Android-x86 Virtual Machines MAC Address

[arif]

How can i find the MAC Address of Android-x86 Virtual Machine ??

ifconfig and netcfg results:
===================
root@generic_x86:/system/bin # ifconfig
eth0
eth0: ip 172.16.8.47 mask 255.255.255.0 flags [up broadcast running
multicast]

root@generic_x86:/system/bin #
netcfg
lo UP 127.0.0.1 255.0.0.0 0x00000049
eth0 UP 172.16.8.47 255.255.255.0 0x00001043




Regards
-Arif

[arif hussain]

Is it the same MAC Address what i see in Network Adapter setting of VirtualBox ?

-Arif

--
You received this message because you are subscribed to the Google Groups "Android-x86" group.
To post to this group, send email to andro...@googlegroups.com.
To unsubscribe from this group, send email to android-x86...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/android-x86?hl=en.

[Stefan]

Hello,

this email is specially for the maintainer of the Tegav2 tree.

There is a little group of peoples running the Tegav2 images of AndroidX86 on their WeTabs and ExoPCs. They have made some changes to the Tegav2 image to get more compatible to that tablet PC. I give you my experiences with it and the links to the modifications.

The target  of this email is to sensiblize you on the needs of the users of WeTab/ExoPC and other Pegatron Lucid based Atom Tablet PCs, so that in the future the Tegav2 images could also run on these without any changes and with full hardware support.

Mostly these peoples at the wetab-community.com forum are using a modified 3.2.2 image based on the Tegav2 image on androidx86.org. They also tryed the 4.0x version but found that it is still too unstable to use it, most testers report that at least 10 minutes it hangs up. The 3.2.2 image is much more stable, I could use it more than 3 hours per session before it hangs up.

First I report you the modifications which have been done to the Android X86 3.2 Tegav2 image to support the Pegatron Lucid platform:

1. Currently the mod is based on: http://android-x86.googlecode.com/files/android-x86-3.2-RC2-tegav2.iso

2. Changes:
2a: WLAN sometimes does not get a DHCP adress after connect, this solves it:
  
chmod 777 /system/etc/dhcpd/*

2b: The touchscreen is detected as a mouse, the follwoing changes this:

cd /system/usr/idc
cp Vendor_03eb_Product_201c.idc Vendor_0eef_Product_72a1.idc

2c: The screen light can only be adjusted after after replacing a file:

cd /system/lib/hw
rm lights.tegav2.so
wget http://www.steckbecken.at/wp-content/uploads/lights.tegav2.so

Note that this change is not compatible to the original Tegav2 platform, so maybe there must be some modifications to this file to be able to support both Tegav2 and Lucid with one file. Would this be possible?

2d: To be able to use a real SD card beside a fake-SD:

cd /system/etc
mv vold.fstab vold.fstab.bak

2e: To make GPS useable on the Lucid platfrom with the GPS/3G-module:

cd /system/lib/hw
rm gps.tegav2.so
wget http://www.steckbecken.at/wp-content/uploads/gps.tegav2.so

Alos this one should not be compatible to the original Tegav2 platform, so maybe a solution could be found to support both with one file?

2f: The webcam can be made useable when replacing two other files:

cd /system/lib
rm libcamera.so
rm libcameraservice.so
wget http://www.steckbecken.at/wp-content/uploads/libcamera.so
wget http://www.steckbecken.at/wp-content/uploads/libcameraservice.so

Also these files should not be compatible to the original Tegav2 platform anymore. Maybe it is possible to change also this in a way that it can suppot both plattforms, Lucid and Tegav2?

2g: The following mod makes the following functions functional, but it still has two negative issues. The negative issues are that the screen light gets a bit darker and on some users WeTab it influences the CPU fan negative, it might not run anymore what is fatal as the tablet is no more cooled. They currently think that this mod is checking the hardware management controller too often, so that it might get overloaded and hangs up. On my own WeTab the fan control still works with this change.

The following files are required, as for the issues in the forum they are marked as experimental:

- http://www.wetab-community.com/index.php?app=core&module=attach&section=attach&attach_id=6072 ( asus-laptop.ko.gz )
- http://www.wetab-community.com/index.php?app=core&module=attach&section=attach&attach_id=6073 ( sensors.tegav2.so.gz )

Then run these commands:

cp asus-laptop.ko /system/lib/modules/3.0.1-android-x86-tegav2+/kernel/drivers/platform/x86
cp sensors.tegav2.so /system/lib/hw

boot with DEBUG=1
exit
depmod -a
exit

2f: The internal 3G modem still does not work well, it is detected, but it seems to hang up after detecting the available networks. The problem has not been solved yet. Further details can be found here: http://www.wetab-community.com/index.php?/topic/17917-android-322-komplett-rc2-image/page__st__320__p__247284#entry247284 and following messages (in german).

