saw patches around in linux-media since the acceptance of MCF into mainline v2.6.39 patching several of omap modules.

Many thanks in adv.
--
Regards,
James

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

Steve Sakoman
On Wed, Nov 23, 2011 at 12:45 AM, James <[hidden email]> wrote:

>> No, the mt9v032 driver is built as a module, i.e. CONFIG_VIDEO_MT9V032=m
>
> Hmm.. So the rootfs already has a mt9v032 driver included or not?

Yes, of course it is included in the rootfs!

> It doesn't explain why that versioning error came about if both the modules
> tarball is the same as included in the rootfs tarball.

Again, I have no idea what manual process you used, so I can't offer
any theories.

I only test that the script results in a working system.

> Understood..
> Getting into trouble is part of learning. :)
> Forgive me if I do offended anyone..

Not offended at all!

> Your image is good for developing natively on top of the machine and board
> and thus I like to be able to build yours from you repo. But I keep banging
> onto walls..

The images and repositories are something I use daily in my own work.

I make them available for others to use, but there is no way I can
provide free support for OE issues!

> BTW, other than Laurent's yavta application to test mt9v032, do you have
> success in doing so with either mplayer or gst-launch to display streaming
> video on screen?

No, but I haven't really tried to get mplayer or gst-launch working so
it very well may be possible.  I just don't have any personal
experience.

Steve

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

JamesAng
Hi Steve,

On Wed, Nov 23, 2011 at 10:45 PM, Steve Sakoman <[hidden email]> wrote:
On Wed, Nov 23, 2011 at 12:45 AM, James <[hidden email]> wrote:

>> No, the mt9v032 driver is built as a module, i.e. CONFIG_VIDEO_MT9V032=m
>
> Hmm.. So the rootfs already has a mt9v032 driver included or not?

Yes, of course it is included in the rootfs!

> It doesn't explain why that versioning error came about if both the modules
> tarball is the same as included in the rootfs tarball.

Again, I have no idea what manual process you used, so I can't offer
any theories.

I followed the step-by-step for creating a bootable SDcard as per both Gumstix and your site.

After doing the bootable partition related stuffs.
I unpacked the rootfs tarball by "sudo tar xvaf <rootfs-tarball> -C /media/rootfs"
Follow by, "sudo tar xvaf <modules-tarball> -C /media/rootfs"

That is only the additional step that I execute to place those modules files onto the file-system.

Unless, I miss understood the steps/purpose about unpacking and adding modules to the file-system.

My understanding from Gumstix's wiki is that all kernel's drivers configuration marked as <m>odules will be packed separately and I've to add them manually to the file-system, if I want to use them.

But, your saying that MT9V032 is marked as module and yet it's included into the rootfs-tarball further confused me.. I verified that it is in the rootfs-tarball.

> Getting into trouble is part of learning. :)
> Forgive me if I do offended anyone..

> Understood..
Not offended at all!

Phew.. I was worried that my consistent questioning will offends the community.
 
> Your image is good for developing natively on top of the machine and board
> and thus I like to be able to build yours from you repo. But I keep banging
> onto walls..

The images and repositories are something I use daily in my own work.

I make them available for others to use, but there is no way I can
provide free support for OE issues!

Understood. Thanks a lot.
 
> BTW, other than Laurent's yavta application to test mt9v032, do you have
> success in doing so with either mplayer or gst-launch to display streaming
> video on screen?

No, but I haven't really tried to get mplayer or gst-launch working so
it very well may be possible.  I just don't have any personal
experience.

Steve

Who at Gumstix, if any, is actually responsible for providing the Caspa related work?
With your new v3.0 kernel, the Caspa's wiki is no longer relevant.

At my side, I tried both mplayer and gst-launch but unable to get it to work.
Attached is the resultant log that I captured for using mplayer on your gnome-r13 image.

BTW, what is the version of mplayer in your image? it stated "UNKNOWN-4.5.3"

Mplayer status highlighted "0 frame processed" and the console gave a section of code traces.

yavta has verified that the camera is working.

Hope you can route this to the personal at Gumstix who deals with Caspa project and see if there is a solution to this.

Many thanks in adv.

--
Regards,
James

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users

mplayer-result.log (11K) Download Attachment
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

Steve Sakoman
On Wed, Nov 23, 2011 at 6:40 PM, James <[hidden email]> wrote:

> I followed the step-by-step for creating a bootable SDcard as per both
> Gumstix and your site.
>
> After doing the bootable partition related stuffs.
> I unpacked the rootfs tarball by "sudo tar xvaf <rootfs-tarball> -C
> /media/rootfs"
> Follow by, "sudo tar xvaf <modules-tarball> -C /media/rootfs"

You won't find that last step on any instructions that I wrote!

> That is only the additional step that I execute to place those modules files
> onto the file-system.

It is not needed.

> My understanding from Gumstix's wiki is that all kernel's drivers
> configuration marked as <m>odules will be packed separately and I've to add
> them manually to the file-system, if I want to use them.

It is true that you have to include them in your *image* recipe (or in
a conf file).

My builds include all required modules in the image creation, so you
don't need to do anything further.

> Who at Gumstix, if any, is actually responsible for providing the Caspa
> related work?

I'll let the Gumstix folks answer that question.

> BTW, what is the version of mplayer in your image? it stated "UNKNOWN-4.5.3"

The image uses the mplayer_svn recipe which uses svn r32735.

> Hope you can route this to the personal at Gumstix who deals with Caspa
> project and see if there is a solution to this.

AFAIK the Gumstix folks read this list.

Steve

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: Building Sakoman's GNOME-R13 image

Neil MacMunn
In reply to this post by JamesAng
Hi James,

> With your new v3.0 kernel, the Caspa's wiki is no longer relevant.

You are right.. the instructions provided on the wiki pertain to the
2.6.34 kernel only.

The latest images have a working driver but, last I checked, no user
space support other than yavta. If you find any information that might
help other users, feel free to add to the wiki.

-
Neil

------------------------------------------------------------------------------
All the data continuously generated in your IT infrastructure
contains a definitive record of customers, application performance,
security threats, fraudulent activity, and more. Splunk takes this
data and makes sense of it. IT sense. And common sense.
http://p.sf.net/sfu/splunk-novd2d
_______________________________________________
gumstix-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gumstix-users
12
Loading...