> I want to run WebGL application on
> Ubuntu 12.04 on pandaboard ES Rev B1.

> It move smooth but there are some artifacts.

> * Environment
> Image: http://cdimage.ubuntu.com/releases/12.04/release/ubuntu-12.04-preinst...
> Install TI OMAP Release PPA Update October 2012,
> Install chromium-browser

> $ chromium-browser --use-gl=egl
> Go to the page
> http://mrdoob.github.com/three.js/examples/webgl_materials_cubemap_ba...

> * Problem
> Sometimes a ball 'A' which located behind another ball 'B'
> shows up in front of 'B'.

> I tried to use graphic driver via Jockey,
> not add TI OMAP PPA,
> but there are strange balls too.

> Any ideas what I'm missing?

> Cheers,
> Kumadasu

> Hi, Nicolas

> Thank you for your information.

> I have finished to install 12.10.  But I couldn't find
> driver for SGX.  I installed jockey-common
> and run jockey-text.
> But nothing is occur.

> I have not tried TI OMAP PPA,
> there is no TI OMAP PPA for quantal isn't it.

> Now I'm trying to rebuild chromium22 for precise.
> It takes some hour.  I'll report here after finishing rebuild.
> I hope no configuration is needed.
> I don't know about porting ;-)

> 2012/10/26 Nicolas Dechesne <nde...@gmail.com>:
> > On Thu, Oct 25, 2012 at 1:41 PM, Kumadasu <kumad...@gmail.com> wrote:
> >> Hi, all

> >> I want to run WebGL application on
> >> Ubuntu 12.04 on pandaboard ES Rev B1.

> >> It move smooth but there are some artifacts.

> >> * Environment
> >> Image: http://cdimage.ubuntu.com/releases/12.04/release/ubuntu-12.04-preinst...
> >> Install TI OMAP Release PPA Update October 2012,
> >> Install chromium-browser

> >> $ chromium-browser --use-gl=egl
> >> Go to the page
> >> http://mrdoob.github.com/three.js/examples/webgl_materials_cubemap_ba...

> >> * Problem
> >> Sometimes a ball 'A' which located behind another ball 'B'
> >> shows up in front of 'B'.

> >> I tried to use graphic driver via Jockey,
> >> not add TI OMAP PPA,
> >> but there are strange balls too.

> >> Any ideas what I'm missing?

> > the chrome browser is the ubuntu archives is fairly old, and that's
> > more or less the problem. we did a build from upstream trunk recently
> > (v22) and WebGL is working much better, even performance wise. the
> > problems we were chasing with v20 + SGX really went away with v22
> > without any change in the GFX driver...

> > i can't tell for sure that all your problems would be fixed though...
> > but doing webgl on old chrome is pretty much a bad idea...

> > v22 was pushed in 12.10 recently, so you can try a 12.10 image with
> > the jokey driver first. and btw I would be interested by your
> > conclusion...

> > then if you need 12.04, you can attempt a rebuild of v22 into 12.04. I
> > have spent a little bit of time on that and wasn't too much
> > successful... and building chrome browser is not something you really
> > want to do too often ;-)

> >> Cheers,
> >> Kumadasu

>> I want to run WebGL application on
>> Ubuntu 12.04 on pandaboard ES Rev B1.

>> It move smooth but there are some artifacts.

>> * Environment
>> Image: http://cdimage.ubuntu.com/releases/12.04/release/ubuntu-12.04-preinst...
>> Install TI OMAP Release PPA Update October 2012,
>> Install chromium-browser

>> $ chromium-browser --use-gl=egl
>> Go to the page
>> http://mrdoob.github.com/three.js/examples/webgl_materials_cubemap_ba...

>> * Problem
>> Sometimes a ball 'A' which located behind another ball 'B'
>> shows up in front of 'B'.

>> I tried to use graphic driver via Jockey,
>> not add TI OMAP PPA,
>> but there are strange balls too.

>> Any ideas what I'm missing?

> the chrome browser is the ubuntu archives is fairly old, and that's
> more or less the problem. we did a build from upstream trunk recently
> (v22) and WebGL is working much better, even performance wise. the
> problems we were chasing with v20 + SGX really went away with v22
> without any change in the GFX driver...

> i can't tell for sure that all your problems would be fixed though...
> but doing webgl on old chrome is pretty much a bad idea...

