ICS-x86 is ready for developers
Chih-Wei Huang
The ics-x86 branch based on Android 4.0.1 is ready.
You may download the source as usual:
$ repo init -u git://git.android-x86.org/manifest.git -b ics-x86
$ repo sync
or from the SourceForge mirror
$ repo init -u git://android-x86.git.sf.net/gitroot/android-x86/x86/platform/manifest.git
-b ics-x86
$ repo sync
What work:
* Wifi
* Multitouch
* OpenGL ES Hardware acceleration for AMD Radeon chipset
What NOT work (yet):
* Sound
* Camera
* Ethernet
* Hardware acceleration for Intel platform
I've uploaded an iso for AMD brazos tablet in the download page.
Enjoy it!
--
Chih-Wei
Android-x86 project
http://www.android-x86.org
Chih-Wei Huang
Alexandru Stoian
android x86. I remembered they gave some hardware for development but
is their role bigger?
Ollie
android-x86 before so it's all new to me other than downloading the
pre-built images.
Касаткин Илья
Marc Mauri Alloza
Were did you get te kernel? I would like to build an image for exopc.
please make iso for viewpad. And include the driver for the broadcom.
--
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.
Rich
are some driver issues that I summarized in issue 505. Please take a
look. Thank you
Chih-Wei Huang
No.
Chih-Wei Huang
Yes. AMD provides great support to us, including devices donation and engineer's support.
On the other hand, Intel still refuses to provide any help to this project. They closed all contact windows I've ever tried.
Therefore, if you plan to ship android-x86 product, I recommend the AMD platform, which is the best platform to run ics-x86 so far.
fgdn17
ics@ubuntu-10-4:~/ics-x86$ repo init -u git://git.android-x86.org/manifest.git
-b ics-x86
Get https://android.googlesource.com/tools/repo
remote: Counting objects: 1462, done
remote: Finding sources: 100% (80/80)
remote: Total 1462 (delta 952), reused 1462 (delta 952)
Receiving objects: 100% (1462/1462), 438.11 KiB, done.
Resolving deltas: 100% (952/952), done.
From https://android.googlesource.com/tools/repo
* [new branch] maint -> origin/maint
* [new branch] master -> origin/master
* [new branch] stable -> origin/stable
* [new tag] v1.0 -> v1.0
* [new tag] v1.0.1 -> v1.0.1
* [new tag] v1.0.2 -> v1.0.2
* [new tag] v1.0.3 -> v1.0.3
* [new tag] v1.0.4 -> v1.0.4
* [new tag] v1.0.5 -> v1.0.5
* [new tag] v1.0.6 -> v1.0.6
* [new tag] v1.0.7 -> v1.0.7
* [new tag] v1.0.8 -> v1.0.8
.
.
.
* [new tag] v1.7.8 -> v1.7.8
* [new tag] v1.7.8.1 -> v1.7.8.1
Get git://git.android-x86.org/manifest.git
remote: Counting objects: 2689, done.
remote: warning: suboptimal pack - out of memory
remote: fatal: inflateInit: out of memory (no message)
remote: aborting due to possible repository corruption on the remote
side.
fatal: protocol error: bad pack header
remote: Counting objects: 2689, done.
remote: Compressing objects: 100% (1266/1266), done.
remote: Total 2689 (delta 1049), reused 2396 (delta 933)
Receiving objects: 100% (2689/2689), 458.05 KiB | 207 KiB/s, done.
Resolving deltas: 100% (1049/1049), done.
From git://git.android-x86.org/manifest
* [new branch] android-x86-1.6 -> origin/android-x86-1.6
* [new branch] android-x86-2.2 -> origin/android-x86-2.2
* [new branch] android-x86-2.2-r2 -> origin/android-x86-2.2-r2
* [new branch] android-x86-v0.9 -> origin/android-x86-v0.9
* [new branch] donut-x86 -> origin/donut-x86
* [new branch] eclair-x86 -> origin/eclair-x86
* [new branch] froyo-x86 -> origin/froyo-x86
* [new branch] gingerbread-x86 -> origin/gingerbread-x86
* [new branch] honeycomb-x86 -> origin/honeycomb-x86
* [new branch] ics-x86 -> origin/ics-x86
Your Name [Android ICS]:
thanks
Stephen Ferrell
took several attempts to successfully connect to my router. No touch
screen or audio. But everything else worked quite well. Google
Market also worked just fine.
smarte
into ICS 4.0.
