Re: [yocto] Intel(R) Core(TM) i7-3612QM CPU @ 2.10GHz

2013-09-03 Thread Eren Türkay
On Mon, Sep 02, 2013 at 10:20:41PM -0700, Sean Liming wrote:
 It has been awhile since I looked this one up. I believe the installer
 script is called init-install.sh under the /meta/recipes-core. If you don't
 use a serial debug port with a terminal to catch the boot menu, you will
 have to modify the syslinux.cfg file (or grub.cfg for EFI boot) to run the
 install option rather than boot option.

I believe that you do not have to edit syslinux.cfg for boot and
install option. live image type adds boot and install labels to
syslinux.cfg while creating the image. When you add live to
IMAGE_FSTYPES, image-live.bbclass is run and it sets related labels.

Regards,
Eren

-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgpMe1O4HRDTa.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


Re: [yocto] [FIXED] Partitioned Image Creation and Initramfs Problem for ALIX3D3

2013-09-02 Thread Eren Türkay
On Thu, Aug 29, 2013 at 06:20:14PM +0200, Eren Türkay wrote:
 Hello all,

Hi again,

 The possible solution that came to my mind is to get
 boot-directdisk.bbclass, add initramfs installation, put the code
 in a seperate bbclass, rename the function as IMAGE_CMD_alix-hddimage,
 and add alix-hddimage to IMAGE_FSTYPES. The same technique is used by
 meta-raspberrypi layer [1]

 However, the problem with this is that core-image-minimal-initramfs
 includes initramfs-live-boot by default. This script tries to find
 rootfs.img and if it cannot find, it fails to boot the board. In this
 case, we will have extracted rootfs in /dev/sda2, so rootfs.img will not
 be there.
 
 I looked at initramfs-framework recipe and it seems to fit for it. For
 initramfs problem, should I create alix3d3-minimal-initramfs recipe,
 add initramfs-framework-base initramfs-module-udev to IMAGE_INSTALL
 and make IMAGE_CMD_alix-hddimage depend on
 alix3d3-minimal-initramfs? Would it be a proper solution?

I applied the solution here and the generated image works as expected. I
needed to create seperate initramfs image for the board, and have a
bbclass file for image creation (IMAGE_CMD_alix-hddimage). This bbclass
for image creation is the mix of boot-directdisk.bbclass and
sdimage_rpi.

https://github.com/eren/meta-alix3d3

Initramfs image is in recipes-core/images, and the bbclass file is in
classes directory. For those who are interested, taking a look at
conf/machine/alix3d3.conf would be helpful as well.

Cheers,
Eren

-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgpSXZVe_47Z1.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


[yocto] ALIX3D3 Live Image Waiting for removable media Problem

2013-08-08 Thread Eren Türkay
Hello,

I am working on ALIX3D3 [0] BSP [1]. It is an x86-compatible PC with AMD
Geode LX Processor. It boots from CF Card and has Award BIOS.

I enabled Geode specific features in the kernel (CS5536 PATA, Sound,
Geode Framebuffer, etc) and I produce live image so that I can
directly dd it into CF Card and boot. However, when I boot it, I see
waiting for removable media and cannot go forward. In the dmesg
output, I see that all geode specific features are OK. The kernel probes
CF Card and finds it as sda, I can see sound card and framebuffer
works as expected. However, udevadm complains about not finding
/dev/sda*

I have looked at how live image was generated and saw that it creates an
msdos file system, not another volume. So, I dig into other bootable
image and found out that vmdk image actually creates .hdddirect with
valid partitions. I tried with that image type, and dd'ed .hdddirect
image. In the end, it complained about not finding initrd as the boot
partition of .hdddirect only includes syslinux and vmlinuz, not
initrd.

I would appreciate any help on deploying these image on the hardware. It
seems that having initrd in .hdddirect would solve the problem but I'm
not sure how I can accomplish this.