> v22 was pushed in 12.10 recently, so you can try a 12.10 image with
> the jokey driver first. and btw I would be interested by your
> conclusion...

> then if you need 12.04, you can attempt a rebuild of v22 into 12.04. I
> have spent a little bit of time on that and wasn't too much
> successful... and building chrome browser is not something you really
> want to do too often ;-)

>> Cheers,
>> Kumadasu

> I tried to rebuild chromium 22 for panda precise.
> It failed, probably my instruction is wrong or
> cannot rebuild without some modification.

> The instruction and error log is below.
> Do you have any suggestion?

> BTW should I ask this kind of question to Ubuntu Mailing list ?

> * Instruction
> Download .orig.tar.gz, .diff.gz, .dsc from here.
> https://launchpad.net/ubuntu/+source/chromium-browser/22.0.1229.94~r1...

> $ dpkg-source -x chromium-browser_22.0.1229.94~r161065-0ubuntu1.dsc
> $ pbuilder-dist precise armhf create
> $ pbuilder-dist precise armhf build
> chromium-browser_22.0.1229.94~r161065-0ubuntu1.dsc

> * log
> $ tail -n20 pbuilder/precise-armhf_result/last_operation.log
>   AR(target) out/Release/obj.target/third_party/libvpx/libvpx_asm_offsets.a
>   LINK(host) out/Release/libvpx_obj_int_extract
>   AR(target) out/Release/obj.target/remoting/libremoting_jingle_glue.a
>   AR(target) out/Release/obj.target/jingle/libjingle_glue.a
>   AR(target) out/Release/obj.target/third_party/libjingle/libjingle.a
>   AR(target) out/Release/obj.target/third_party/jsoncpp/libjsoncpp.a
> /usr/bin/ld.gold.real: fatal error:
> out/Release/libvpx_obj_int_extract: Invalid argument
> collect2: ld returned 1 exit status
> make[1]: *** [out/Release/libvpx_obj_int_extract] Error 1
> make[1]: *** Waiting for unfinished jobs....
>   AR(target) out/Release/obj.target/third_party/libjingle/libjingle_p2p.a
> make[1]: Leaving directory
> `/tmp/buildd/chromium-browser-22.0.1229.94~r161065/build-tree/src'
> make: *** [debian/stamp-makefile-build] Error 1
> dpkg-buildpackage: error: debian/rules build gave error exit status 2
> E: Failed autobuilding of package
> I: unmounting /var/cache/pbuilder/ccache filesystem
> I: unmounting dev/pts filesystem
> I: unmounting proc filesystem
> I: cleaning the build env
> I: removing directory /var/cache/pbuilder/build//3040 and its subdirectories

> Cheers,
> Kumadasu

> 2012/10/26 21:43 "Kumadasu" <kumad...@gmail.com>:

>> Hi, Nicolas

>> Thank you for your information.

>> I have finished to install 12.10.  But I couldn't find
>> driver for SGX.  I installed jockey-common
>> and run jockey-text.
>> But nothing is occur.

>> I have not tried TI OMAP PPA,
>> there is no TI OMAP PPA for quantal isn't it.

>> Now I'm trying to rebuild chromium22 for precise.
>> It takes some hour.  I'll report here after finishing rebuild.
>> I hope no configuration is needed.
>> I don't know about porting ;-)

>> 2012/10/26 Nicolas Dechesne <nde...@gmail.com>:
>> > On Thu, Oct 25, 2012 at 1:41 PM, Kumadasu <kumad...@gmail.com> wrote:
>> >> Hi, all

>> >> I want to run WebGL application on
>> >> Ubuntu 12.04 on pandaboard ES Rev B1.

>> >> It move smooth but there are some artifacts.

>> >> * Environment
>> >> Image: http://cdimage.ubuntu.com/releases/12.04/release/ubuntu-12.04-preinst...
>> >> Install TI OMAP Release PPA Update October 2012,
>> >> Install chromium-browser

>> >> $ chromium-browser --use-gl=egl
>> >> Go to the page
>> >> http://mrdoob.github.com/three.js/examples/webgl_materials_cubemap_ba...

>> >> * Problem
>> >> Sometimes a ball 'A' which located behind another ball 'B'
>> >> shows up in front of 'B'.

>> >> I tried to use graphic driver via Jockey,
>> >> not add TI OMAP PPA,
>> >> but there are strange balls too.