I get on the boot up screen "loading android".. but noting is
happend.
Thnaks for the hard work... developers..
> > Android-x86 projecthttp://www.android-x86.org- Tekst uit oorspronkelijk bericht niet weergeven -
>
> - Tekst uit oorspronkelijk bericht weergeven -
dale daigh
You guys do great work, thanks for everything!
I have the north American model 8Z467 if that makes a difference.
vgernyc
except for the Wi-Fi and audio.
I purchased the AzureWave AW-NB037 which is the WLAN nic from the MSI
Wind 110W to get Wi-Fi to work.
I could see this working on a WYSE Z90 too easily. An Android thin
client/cloud device?
Maybe convert a Zotac Zbox A03 into an Android TV device?
fuzzy7k
The most dummy proof thing I've tried is creating an amd_brazos build
with one small tweak, device/amd/brazos/
BoardConfig.mk:#TARGET_ARCH_VARIANT := x86-atom. I'm using a Turion
X2, which doesn't support atom optimizations. The next most dummy
proof thing I've tried is doing the same with the asus_laptop build in
addition to changing the kernel target arch from core2 to k8 and
adding support for amd cpus.
The results I get under normal boot are:
-"Detecting Android-x86... found at /dev/sda8" with flashing caps
lock.
The results I get under Debug mode are:
-no early debug shell, I see the message to type exit and the message
that it's running the shell, but it just blows through without
skipping a beat
-no late debug shell, again I get the type exit message, and the
running shell message with a blinking cursor on the line below, but no
prompt. I can type, but there is no prompt so nothing happens when I
hit enter.
-I can switch to ps2,3, and run dmesg which looks ok, but logcat
gives me "Illegal instruction".
On Dec 1, 7:12 am, smarte <raf...@gmail.com> wrote:
> On the Acer AO521 AMD Athlon Neo netbook is it not possible to boot
> intoICS4.0.
> I get on the boot up screen "loading android".. but noting is
> happend.
>
> Thnaks for the hard work... developers..
>
> On 1 dec, 03:45, Stephen Ferrell <stephen.ferr...@gmail.com> wrote:
>
> > Just finished trying this on an Iconia W500. Wireless worked but it
> > took several attempts to successfully connect to my router. No touch
> > screen or audio. But everything else worked quite well. Google
> > Market also worked just fine.
>
> > On Nov 30, 11:53 am, Chih-Wei Huang <cwhu...@android-x86.org> wrote:
>
> > > Hi all,
>
> > > Theics-x86 branch based on Android 4.0.1 is ready.
> > > You may download the source as usual:
>
> > > $ repo init -u git://git.android-x86.org/manifest.git -bics-x86
> > > $ repo sync
>
> > > or from the SourceForge mirror
>
> > > $ repo init -u git://android-x86.git.sf.net/gitroot/android-x86/x86/platform/manifest.git
> > > -bics-x86
> > > $ repo sync
>
> > > What work:
> > > * Wifi
> > > * Multitouch
> > > * OpenGL ES Hardware acceleration for AMD Radeon chipset
>
> > > What NOT work (yet):
> > > * Sound
> > > * Camera
> > > * Ethernet
> > > * Hardware acceleration for Intel platform
>
> > > I've uploaded an iso for AMD brazos tablet in the download page.
> > > Enjoy it!
>
> > > --
> > > Chih-Wei
> > > Android-x86 projecthttp://www.android-x86.org-Tekst uit oorspronkelijk bericht niet weergeven -
Chih-Wei Huang
I've got a reply from an Intel engineer (privately).
He said "Intel platform does not yet supportICS, but Intel developers
are working on it.
i965 support will arrive in several weeks."
However, I still have some unclear questions
that are not answered:
* Whether Intell will support i915 chip?
That is, the video chip used in most atom based
tablet or netbook.
* Whether Intel will release the solution to public
so all the community can benefit from it.
Bzbit
Can you please elaborate about what is the issues - what is missing in
order Intel platform to support ICS?
Thx
Chih-Wei Huang
We still have troubles to bring OpenGL ES
hardware acceleration (by GPU) to
Intel video chipsets, including i915 and i965.
More technically, when start surfaceflinger,
the screen becomes whole black, and
the errors last in logcat:
Surfaceflinger: GL error 0x500
Are you interesting in resolving it?
Welcome on board.