= Bitbake Output =
Build Configuration:
BB_VERSION= 1.18.0
BUILD_SYS = x86_64-linux
NATIVELSBSTRING   = Ubuntu-12.04
TARGET_SYS= i586-poky-linux
MACHINE   = alix3d3
DISTRO= poky
DISTRO_VERSION= 1.4.1
TUNE_FEATURES = m32 i586
TARGET_FPU= 
meta
meta-yocto
meta-yocto-bsp= dylan:e4a0a4c5e419154a34710a1b6c28d4180c6304c3
meta-hamradio = master:243f71144a3fb9a1dbeb81cfc6c46157627b5aa8
meta-alix3d3  = master:46afebe021f259206941445a643afe2c6dbc0d2d
= End: Bitbake Output =

Regards,
Eren

[0] http://pcengines.ch/alix3d3.htm
[1] https://github.com/eren/meta-alix3d3
-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgp0AuCaE6CHI.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


Re: [yocto] [FIXED] ALIX3D3 Live Image Waiting for removable media Problem

2013-08-08 Thread Eren Türkay
On Thu, Aug 08, 2013 at 03:13:49PM +0200, Eren Türkay wrote:
 output, I see that all geode specific features are OK. The kernel probes
 CF Card and finds it as sda, I can see sound card and framebuffer
 works as expected. However, udevadm complains about not finding
 /dev/sda*

After hours of reading udev output and trying to figure out the problem,
I learned that udev requires CONFIG_DEVTMPFS to work properly. After
adding the following fragments to my kernel configuration, I was able to
boot live image properly.

CONFIG_DEVTMPFS=y
CONFIG_DEVTMPFS_MOUNT=y

Message to the future: I hope this information will be beneficial for
anyone who is struggling with that problem.

Best regards,
Eren

-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgp8dHD8847jj.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


Re: [yocto] Ubuntu 11.10 to 12.04 upgrade

2013-07-11 Thread Eren Türkay
On Thu, Jul 11, 2013 at 11:49:47AM +0100, Chris Tapp wrote:
 My current build system is an Ubuntu 11.10 64-bit (server). This is
 now out of support, so I plan to upgrade to 12.04 (LTS).

Hello,

 Is this likely to go smoothly, or are there some gotchas to worry
 about? I'm currently building for Poky 8 if that makes a difference.

I am using 12.04 without a problem. Once you install build-essential and
other required packages by bitbake, there shouldn't be any problem.

 Thanks!
 Chris Tapp

Regards,
Eren

-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgp_01M0P1B4t.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


Re: [yocto] OE/Yocto Talk in Free Software and Linux Days in Turkey

2013-03-30 Thread Eren Türkay
On Thu, Mar 28, 2013 at 04:52:29PM -0400, Trevor Woerner wrote:
 If I had to create a talk I'd probably spend no more than 40% of the
 time showing them how neat Yocto is in that you can add various
 packages to an image and how the package list is independent of the
 target CPU. For example, you can have 2 different devices, but build
 an image for both that contains the same packages and all that's
 needed is a one-line, local, configuration change. I'd then spend the
 rest of the talk discussing the SDK that can be generated and deployed
 to the various team members. Then, if a package is added, a new SDK is
 generated, the team members install the new SDK, update their
 environment, and voila! they can now build an application that uses
 sqlite3 (for example).

Thanks for your suggestion! You are right about SDK and most people in
the development team would normally write applications for the board
using the SDK.

I will probably allocate 50% of the talk to SDK after explaining what
Yocto is, and how we can use it to build a root image. I'm thinking of
showing how the kernel/fs guy can work, and how other team members can
use the image easily.

 There are many previous Yocto/OE talks from which you can draw
 inspiration. free-electrons.com has archives of many such talks from
 various embedded linux conferences. These archives include videos as
 well as slides. Maybe something in that pile can help you get ready?
 
 E.g. http://free-electrons.com/blog/elc-2012-videos/

Thank you for the links. I also got yocto project presentations.

Regards,
Eren

-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgpTwlmqSG3o_.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


[yocto] OE/Yocto Talk in Free Software and Linux Days in Turkey

2013-03-28 Thread Eren Türkay
Hello,

Free software and Linux days [0] is held annually by Istanbul Bilgi
University [1] and Turkish Linux Users Association [2] for 12 years. It
is one of the most well known events, which will be on 5-6 April 2013
this year. My application has been accepted and I will be giving talk
about OE/Yocto.  My seminar title is: Building an Embedded Linux
Distribution using OpenEmbedded and Yocto

The time allocated for the seminar is 45-minutes. The audience generally
has a basic knowledge of linux and they are eager to learn new topics.