>> >> Any ideas what I'm missing?

>> > the chrome browser is the ubuntu archives is fairly old, and that's
>> > more or less the problem. we did a build from upstream trunk recently
>> > (v22) and WebGL is working much better, even performance wise. the
>> > problems we were chasing with v20 + SGX really went away with v22
>> > without any change in the GFX driver...

>> > i can't tell for sure that all your problems would be fixed though...
>> > but doing webgl on old chrome is pretty much a bad idea...

>> > v22 was pushed in 12.10 recently, so you can try a 12.10 image with
>> > the jokey driver first. and btw I would be interested by your
>> > conclusion...

>> > then if you need 12.04, you can attempt a rebuild of v22 into 12.04. I
>> > have spent a little bit of time on that and wasn't too much
>> > successful... and building chrome browser is not something you really
>> > want to do too often ;-)

>> >> Cheers,
>> >> Kumadasu

> Or where should I start to investigate from?
> I want to help development, and run new chromium!!

> good. same status here ;-)

>> Or where should I start to investigate from?
>> I want to help development, and run new chromium!!

> if you can debug that specific issue, it would be really nice. iirc,
> it segfaults very quickly, there must be a build or lib depends
> issues...

> let us know

> I'm not good at debugging.
> I sent bug report and I checked backtrace.
> I don't know where I sent the report to, you or someone else.
> I'll send you crash
> log(/var/crash/_usr_lib_chromium-browser_chromium-browser.1000.crash)
> after this email.

> It is similar to the post on Oct 15, 2012 by robe.
> http://code.google.com/p/chromium/wiki/LinuxChromiumArm

> He or she said that is caused by gold bug.

> The result in my case is like below.

> ***@***:~$ apport-retrace -g
> /var/crash/_usr_lib_chromium-browser_chromium-browser.1000.crash
> GNU gdb (Ubuntu/Linaro 7.4-2012.04-0ubuntu2) 7.4-2012.04
> Copyright (C) 2012 Free Software Foundation, Inc.
> License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
> This is free software: you are free to change and redistribute it.
> There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
> and "show warranty" for details.
> This GDB was configured as "arm-linux-gnueabihf".
> For bug reporting instructions, please see:
> <http://bugs.launchpad.net/gdb-linaro/>.
> Reading symbols from
> /usr/lib/chromium-browser/chromium-browser...Reading symbols from
> /usr/lib/debug/usr/lib/chromium-browser/chromium-browser...(no
> debugging symbols found)...done.
> (no debugging symbols found)...done.
> warning: core file may not match specified executable file.
> [New LWP 10533]
> [Thread debugging using libthread_db enabled]
> Using host libthread_db library "/lib/arm-linux-gnueabihf/libthread_db.so.1".
> Core was generated by `/usr/lib/chromium-browser/chromium-browser'.
> Program terminated with signal 11, Segmentation fault.
> #0  0x0020d978 in ?? ()
> (gdb) bt
> #0  0x0020d978 in ?? ()
> #1  0xb611542e in __libc_csu_init ()
> #2  0xb37ecf98 in __libc_start_main () from /lib/arm-linux-gnueabihf/libc.so.6
> #3  0xb45704da in _start ()
> (gdb)

> 2012/11/1 Nicolas Dechesne <nde...@gmail.com>:
>> On Wed, Oct 31, 2012 at 3:08 PM, Kumadasu <kumad...@gmail.com> wrote:
>>> I got segmentation fault, no window.
>>> Anybody could run it?

>> good. same status here ;-)

>>> Or where should I start to investigate from?
>>> I want to help development, and run new chromium!!

>> if you can debug that specific issue, it would be really nice. iirc,
>> it segfaults very quickly, there must be a build or lib depends
>> issues...

>> let us know

>> I'm not good at debugging.
>> I sent bug report and I checked backtrace.
>> I don't know where I sent the report to, you or someone else.
>> I'll send you crash
>> log(/var/crash/_usr_lib_chromium-browser_chromium-browser.1000.crash)
>> after this email.

>> It is similar to the post on Oct 15, 2012 by robe.
>> http://code.google.com/p/chromium/wiki/LinuxChromiumArm

>> He or she said that is caused by gold bug.

>> The result in my case is like below.