From Windows Devicemanager, Huawei Mobile Connect 3G Modem:
- USB/VID_12D1&PID_1404&REV_0000&MI_00

2g: For the Atheros Bluetooth module no driver yet.
From Windows Devicemanager:
- USB\VID_13D3&PID_3304&REV_0001

2h: The Android kernal is compiled without himem option, so even if the tablet PC has 2 GB of memory, it can only use 1 GB.

2i: There is no support in the Tegav2 build for the Broadcom video accelerator which is built into the WeTab 3G. Broadcom Crystal HD Video Decoder, PCI\VEN_14E4Tegav2 image on WeTab/ExoPC and other Pegatron Lucid tablets1615&SUBSYS_161514E&REV_00

2j: Also no support for the Chronotel Videoswitch of the WeTab/ExoPC to enable HDMI-out to mirror the internal display to an external one.

2k: This summary is based on the following german threads in the WeTab community forum:
I. http://www.wetab-community.com/index.php?/topic/17948-zusammenfassung-der-zwei-android-threads/ (list of changes)
II. http://www.wetab-community.com/index.php?/topic/17917-android-322-komplett-rc2-image/page__st__220__p__245469#entry245469 (322 rc2 image discussion)
III. Own experiences and researches.

3. A fully modified image runable on the Lucid/WeTab/ExoPC/... can be downloaded from message No. 1 of this thread, it is always updated with the newest changes:

http://www.wetab-community.com/index.php?/topic/17917-android-322-komplett-rc2-image/

The instruction of this messaqe in the forum shows how to boot this image out of a Windows 7 / 8 partition using NeoGrub and BCDEdit. This is very interesting for those who own an ExoPC which comes by default with Windows 7. Or for those who have deleted WeTabOS and replaced by Win 7.

I hope that one of the Android X86 group will have a close look in this message and helps to improve the developement of Android X86 to support more platforms. If you have any questions or you need help, just contact me or the user Halver of the wetab-community forum as the responsible for the most of the changes at the 3.2.RC2 image.

Thank you for providing the Tegav2 image and hope that it will support the Lucid platform soon fully,

Stefan

[Chih-Wei Huang]

Thank you for the detailed report.
I'm impressed and appreciate.
Please see my comments below.

2012/1/10 Stefan <matthaeu...@googlemail.com>:
> Hello,

> 1. Currently the mod is based on:
> http://android-x86.googlecode.com/files/android-x86-3.2-RC2-tegav2.iso
>
> 2. Changes:
> 2a: WLAN sometimes does not get a DHCP adress after connect, this solves it:
>
> chmod 777 /system/etc/dhcpd/*

This is doubtful.
I didn't see any rationale to do that.
I suggest you find the root cause for a proper fix.

> 2b: The touchscreen is detected as a mouse, the follwoing changes this:
> cd /system/usr/idc
> cp Vendor_03eb_Product_201c.idc Vendor_0eef_Product_72a1.idc

This isn't necessary in the latest code,
as I have it use a default idc now.

2c-2g are all HAL layer changes.
I'm surprised since even the original tegav2 image
doesn't have some of the features described above.
Please provide the source code or patches
so they can be integrated into the source tree.

> 2f: The internal 3G modem still does not work well, it is detected, but it
> seems to hang up after detecting the available networks. The problem has not
> been solved yet. Further details can be found here:
> http://www.wetab-community.com/index.php?/topic/17917-android-322-komplett-rc2-image/page__st__320__p__247284#entry247284
> and following messages (in german).
>
> From Windows Devicemanager, Huawei Mobile Connect 3G Modem:
> - USB/VID_12D1&PID_1404&REV_0000&MI_00
>
> 2g: For the Atheros Bluetooth module no driver yet.
> From Windows Devicemanager:
> - USB\VID_13D3&PID_3304&REV_0001

Hmm, seems you reuse 2f and 2g?
Bluetooth support is buggy in honeycomb-x86.
There are two problems here.
The ath3k is broken, and the honeycomb bluetooth
stack is too old to work with 3.x kernel.
I'm just aware the fact recently.
So it's hard to support Bluetooth in honeycomb-x86.
Fortunately this have been fixed in ics-x86.

