[DEV] AOSP Dev for Zte Blade

[Guest t0mm13b]

its there for sure. no idea why its not working, I guess its the way im packing the boot.img.

anyway on another note. I raplaced boot.img with your experimental boot.img from the previous page but i STILL dont get rild sockets open.

srw-rw---- bluetooth bluetooth 2011-11-29 22:48 bluetooth

srw-rw---- bluetooth bluetooth 2011-11-29 22:46 dbus

srw-rw---- bluetooth bluetooth 2011-11-29 22:48 dbus_bluetooth

srw-rw---- root inet 2011-11-29 22:46 dnsproxyd

srw------- system system 2011-11-29 22:46 installd

srw-rw-rw- root root 2011-11-29 22:46 keystore

srw-rw---- root system 2011-11-29 22:46 netd

srw-rw-rw- root root 2011-11-29 22:46 property_service

srw-rw---- root mount 2011-11-29 22:46 vold

srw-rw-rw- root root 2011-11-29 22:46 zygote

what else did you do?? or what else do I need to do to open? was it only the boot.img that you changed?

Check if /system/bin/rild is executable via doing

ls -l /system/bin/rild

which should yield

-rwxr-xr-x

in the leftmost column. If not, use

chmod 0755 /system/bin/rild

. Tip: obtain busybox binary from busybox.org, using adb, remount and push it into /system/xbin and prefix any regular linux command with busybox as fo example

busybox chmod 0755 /system/bin/rild

:)

[Guest t181]

So now we need only some sound and radio tweaks?I can't believe it. You are the best developers. :)

[Guest AndreiKu]

offtop:

Hi, guys. Can release beta version of your mod with all the fixes that really want to try ICS? :P

[Guest skywave]

Ok i build this rom from git, after t0mm13b's last commit last night (4dca642b2c55f8e0acb440786118f2db9442d184).

On your own risk, especially as this is my first rom build ;) It boots on my phone except the touchscreen doesn't work (which probably is easy to fix)

http://www.mediafire.com/?m4r6d8bt3dwcd9m

[Guest Racerboy]

hw accln working?

why ask these questions, the best way would be to try it yourself !!

[Guest t0mm13b]

@the moboman:

Quite clearly, your insistence in cutting across and jamming on this thread in a way that is trolling.

This is a DEV THREAD only. Leave us guys alone and let us focus on this.

Your postings are about as useful as an ashtray on a motorbike going 120km/hr!

We will open a new thread for public to try out the Rom soon.

[Guest davidnintendo]

Ok i build this rom from git, after t0mm13b's last commit last night (4dca642b2c55f8e0acb440786118f2db9442d184).

On your own risk, especially as this is my first rom build ;) It boots on my phone except the touchscreen doesn't work (which probably is easy to fix)

http://www.mediafire...m4r6d8bt3dwcd9m

The touchscreen fix needed a different boot.img than the one from Gingerbread, so if you are using a boot.img file from the first posts of the thread it is normal that it doesn't work. Grab one of the last ones and it should work.

[Guest fp30]

Check if /system/bin/rild is executable via doing

ls -l /system/bin/rild

which should yield in the leftmost column. If not, use

chmod 0755 /system/bin/rild

. Tip: obtain busybox binary from busybox.org, using adb, remount and push it into /system/xbin and prefix any regular linux command with busybox as fo example

busybox chmod 0755 /system/bin/rild

:)

rild must be executable because its telling me its trying to reconnect. It doesnt make sense though that it should work on one phone and not another. All I did was to place your boot.img into alpha3 and flash. If this is all you did why would it not work correctly for me?

[Guest ricogu4]

Guys, sorry for the offtopic.

I don't understand a bleep you guys are saying or doing but somehow I just can't stop reading this thread.

Please ignore me and keep the good work!