>> ***@***:~$ apport-retrace -g
>> /var/crash/_usr_lib_chromium-browser_chromium-browser.1000.crash
>> GNU gdb (Ubuntu/Linaro 7.4-2012.04-0ubuntu2) 7.4-2012.04
>> Copyright (C) 2012 Free Software Foundation, Inc.
>> License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
>> This is free software: you are free to change and redistribute it.
>> There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
>> and "show warranty" for details.
>> This GDB was configured as "arm-linux-gnueabihf".
>> For bug reporting instructions, please see:
>> <http://bugs.launchpad.net/gdb-linaro/>.
>> Reading symbols from
>> /usr/lib/chromium-browser/chromium-browser...Reading symbols from
>> /usr/lib/debug/usr/lib/chromium-browser/chromium-browser...(no
>> debugging symbols found)...done.
>> (no debugging symbols found)...done.
>> warning: core file may not match specified executable file.
>> [New LWP 10533]
>> [Thread debugging using libthread_db enabled]
>> Using host libthread_db library "/lib/arm-linux-gnueabihf/libthread_db.so.1".
>> Core was generated by `/usr/lib/chromium-browser/chromium-browser'.
>> Program terminated with signal 11, Segmentation fault.
>> #0  0x0020d978 in ?? ()
>> (gdb) bt
>> #0  0x0020d978 in ?? ()
>> #1  0xb611542e in __libc_csu_init ()
>> #2  0xb37ecf98 in __libc_start_main () from /lib/arm-linux-gnueabihf/libc.so.6
>> #3  0xb45704da in _start ()
>> (gdb)

> It was a while ago that I last built chromium, but I think this was
> the error that I was seeing until I disabled neon.. I don't really
> know what the root issue was but somehow having neon enabled was
> resulting in something getting corrupted at startup.

> BR,
> -R

>> 2012/11/1 Nicolas Dechesne <nde...@gmail.com>:
>>> On Wed, Oct 31, 2012 at 3:08 PM, Kumadasu <kumad...@gmail.com> wrote:
>>>> I got segmentation fault, no window.
>>>> Anybody could run it?

>>> good. same status here ;-)

>>>> Or where should I start to investigate from?
>>>> I want to help development, and run new chromium!!

>>> if you can debug that specific issue, it would be really nice. iirc,
>>> it segfaults very quickly, there must be a build or lib depends
>>> issues...

>>> let us know

>> I greped and I found one 'arm_neon=0' in debian/rules.
>> Is it enough?  Or have to add somewhere other 'arm_neon=0'?

> that should be.. fwiw, I used:

> GYP_DEFINES="chromeos=1 target_arch=arm sysroot=/chroot/precise-armhf
> disable_nacl=1 linux_use_tcmalloc=0 armv7=1 arm_thumb=1 arm_neon=0
> arm_float_abi=hard v8_use_arm_eabi_hardfloat=true "

> but I didn't use the deb build mechanism.. I had just built manually.

> BR,
> -R

>> kumadasu@kumadasu-MBA:~/cross/chromium22ti/chromium-browser-22.0.1229.79~r1 58531$
>> ls ..
>> chromium-browser-22.0.1229.79~r158531
>> chromium-browser_22.0.1229.79~r158531-0ubuntu1+ti1.diff.gz
>> chromium-browser_22.0.1229.79~r158531-0ubuntu1%2Bti1.dsc
>> chromium-browser_22.0.1229.79~r158531.orig.tar.gz
>> kumadasu@kumadasu-MBA:~/cross/chromium22ti/chromium-browser-22.0.1229.79~r1 58531$
>> grep -3 -r "neon" .
>> ./debian/rules-ifneq (,$(findstring arm,$(DEB_BUILD_ARCH)))
>> ./debian/rules-GYP_DEFINES += \
>> ./debian/rules- arm_thumb=1 \
>> ./debian/rules: arm_neon=0 \
>> ./debian/rules- arm_fpu=vfpv3-d16 \
>> ./debian/rules- target_arch=arm \
>> ./debian/rules- disable_nacl=1 \
>> --
>> ./debian/copyright-Copyright: Copyright: 2007 Siarhei Siamashka
>> <s...@users.sourceforge.net>:
>> ./debian/copyright-License: LGPL (v2.1 or later)
>> ./debian/copyright-
>> ./debian/copyright:Files:
>> src/third_party/ffmpeg/source/patched-ffmpeg-mt/libavcodec/arm/{dsputil_neo n.c}
>> ./debian/copyright-Copyright: Copyright: 2008 Mans Rullgard <m...@mansr.com>:
>> ./debian/copyright-License: LGPL (v2.1 or later)
>> ./debian/copyright-
>> --
>> ./debian/changelog-  * Promote Uyghur to the list of supported translations
>> ./debian/changelog-    - update debian/rules
>> ./debian/changelog-    - update debian/control
>> ./debian/changelog:  * Fix the FTBFS on arm by re-adding the lost
>> arm_neon=0, and really set armv7=1
>> ./debian/changelog-    on maverick and natty
>> ./debian/changelog-    - update debian/rules
>> ./debian/changelog-  * Fix the broken symlinks in /usr/share/doc
>> created by CDBS (See LP: #194574)
>> kumadasu@kumadasu-MBA:~/cross/chromium22ti/chromium-browser-22.0.1229.79~r1 58531$