> 2h: The Android kernal is compiled without himem option, so even if the
> tablet PC has 2 GB of memory, it can only use 1 GB.

Ok, I think tegav2 has only 1GB memory.
But it's not harmful to enable himem anyway.

I suggest you report issues and send patches
to the forum directly so that more developers can
aware and review your work.

It's important for us to co-work at the source code level.
Not just because this is an open source project,
but because we need to know what you did exactly
to do better integration.
(e.g., make the HAL compatible with both my and your tablets)

> Thank you for providing the Tegav2 image and hope that it will support the
> Lucid platform soon fully,

Let's co-work together.

Regards,
--
Chih-Wei
Android-x86 project
http://www.android-x86.org

[Stefan]

Hi Chih-Wei and all others,

thank you for the reply. Meanwhile have put together a few peoples wich are working on the WeTab/ExoPC integration into Android and specially Wetab-community member xyyzy (real name Rene, I don't know yet under which name he is active here) is already working on ICS. Here is his replay when I contacted him directly.

---- quote

Hi,

First of all a change of tegrav2 is not enough, too many differences.
A seperate device tree would be better I think.

Secondly my efforts are going to ICS, not HC anymore.
But sources are not ready yet.
I am still waiting for some REAL answers to my questions in the x86 mailing list

Maybe later.

Greetings Rene

---- /quote

I hope Rene als read this, and also those where Rene is waiting for answers... Maybe Rene repeats again his questions in context of this discussion?

I understand when he says that there might be too many differences to the Tegav, so an own device tree should be necessary, but how about compatibility then of the apps? Do they check against the device tree (tegav2 or not) or against the processor to claim compatiblity or not?

That's my question here, now I will scroll down into Chih-Wei's reply to my report and answer to his questions there.

Stefan

Am 10.01.2012 18:28, schrieb Chih-Wei Huang:

Thank you for the detailed report.
I'm impressed and appreciate.
Please see my comments below.

And I am pleased for this reaction...

2012/1/10 Stefan <matthaeu...@googlemail.com>:

chmod 777 /system/etc/dhcpd/*

This is doubtful.
I didn't see any rationale to do that.
I suggest you find the root cause for a proper fix.

Looks like this is only a simple fix and later on it could be even a security risk. I agree with you that there should be a propper fix. Rene alias xyyzy???

2b: The touchscreen is detected as a mouse, the follwoing changes this:
cd /system/usr/idc
cp Vendor_03eb_Product_201c.idc Vendor_0eef_Product_72a1.idc

This isn't necessary in the latest code,
as I have it use a default idc now.

That sounds good!

2c-2g are all HAL layer changes.

That might be the diffenrences about what Rene alias xyyzy is talking in the quote above. As I understood, Android doesn't have the Plug'n'Play hardware detection like Windows or a standard Linux kernel has for system components, that's why there are so many device trees and that's why Android boots so fast... right?

(No plans to get more flexible in the future, to support several different devices with one build?)

I'm surprised since even the original tegav2 image
doesn't have some of the features described above.
Please provide the source code or patches
so they can be integrated into the source tree.

Rene alias xyyzy is having them. See above his comment.

- USB\VID_13D3&PID_3304&REV_0001

Hmm, seems you reuse 2f and 2g?
Bluetooth support is buggy in honeycomb-x86.
There are two problems here.
The ath3k is broken, and the honeycomb bluetooth
stack is too old to work with 3.x kernel.

The WeTab/ExoPC is using a mini PCI card from Atheros which is doing Bluetooth and WLAN.

I'm just aware the fact recently.
So it's hard to support Bluetooth in honeycomb-x86.
Fortunately this have been fixed in ics-x86.

I don't care much on Honeycomb, better would be ICS. But HC is just the version which runs the best at the WeTab at the moment. In fact we already can be quite happy with the version provided over the WeTab community. I also would prefer ICS, and xyyzy / Rene is already working on it, again based on the Tegav2-tree. I wish, that this done more inside this community.

2h: The Android kernal is compiled without himem option, so even if the
tablet PC has 2 GB of memory, it can only use 1 GB.