Cheers

[Guest fp30]

has anyone else tried alpha3 with t0mm13b's boot.img?

if you have can you print the output of ls -l /dev/socket here please to see if a rild has been opened on other phones

actually anyone running any version of ics can you run the above command and print the output here

Edited November 30, 2011 by fp30

[Guest t0mm13b]

Has anyone much success with ril running? What I think, after talking to a reputable dev (hi hecatae! :) ), the original frameworks/base/core/res/res/values/config.xml highlights this

<string-array translatable="false" name="networkAttributes">

<item>"wifi,1,1,1,-1,true"</item>

<item>"mobile,0,0,0,-1,true"</item>

<item>"mobile_mms,2,0,2,60000,true"</item>

<item>"mobile_supl,3,0,2,60000,true"</item>

<item>"mobile_hipri,5,0,3,60000,true"</item>

<item>"mobile_fota,10,0,2,60000,true"</item>

<item>"mobile_ims,11,0,2,60000,true"</item>

<item>"mobile_cbs,12,0,2,60000,true"</item>

<item>"wifi_p2p,13,1,0,-1,true"</item>

</string-array>

The connectivity manager (part of radio/wifi interface) looks up those attributes... Originally, this was ripped and pasted into the device's overlay/frameworks/base/core/res/res/values/config.xml, hecatae and myself think its that what the rild was trying the impossible such as mobile_fota/mobile_ims/mobile_cbs which caused it to keel over and end up bootlooping despite receiving telephony signals... With the latest in the github:

<string-array translatable="false" name="networkAttributes">

<item>"wifi,1,1,1,-1,true"</item>

<item>"mobile,0,0,0,-1,true"</item>

<item>"mobile_mms,2,0,2,60000,true"</item>

<item>"mobile_supl,3,0,2,60000,true"</item>

<item>"mobile_dun,4,0,3,60000,true"</item>

<item>"mobile_hipri,5,0,3,60000,true"</item>

</string-array>

That yet still causes to keel over and end up bootloops... so...

My request for those with powerful machines (my netbook's disk is dying tragically from the 31 hour burn in testing - loud grating/knocking sounds... aint not good)

Can you remove the last line that states 'mobile_hipri' and rebuild it and put it here for us to try please? There should be at least, 'wifi', 'mobile_mms', 'mobile_supl', 'mobile_dun', and try mix and match with the ICS's apns-conf.xml (which had a version bump up by the way) and with those from CM7?

Thanks :)

Edit: t0mm13b_ics4blade_boot.tar.bz2 is found here. This contains the modded init.rc and init.blade.rc :)

Edit #2:

A big thank you to Skywave for testing it and yes, Skywave confirmed telephony is working :D

Edited November 30, 2011 by t0mm13b

[Guest RR2H]

So now we need only some sound and radio tweaks?I can't believe it. You are the best developers. :)

im pretty sure it won't be usable, just for experimentation perhaps. it will lag and drain battery like crazy i guess :D

Edited November 30, 2011 by RR2H

[Guest fp30]

another thing that might help to understand whats going on...

if the rild socket is not being opened then would it be worth outputing exception text here:

try {for (;;) {

            	LocalSocket s = null;

            	LocalSocketAddress l;

            	try {

                	s = new LocalSocket();

                	l = new LocalSocketAddress(SOCKET_NAME_RIL,

                        	LocalSocketAddress.Namespace.RESERVED);

                	s.connect(l);

            	} catch (IOException ex){

                	try {

                    	if (s != null) {

                        	s.close();

                    	}

                	} catch (IOException ex2) {

                	}

               	if (retryCount == 8) {

                    	Log.e (LOG_TAG,

                        	"Couldn't find '" + SOCKET_NAME_RIL

                        	+ "' socket after " + retryCount

                        	+ " times, continuing to retry silently");

                	} else if (retryCount > 0 && retryCount < 8) {

                    	Log.i (LOG_TAG,

                        	"Couldn't find '" + SOCKET_NAME_RIL

                        	+ "' socket; retrying after timeout");

                	}

i.e. if its failing to open a socket would outputing IOexception ex to the logcat shed any light as to why? perhaps adding Log.i(LOG_TAG, ex) would help?

I dont know. If someone does try this you will have to convert ex to a string. I dont know how to do this though. anybody?