>> 2012/11/6 Rob Clark <robdcl...@gmail.com>:
>>> On Sat, Nov 3, 2012 at 12:50 PM, Kumadasu <kumad...@gmail.com> wrote:
>>>> Hi Nicolas

>>>> I'm not good at debugging.
>>>> I sent bug report and I checked backtrace.
>>>> I don't know where I sent the report to, you or someone else.
>>>> I'll send you crash
>>>> log(/var/crash/_usr_lib_chromium-browser_chromium-browser.1000.crash)
>>>> after this email.

>>>> It is similar to the post on Oct 15, 2012 by robe.
>>>> http://code.google.com/p/chromium/wiki/LinuxChromiumArm

>>>> He or she said that is caused by gold bug.

>>>> The result in my case is like below.

>>>> ***@***:~$ apport-retrace -g
>>>> /var/crash/_usr_lib_chromium-browser_chromium-browser.1000.crash
>>>> GNU gdb (Ubuntu/Linaro 7.4-2012.04-0ubuntu2) 7.4-2012.04
>>>> Copyright (C) 2012 Free Software Foundation, Inc.
>>>> License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
>>>> This is free software: you are free to change and redistribute it.
>>>> There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
>>>> and "show warranty" for details.
>>>> This GDB was configured as "arm-linux-gnueabihf".
>>>> For bug reporting instructions, please see:
>>>> <http://bugs.launchpad.net/gdb-linaro/>.
>>>> Reading symbols from
>>>> /usr/lib/chromium-browser/chromium-browser...Reading symbols from
>>>> /usr/lib/debug/usr/lib/chromium-browser/chromium-browser...(no
>>>> debugging symbols found)...done.
>>>> (no debugging symbols found)...done.
>>>> warning: core file may not match specified executable file.
>>>> [New LWP 10533]
>>>> [Thread debugging using libthread_db enabled]
>>>> Using host libthread_db library "/lib/arm-linux-gnueabihf/libthread_db.so.1".
>>>> Core was generated by `/usr/lib/chromium-browser/chromium-browser'.
>>>> Program terminated with signal 11, Segmentation fault.
>>>> #0  0x0020d978 in ?? ()
>>>> (gdb) bt
>>>> #0  0x0020d978 in ?? ()
>>>> #1  0xb611542e in __libc_csu_init ()
>>>> #2  0xb37ecf98 in __libc_start_main () from /lib/arm-linux-gnueabihf/libc.so.6
>>>> #3  0xb45704da in _start ()
>>>> (gdb)

>>> It was a while ago that I last built chromium, but I think this was
>>> the error that I was seeing until I disabled neon.. I don't really
>>> know what the root issue was but somehow having neon enabled was
>>> resulting in something getting corrupted at startup.

>>> BR,
>>> -R

>>>> 2012/11/1 Nicolas Dechesne <nde...@gmail.com>:
>>>>> On Wed, Oct 31, 2012 at 3:08 PM, Kumadasu <kumad...@gmail.com> wrote:
>>>>>> I got segmentation fault, no window.
>>>>>> Anybody could run it?

>>>>> good. same status here ;-)

>>>>>> Or where should I start to investigate from?
>>>>>> I want to help development, and run new chromium!!

>>>>> if you can debug that specific issue, it would be really nice. iirc,
>>>>> it segfaults very quickly, there must be a build or lib depends
>>>>> issues...

>>>>> let us know