2011/12/5 Bzbit <benz...@gmail.com>:
> Hi Chih-Wei
> Can you please elaborate about what is the issues - what is missing in
> order Intel platform to support ICS?
--
Bzbit
In my case - I can see the UI, but I get "Unfortunately, System UI has
stopped." massage box will popup every few seconds. I cant use the
console, but still I will try to see if its the same "GL error 0x500"
issue.
On Dec 5, 11:04 am, Chih-Wei Huang <cwhu...@android-x86.org> wrote:
> I've said that in another thread.
> We still have troubles to bring OpenGL ES
> hardware acceleration (by GPU) to
> Intel video chipsets, including i915 and i965.
>
> More technically, when start surfaceflinger,
> the screen becomes whole black, and
> the errors last in logcat:
>
> Surfaceflinger: GL error 0x500
>
> Are you interesting in resolving it?
> Welcome on board.
>
> 2011/12/5 Bzbit <benziz...@gmail.com>:
Chih-Wei Huang
you won't have the issue "System UI has stopped."
2011/12/5 Bzbit <benz...@gmail.com>:
> Thx for your replay.
>
> In my case - I can see the UI, but I get "Unfortunately, System UI has
> stopped." massage box will popup every few seconds. I cant use the
> console, but still I will try to see if its the same "GL error 0x500"
> issue.
--
Bzbit
when typing logcat in the console I get - "Unable to open log device '/
dev/log/main':No such file or directory", why is that?
Thx
On Dec 5, 1:36 pm, Chih-Wei Huang <cwhu...@android-x86.org> wrote:
> If you have succeeded to enable hardware acceleration,
> you won't have the issue "System UI has stopped."
>
> 2011/12/5 Bzbit <benziz...@gmail.com>:
fuzzy7k
I believe I have tracked down the culprit to my problem:
bionic/libc/Android.mk:
# libc_common_cflags += -mtune=i686 -DUSE_SSSE3
This can and is already set in build/core or within an individual
target.
I'm doing a build now to verify that this is the only change needed.
> > > > Android-x86 projecthttp://www.android-x86.org-Tekstuit oorspronkelijk bericht niet weergeven -
nuri
Ben Rivera
Any chance of a one for the 3.2 or 4.0 builds?
Thanks for the great work,
Ben
fuzzy7k
> dev/log/main':No such file or directory", why is that?
> Thx
Bzbit, after building without -DUSE_SSSE3, I now get the same error.
What processor & video chip are you trying to run on? Mine is k8 w/
RS780.
> You have posted tx2500 builds in the past that worked great for me.
> Any chance of a one for the 3.2 or 4.0 builds?
Sure. CW is the one who posts the iso's. Due to google not releasing
the honeycomb source until ics, I didn't have the code until after he
posted the other RC1's. But I have made the necessary modifications
and have been testing honeycomb and it is worthy of an RC release.
I'll let him know.
Chih-Wei Huang
> Thanks to Henri, http://groups.google.com/group/android-x86/browse_thread/thread/dc6087e986f2113,
> I believe I have tracked down the culprit to my problem:
>
> bionic/libc/Android.mk:
> # libc_common_cflags += -mtune=i686 -DUSE_SSSE3
This is strange. I just checked.
These flags has been moved to build/ since gingerbread.
(I patched it and it has been merged to AOSP)
Don't understand why it re-appear in ICS?
I guess it's probably due to some incorrect
conflicts resolving process.
I'll try to contact the Google engineer
to figure it out.
Bzbit
4500MHA
fuzzy7k
uses r600g. The first thing that we have in common is that we don't
have low power cpu's, atom/brazos. Yours supports SSSE3, and I have
fixed my issue with that. I wonder if there is some other special atom
instruction that we are missing.
The second thing that we have in common is that we are doing our own
builds. Maybe we are missing something required. How about you
download the asus_laptop and/or eeepc isos to see if they boot. That
should give us a handle on whether this is a hardware incompatibility
or a build issue. They won't run on hardware because they have SSSE3
optimizations.
Probably not the cause, but at least a symptom of our troubles is it
seems that mdev is not doing it's job, or doing it poorly. There are
no sub directories under /dev (e.g. /dev/log/, /dev/input/, /dev/
block/...) but all of the events and devices are there (e.g. /dev/
log_main, /dev/eventX, /dev/sdaX)
Ben Rivera
Ben Atkinson
http://code.google.com/p/android-x86/downloads/list on an Asus EeePC
1215N, the Wifi doesn't work with any of the builds. The wifi is on,
the BIOS says so and the blue light on the laptop is on (which toggles
when I press Fn+F2), but ICS gets stuck on "Turning Wifi on..." and
doesn't do anything.