[Guest t0mm13b]

another thing that might help to understand whats going on...

if the rild socket is not being opened then would it be worth outputing exception text here:

try {for (;;) {

            	LocalSocket s = null;

            	LocalSocketAddress l;

            	try {

                	s = new LocalSocket();

                	l = new LocalSocketAddress(SOCKET_NAME_RIL,

                        	LocalSocketAddress.Namespace.RESERVED);

                	s.connect(l);

            	} catch (IOException ex){

                	try {

                    	if (s != null) {

                        	s.close();

                    	}

                	} catch (IOException ex2) {

                	}

               	if (retryCount == 8) {

                    	Log.e (LOG_TAG,

                        	"Couldn't find '" + SOCKET_NAME_RIL

                        	+ "' socket after " + retryCount

                        	+ " times, continuing to retry silently");

                	} else if (retryCount > 0 && retryCount < 8) {

                    	Log.i (LOG_TAG,

                        	"Couldn't find '" + SOCKET_NAME_RIL

                        	+ "' socket; retrying after timeout");

                	}

i.e. if its failing to open a socket would outputing IOexception ex to the logcat shed any light as to why? perhaps adding Log.i(LOG_TAG, ex) would help?

I dont know. If someone does try this you will have to convert ex to a string. I dont know how to do this though. anybody?

Further to understanding, the groups used for the starting up of the rild, net_admin and net_raw is absent in init.rc, would could explain why it was not able to access the socket. The boot image that I have posted previously contains the mod, But other than that its all related to permissions/ownership of the rild binary I guess... xD

[Guest fp30]

Further to understanding, the groups used for the starting up of the rild, net_admin and net_raw is absent in init.rc, would could explain why it was not able to access the socket. The boot image that I have posted previously contains the mod, But other than that its all related to permissions/ownership of the rild binary I guess... xD

ok fair enough. Im still a bit confused though because I used your earlier boot.img with the changes in init.rc (I think it was called telephone_almost_working) which you said had given you a rild socket but I still didnt have a rild socket. Its possible I did something wrong but without further confirmation that the rild socket is indeed there with the amended init.rc can we defnitely conclude thats its just a permissions thing?

[Guest t0mm13b]

ok fair enough. Im still a bit confused though because I used your earlier boot.img with the changes in init.rc (I think it was called telephone_almost_working) which you said had given you a rild socket but I still didnt have a rild socket. Its possible I did something wrong but without further confirmation that the rild socket is indeed there with the amended init.rc can we defnitely conclude thats its just a permissions thing?

As a fyi for you and other devs, to make this easier for you to try out, and know you have had some trouble with boot.img, suppose you have the split_bootimg.pl script, (For others, its found here) and do this:

/home/joe/boot $ split_bootimg.pl boot.img 

Extract the ramdisk this way:

/home/joe/boot $ mkdir RAM; cd RAM

/home/joe/boot/RAM $ gzip -dc ../boot.img-ramdisk.gz | cpio -i

The start up scripts are extracted and you would be able to double check on the init.rc/init.blade.rc scripts, modify it and then:

/home/joe/boot $ mkbootfs RAM | gzip > myramdisk.gz

Finally to recreate the boot.img

/home/joe/boot $ mkbootimg --cmdline 'androidboot.hardware=blade console=null' --base 0x2600000 --kernel boot.img-kernel --ramdisk myramdisk.gz -o myboot.img

then fastboot flash it and away you go (If you have difficulty with the scripts, let me know I will post it here for others)

Now on to the real meat, if rild is not executable, then no socket will be created - simple, hence I asked for you to check if its indeed chmod'd 0755. What I am not sure, is why you're not getting the rild socket in /dev/socket which is a bit of a mystery :)

[Guest fp30]

then fastboot flash it and away you go (If you have difficulty with the scripts, let me know I will post it here for others)

Now on to the real meat, if rild is not executable, then no socket will be created - simple, hence I asked for you to check if its indeed chmod'd 0755. What I am not sure, is why you're not getting the rild socket in /dev/socket which is a bit of a mystery :)

thanks for the guide. I think I was missing a command in the mkbootimg step which was stopping adb from seeing the phone correctly. Now thats fixed I still cant get a rild socket im afraid. Even with your boot.img.