Ok, I think tegav2 has only 1GB memory.
But it's not harmful to enable himem anyway.

I checked the Tegav2 datasheet on the web  and could see that it also can support 2 GB memory. Just open it, plugout the 1GB SoDimm and replace with a 2 GB DDR2 type... I did the same with my WeTab. (next thing I plan is to replace that tiny 32 GB SSD, I have seen an affordable 120 GB card, for a dual boot Win 7(8) / Android that would be much better.)

I suggest you report issues and send patches
to the forum directly so that more developers can
aware and review your work.

I fully support you in this message. Rene/xyyzy, please help.

It's important for us to co-work at the source code level.
Not just because this is an open source project,
but because we need to know what you did exactly
to do better integration.
(e.g., make the HAL compatible with both my and your tablets)

right.

Thank you for providing the Tegav2 image and hope that it will support the
Lucid platform soon fully,

Let's co-work together.

Regards,

Nice!

Stefan

[rvdb]

Hi,

> member xyyzy (real name Rene, I don't know yet under which name he is active here)

Well active :-)
rvdb <rene.van....@gmail.com>

> I understand when he says that there might be too many differences to

> the Tegav, so an own device tree should be necessary.
Oke, most of the differences like the HAL modules can be handled
during boot
or dynamical by setting the correct name.
This would require multiple device HAL's in one device tree (the tega
tree)

But there are some kernel modules like athk3 and btusb, that need to
change
for the wetab. because of the "strange" pci-id handling.
(In the kernel tree they are changing them every other day, because
different hardware is not doing this the same way).
pci-ids change before and after firmware load.

Also hid-multitouch needs wetab specific changes, altough that might
be solved in ICS.

The most difficult is asus-laptop.ko or pega-ext.ko.
For reading accelerometer (auto-rotate)
If using asus-laptop.ko, I used some changes that never will make it
into kernel tree.
If going for pega-ext.ko that is less of a problem although a major
rewrite
of libsensor is needed, and ofcourse we would need the sources :-)
(asus-laptop is using uevent, pega-ext is IOCTL based).
But even with pega-ext a not in kernel tree module is needed.
How do we do this in a nice way in x86 build ???
(out of kernel build is possible, but what about the depmod ?)

For suspend/resume the wetab needs to have acpi_sleep removed from
boot cmdline.
How to do this in the tega three ?
(and an other powerbtnd).

As for audio: I can not find that stuff in current ICS git.

> >> chmod 777 /system/etc/dhcpd/*
> > This is doubtful.
> > I didn't see any rationale to do that.
> > I suggest you find the root cause for a proper fix.
>
> Looks like this is only a simple fix and later on it could be even a
> security risk. I agree with you that there should be a propper fix. Rene
> alias xyyzy???

Not my change.

> > 2c-2g are all HAL layer changes.

Yep, they are :-)
Mostly stolen from other sources, with small adaptions.

> > Bluetooth support is buggy in honeycomb-x86.
> > There are two problems here.
> > The ath3k is broken, and the honeycomb bluetooth
> > stack is too old to work with 3.x kernel.

With my little changes on ath3k and btusb, command line BT is
reasonably oke.

> I don't care much on Honeycomb, better would be ICS.

Lets go ICS.
After getting camera working on HC with a lot of work, I found out
that in ICS its changed a lot.
Current version I have for ICS is not 100% oke yet.

> >> 2h: The Android kernal is compiled without himem option, so even if the
> >> tablet PC has 2 GB of memory, it can only use 1 GB.
> > Ok, I think tegav2 has only 1GB memory.
> > But it's not harmful to enable himem anyway.

Easy to change.

> >> Thank you for providing the Tegav2 image

I fully agree. This Android-x86 work earns our respect.

Greetings xyzzy-rvdb

[Chih-Wei Huang]

Hi rvdb,
Please send your changes in patch format to the forum
so we can review and integrate it.
I prefer to integrate it to the current targets.
But if not possible, let's create another target for it.

Regards,

2012/1/14 rvdb <rene.van....@gmail.com>:

--

[Phil Osoph]

On 15 Jan., 17:11, Chih-Wei Huang <cwhu...@android-x86.org> wrote:

> I prefer to integrate it to the current targets.
> But if not possible, let's create another target for it.

Oh yeah that would be really nice =)

[Corvus]

We have a really big problem with sound. Sound is not working in
updated code, but in December image was working. What changes has been
done from December to today in sound code?