Ben Atkinson
using Ubuntu, and the Wifi drivers for Ubuntu?
On Dec 9, 4:07 pm, Ben Atkinson <ben1...@googlemail.com> wrote:
> I've tried all four of the builds available fromhttp://code.google.com/p/android-x86/downloads/liston an Asus EeePC
fuzzy7k
tx2500 does not have.
Ben Rivera
According to AMD's website the Turion X2 and X2 Ultra support SSE,
SSE2, SSE3 and MMX. I was able to get the 3.2 ASUS build to run, but
there was not touch screen or wifi. I would be more than happy to
provide testing for a tx2500 3.2 builld. Also, If all that is needed
is SSE3 support then I think a 4.0 build should work too, atleast
according to AMD.
http://www.amd.com/us/infrastructure/processors/turion-x2/Pages/turion-x2-mobile-features.aspx
> > > I'll let him know.- Hide quoted text -
>
> - Show quoted text -
Ben Atkinson
for modprobe is b43, and when I try that it says "modprobe: module b43
not found in modules.dep"
On Dec 9, 6:47 pm, tabletsx86 <rbg.tablets...@gmail.com> wrote:
> search the forums as there is a way to include the driver is it's
> there...
>
> something like
>
> modprobe driver
> to see if your driver can load correctly.
>
> On Dec 9, 12:29 pm, Ben Atkinson <ben1...@googlemail.com> wrote:
>
>
>
>
>
>
>
> > So I've been searching around, it looks like I have to build it myself
> > using Ubuntu, and the Wifi drivers for Ubuntu?
>
> > On Dec 9, 4:07 pm, Ben Atkinson <ben1...@googlemail.com> wrote:
>
> > > I've tried all four of the builds available fromhttp://code.google.com/p/android-x86/downloads/listonanAsus EeePC
fuzzy7k
On Dec 10, 9:34 am, Ben Rivera <benrivera...@gmail.com> wrote:
> I checked my tablet and it has an AMD Turion X2 Ultra ZM-82.
> According to AMD's website the Turion X2 and X2 Ultra support SSE,
> SSE2, SSE3 and MMX. I was able to get the 3.2 ASUS build to run, but
> there was not touch screen or wifi. I would be more than happy to
> provide testing for a tx2500 3.2 builld. Also, If all that is needed
> is SSE3 support then I think a 4.0 build should work too, atleast
> according to AMD.
>
> http://www.amd.com/us/infrastructure/processors/turion-x2/Pages/turio...
Ben Rivera
I didn't realize there were 3 S's. If all I can get 3.2 on this tablet, that would still be great.
--
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.
Bzbit
> > > - Show quoted text -- Hide quoted text -
fuzzy7k
precompiled iso's that I mentioned and taking into consideration that
you have SSSE3 support, the only thing I can think is that there are
other, non-sse optimizations for atom cpus that are in use somewhere.
In order to find the root cause we'd probably need to look at the git
history and bisect the commits between honeycomb & ics.
fuzzy7k
b682731af78b77cd50ae0ea5d9ba3abb19c98add is a major concern, but a
simple revert did not get me around the issue. There are also a couple
other march and mtune=atom flags that I changed to suit my cpu, but no
luck. Check this to see what I'm talking about, grep -ris =atom ndk/
build/tools/toolchain-patches/gcc.
This is what I did to try to get a working build: check out a fresh
tree, perform the revert and sed the =atom lines in ndk, make sure
that bionic wasn't set to use ssse3, and make things the same as I
have for honeycomb. It all comes out the same with a locked ui and no /
dev sub directories.
Bzbit, you may have more luck changing the =atom optimizations since
you have support for ssse3. You might try just checking out a fresh
tree and changing the march/mtune= flags from atom to nocona (after
verifying this is what you have).
I'm not certain that the above commit is causing the problem because I
did a search through an old build log and did not find any =atom flags
set. However, I have little clue as to what the patch is doing and
since I've not seen any comments about ics working on something other
than atom arch, clearly there must be something atom embedded. It
sounds as though google did not intend for this, since their
documentation clearly states that ssse3 is not to be a requirement,
but an option, and mtune=atom goes a level beyond requiring ssse3.
