PPCGeeks - NAND Testing - 05-25 Update: New LK, Recovery.img, Kernel Updates through Recovery

PPCGeeks (http://forum.ppcgeeks.com/index.php)

- Android On TP2 Development (http://forum.ppcgeeks.com/forumdisplay.php?f=319)

- - NAND Testing - 05-25 Update: New LK, Recovery.img, Kernel Updates through Recovery (http://forum.ppcgeeks.com/showthread.php?t=134598)

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

Ok got my Rhodium! It's Rhod300 therefore GSM! And yes it won't boot up! Will keep you updated if I make any progress on this!

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

Quote:

Originally Posted by neopeek (Post 2069673)

Ok got my Rhodium! It's Rhod300 therefore GSM! And yes it won't boot up! Will keep you updated if I make any progress on this!

yay finally a dev with a gsm fone that means that one day gsm might even be working with the cdma wu edition lol\\:D/\\:D/\\:D/

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

Quote:

Originally Posted by natemcnutty (Post 2069663)

So, removing /bin from sysinit.rc caused some weird data issues for me. It would start out on 3G but could never download anything. Then it would switch over to Edge and downloads would work but of course be really slow. Anyone else experience this? For now, I've just removed the link for /bin/su and put /bin back in the path. I'm thinking there are some things that reference it directly that are no longer getting called (like the ppp scripts or something).

Also, I just created the /sbin/ueventd link (to /system/init) and put init.android from rootfs as /system/init. It seems to be running like a dream!

I had the same issue. Currently I'm using this line in my sysinit.sh:
mount -o remount,rw /
rm /bin/su
mount -o remount,ro /

Crude but it works for now. I guess we just need to remove the su from the bin folder.

Quote:

Originally Posted by [ACL] (Post 2069666)

oo.. need to update the frx then. I'm uploading the last recovery for testing. Did you play with the last one? there was a bug with the backup and restore.

Another new one... :) or was it just the updated restore script.

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

Quote:

Originally Posted by Lmiller1708 (Post 2069713)

no it was just the restore script. also added the splash so yo dont freak out :-p its not the original splash i wanted but its a start.

ok, for the issue nate had. Can we check the busy box version inside /system/bin No one cares about that in haret since they use /bin , so we should double check.

otherwise there isnt much difference.

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

Got past that hang! Booted up fine! But now screen's black and can't awake! I am at work now! Will do a soft-reset now and later in the evening I will provide a solution for GSM!

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

Quote:

Originally Posted by neopeek (Post 2069750)

Got past that hang! Booted up fine! But now screen's black and can't awake! I am at work now! Will do a soft-reset now and later in the evening I will provide a solution for GSM!

we may not have the init for the panel for all gsm devices. Once the screen goes down it wont re-init.

We have the code actually just never thew it in there. Do you have a patch to get past the boot for gsm?

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

Checkked. 1st post andd didn't fffinndd "GPPPPSSS"" on list offf fffeeeattturrress...
GPS works but its having same issuess with the services? Weather widget and. Other services can't usee it.

Yyyes, hharddddware keyboaarrrd very buggy. Key downn key up not being ddeetectedd propperly?

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

Of course I have! The trick was to change the folowing in /system/sysinit.rc as it always used to hang on RIL initialization on smd 0:

Changed

PHP Code:


		
			
service ril-daemon /system/bin/rild -l /system/lib/libhtcgeneric-ril.so -- -d /dev/smd0

PHP Code:


		
			
service ril-daemon /system/bin/rild -l /system/lib/libhtcgeneric-ril.so

Quote:

we may not have the init for the panel for all gsm devices. Once the screen goes down it wont re-init.

We have the code actually just never thew it in there.

Please threw it in as I can do further testing here!

Also found a bug. Forgot once to rename androidupdate-latest.tgz to androidupdate.tgz afterwards "isCheckpointed" option wouldn't even accept whatever I did and always wanted me to rename it to androidupdate.tgz although I already renamed it and had it in right place (andboot). Had to reflash RHODIMG.NBH and start from scratch! Have to finish now. C ya later!

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

Quote:

Originally Posted by neopeek (Post 2069766)

Of course I have! The trick was to change the folowing in /system/sysinit.rc as it always used to hang on RIL initialization on smd 0:

Changed

PHP Code:


		
			
service ril-daemon /system/bin/rild -l /system/lib/libhtcgeneric-ril.so -- -d /dev/smd0

PHP Code:


		
			
service ril-daemon /system/bin/rild -l /system/lib/libhtcgeneric-ril.so

This simple trick made me boot up! But once the screen's off it won't reawake! Didn't test out if ril can work this way as I have no SIM card here at work. But now we should be able to find a proper solution for GSM + CDMA for RIL initialization. As I have seen on tinboot code you are still using force cdma option. Have seen also a lot of build.prop's which have separate rild.libarg's to do some smd 0 initialization of RIL! Will do some logcat later!

Please threw it in as I can do further testing here!

Also found a bug. Forgot once to rename androidupdate-latest.tgz to androidupdate.tgz afterwards "isCheckpointed" option wouldn't even accept whatever I did and always wanted me to rename it to androidupdate.tgz although I already renamed it and had it in right place (andboot). Had to reflash RHODIMG.NBH and start from scratch! Have to finish now. C ya later!

Ahh ok.. stinedb did the same to boot but its not a real fix. I know i removed alot of code for gsm on the kernel so eventually we will need to see whats needed. good news is wistilt is working on a ril for everyone.

Yea most of our files are hardcoded for cdma, but we can easily fix that. Also the force cdma option wont be used anymore after the next round of updates since we have variant recognition at the kernel level, so we will remove it from tinboot as well (thanks for reminding me).

once you boot into android, dump a dmesg. I'll take alook at the clocks to make sure we have everything in working order. Thanks dood.

edit: also we run a modified ril for now. So after dmesg, get a "logcat -b radio" to see how our changes are affecting gsm. Thanks again :-)

Re: NAND Boot Testing - 03-11: FRX05 on NAND (data working again!)

ACL here's the dmesg!