Since OE/Yocto is not well known in Turkey, I thought it would be better
to talk about the basics about OE/Yocto. I will give an overview
throughout the seminar, and show how to grab a layer and start building
the distribution.

My draft list of topics to include are:

- What is an embedded system?
- How embedded linux systems are produced? (traditional way of creating
  cross-toolchain, using buildroot, scratchbox, etc)
- What are the disadvantages of these methods? How can it be done
  differently?
- Introduction to OpenEmbedded and Yocto
- What problems does it solve. Advantages/Disadvantages.
- By whom and how is it developed?
- The structure of OpenEmbedded
- How to create an image using OpenEmbedded

I haven't prepared slides and I have one week until the event.  I would
really appreciate a previous seminar materials and slides. I would be
grateful if we can discuss further on items in detail (e.g: what to
explain). If you were to give a talk about OE, what would you explain?
:)

Thanks for your help!

[0] http://www.ozguryazilimgunleri.org.tr/2013/
[1] http://www.bilgi.edu.tr/en/
[2] http://www.lkd.org.tr/en/

Regards,
Eren

-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgpAQld3WMrZg.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


Re: [yocto] Where does the bitbake get the variable TOPDIR

2012-12-18 Thread Eren Türkay
On Tue, Dec 18, 2012 at 08:28:32PM +0800, Biao wrote:
 http://hambedded.org/blog/2012/11/24/from-bitbake-hello-world-to-an-image/
 I will try to give some feedback after finishing the reading.
 But for now, there are some silly questions:
 1. There is a base_do_fetch() {}, I did not find the explanation of the 
 keyword of 'base' in the manual, do i miss something?

I don't know the internals of bitbake on that level but as far as I
understand, bitbake uses the name of the bbclass file as a prefix in the
function names for the sake of abstraction. So, as in autotools.bbclass,
the tasks defined in base.bbclass gets prefixes with base_ keyword.
Probably, it's related with mapping the function names when 'inherit'
keyword is used.

I'm not clear on this topic either. A hand from an experienced bitbake
guru would be really helpful here. The type of abstraction in bitbake
should be explained in detail as well as how EXPORT_FUNCTION works.

Any volunteers?

 2. There is 'bb.note' and 'bbnote' in the code example, is it a type-mistake 
 or something else?

No, this is not a mistake. You can use python as well as shell functions
in bitbake recipes and classes. See 'python' keyword in function
definitons where bb.note is used. That's why python function bb.note()
is used to print log information. Without python keyword, the bitbake
function is treated as shell.

Regards,
Eren

-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgp5Uyi8teb7I.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


Re: [yocto] Where does the bitbake get the variable TOPDIR

2012-12-17 Thread Eren Türkay
On Mon, Dec 17, 2012 at 05:06:26PM +0800, Biao wrote:
 Greetings, 
 
 I am a newbie tying to understand how bitbake works.
 There is one line as BBPATH = ${TOPDIR} in the mybuild/conf/bblayers.conf, 
 i would like to know where the TOPDIR is defined? 

It's defind in lib/bb/parse/parse_py/ConfHandler.py:36. When TOPDIR is
not defined in any of the bitbake configuration files, it's set to
current working directory automatically.

(...)

def init(data):
topdir = data.getVar('TOPDIR')
if not topdir:
data.setVar('TOPDIR', os.getcwd())

(...)

You may want to read the newly-written documentation about bitbake and
open embedded. I tried to explain how things fit together. I would like
to have your feedback.

http://hambedded.org/blog/2012/11/24/from-bitbake-hello-world-to-an-image/

-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgpgNt3znqf7T.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


Re: [yocto] [Draft] From Bitbake Hello World To an Image

2012-12-13 Thread Eren Türkay
On Thu, Dec 13, 2012 at 05:26:29PM -0500, Trevor Woerner wrote:
 I'm not sure if there's something about my system that is different
 than what you expect, but I had to delete my 'tmp' folder before
 invoking './bin/bitbake firstrecipe -vDD' after inheriting the
 autotool versions in order to get the new, overridden tasks to run
 (since there's no 'clean' tasks).

Thank you for your feedback. I am glad you like it.