rild is executable

-rwxr-xr-x root shell 9932 2008-08-01 12:00 rild

but no rild sockets:

srw-rw---- bluetooth bluetooth 2011-11-30 17:19 bluetooth

srw-rw---- bluetooth bluetooth 2011-11-30 17:18 dbus

srw-rw---- bluetooth bluetooth 2011-11-30 17:19 dbus_bluetooth

srw-rw---- root inet 2011-11-30 17:18 dnsproxyd

srw------- system system 2011-11-30 17:18 installd

srw-rw-rw- root root 2011-11-30 17:18 keystore

srw-rw---- root system 2011-11-30 17:18 netd

srw-rw-rw- root root 2011-11-30 17:18 property_service

srw-rw---- root mount 2011-11-30 17:18 vold

srw-rw---- wifi wifi 2011-11-30 17:20 wpa_wlan0

srw-rw-rw- root root 2011-11-30 17:18 zygote

when I try to manually run rild with:

/system/bin/rild -l /system/lib/libreference-ril.co -- -d /dev/socket/rild

I just see this in logcat:

( 927): Opening tty device /dev/socket/rild

E/RILC ( 927): RIL_register: RIL version 6

E/RILC ( 927): Failed to get socket 'rild'

Edited November 30, 2011 by fp30

[Guest t0mm13b]

-rwxr-xr-x root shell 9932 2008-08-01 12:00 rild

The group ownership looks incorrect - think that should be root... using busybox (you do have that in your /system/xbin?) and that the filesystem is remounted as r/w

busybox chgrp 0 /system/bin/rild

[Guest fp30]

The group ownership looks incorrect - think that should be root... using busybox (you do have that in your /system/xbin?) and that the filesystem is remounted as r/w

busybox chgrp 0 /system/bin/rild

yes I have busybox but running that command didnt change the ownership.

root@android:/ # /system/xbin/busybox chgrp 0 /system/bin/rild

255|root@android:/ # ls -l /system/bin/rild

-rwxr-xr-x root root 9932 2008-08-01 12:00 rild

[Guest tilal6991]

Please moboman. Stop trolling the thread. You seem to have no idea of what is going on and are making completely inappropriate comments.

Unrelated but has anyone tried to compile with adreno fix posted earlier. If not then I will do so.

[Guest t0mm13b]

Unrelated but has anyone tried to compile with adreno fix posted earlier. If not then I will do so.

Why not? Go for it and see what happens - would love to know what's the results on that :)

[Guest fp30]

Please moboman. Stop trolling the thread. You seem to have no idea of what is going on and are making completely inappropriate comments.

Unrelated but has anyone tried to compile with adreno fix posted earlier. If not then I will do so.

I don't think anyone has yet but from what I read in a few threads on XDA its not actually a fix for adreno its more of an egl workaround. The guy on the hd2 forum who was working on the patches hadn't released the code for the patches when I checked this morning so I'm not sure where the other link came from (in the earlier post in this thread) but it's still worth a shot! Go for it :)

[Guest tilal6991]

Why not? Go for it and see what happens - would love to know what's the results on that :)

well if someone has already tried it and it didn't work for them then my whole compile time will have been wasted. So just wanted to check before starting.

[Guest FelixL]

Just to mention it, mobomans link also states this:

. Verify that applications have started

Once the shell becomes available, execute % ps to confirm that the following applications are running:

* /system/bin/logd

* /sbin/adbd

* /system/bin/usbd

* /system/bin/debuggerd

* /system/bin/rild

* /system/bin/app_process

* /system/bin/runtime

* /system/bin/dbus-daemon

* system_server

Each of these applications is embedded Linux C/C++ and you can use any standard Linux debugging tool to troubleshoot applications that aren't running. Execute % make showcommands to determine precise build commands. gdbserver (the GNU debugger) is available in the bin directory of the system partition (please see http://sourceware.org/gdb/ for more information).

You propably already knew it, but didn't want to let this slip if it's new to you. And moboman should be banned from this thread, really.

[Guest skywave]

the problem was the mobile_hipri in overlay/frameworks/base/core/res/res/values/config.xml