You are right that 'tmp' directory needs to be removed after changing
the recipe (adding inherit, or removing it). I guess we are hitting
bitbake cache. Since clean tasks don't exist, 'tmp' needs to be removed
to make changes take effect.

I must have removed 'tmp' and missed adding it while writing. I will add
the explanation.

Regards,
Eren

-- 
. 73! DE TA1AET
  http://linkedin.com/in/erenturkay


pgpNVpzkw4DD2.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


[yocto] [Draft] From Bitbake Hello World To an Image

2012-12-02 Thread Eren Türkay
Hello,

I have completed a document that tries to explain how image is created
using openembedded starting from bitbake fundementals. The document
reflects my own understanding and includes the information that I wished
to have when I first involved in understanding openembedded.

This is aimed for people who want get the big picture with some details
under the hood (What's the sequence of creating an image, how is source
fetched, patched, configured, and installed? How is ipk, rpm, and deb
created? How is image produced?)

People who know bitbake can skip to Understanding OpenEmbedded part.
However, I suggest them to play with plain bitbake and read the first
chapter as bitbake is the hearth of OpenEmbedded.

There are additional information required which are marked with FIXME
keywords. Suggestions and contributions are welcomed.

I don't know how it can be integrated to current documentation but I
would like to help if that's the case. The licence of the work is CC
BY-SA. Use it, change it, adapt it, distribute it, as long as you
attribute the original work and the author :)

Document:
http://hambedded.org/blog/2012/11/24/from-bitbake-hello-world-to-an-image/

Markdown source:
https://github.com/hambedded-linux/hambedded-linux.github.com/tree/source/source/_posts

Regards,
Eren

-- 
. 73! DE TA1AET
  http://erenturkay.com/


pgplZR4ZjrBmn.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


Re: [yocto] IMAGE_INSTALL_append workflow

2012-11-27 Thread Eren Türkay
On Tue, Nov 27, 2012 at 08:40:26AM -0500, Trevor Woerner wrote:
 Thank you, Eric, for the confirmation.
 
 Just to clarify for anyone following along: the problem isn't with
 bash or trace-cmd themselves. If I perform the same steps but use
 'bash' first, then 'trace-cmd', adding 'bash' succeeds but later
 adding 'trace-cmd' causes 'trace-cmd' to fail in the same way 'bash'
 did when it was added second.

Since it's confirmed from different people, I think it would be better to report
it using bugzilla. I guess that it's not an intended behaviour/feature of
bitbake. Please inform the bug number so that anyone who is interested can
follow.

-- 
. 73! DE TA1AET
  http://erenturkay.com/


pgpIGycH7Ifpr.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto


Re: [yocto] IMAGE_INSTALL_append workflow

2012-11-26 Thread Eren Türkay
On Mon, Nov 26, 2012 at 10:47:31AM -0500, Trevor Woerner wrote:
 I'm kinda surprised nobody jumped in on this one; either my question
 is so silly it isn't worth anyone's time, or I'm the only one who
 wants to add packages to an image after a full build. I know my
 question has nothing to do with trace-cmd specifically, adding any
 package after a full build results in the same problem.

I don't think you are the first person who wants to add a package to image after
the image has been built :)

The obvious problem would have been leading spaces in IMAGE_INSTALL_append but
you explicitly stated its correct. Normally, the class that's responsible for
image creation (image.bbclass) depends on the packages defined in IMAGE_INSTALL
so that the packages are built before the image creation begins.

image.bbclass, line 9
-
RDEPENDS += ${IMAGE_INSTALL} ${LINGUAS_INSTALL} ${NORMAL_FEATURE_INSTALL}
${ROOTFS_BOOTSTRAP_INSTALL}

On irc (#oe), Eric Benard tried IMAGE_INSTALL_append =  trace-cmd after a full
build and reported that it built fine. I'm CCing him.

I believe that it will be better to include your image file and the files that
you changed so that it will be easier to debug. Mentioning your layers,
configuration, version of Yocto you use will also be helpful.

Regards,
Eren

-- 
. 73! DE TA1AET
  http://erenturkay.com/


pgpbu9aAsczjX.pgp
Description: PGP signature
___
yocto mailing list
yocto@yoctoproject.org
https://lists.yoctoproject.org/listinfo/yocto