• After 15+ years, we've made a big change: Android Forums is now Early Bird Club. Learn more here.

Root ICS/CM9-JB/CM10 Development Thread

Thadd Valeika said:
which google apps package do i flash? i flashed a gb version of google apps and all it did was come up with the setup that pops up when you tuen on your lg marquee for the first time and it started to force close and wouldnt let me go through the set up so im pretty sure i flashed the wrong google apps package but idk which one to flash on LS855.can someone help?
Click to expand...

Flash the version that corresponds to the ROM you have.

If you are installing gb, flash the gb gapps.

If you are installing ics, flash the ics gapps.

If you are installing jb, install the jb gapps.

If you are installing... Well, you get the picture. :)
 
I took all the hwmod files in arch/arm from the 2.6.35 kernel and put them into the 3.0.8 kernel, then fixed things up. Unfortunately, it hangs at boot.

I think there's three main options now:

Try to figure out what's causing it to hang. That is pretty much shooting in the dark without a fbcon. I could try to get the fbcon working, but I doubt that would even matter, as this stuff all happens waaaay early in the boot process, probably before the fbcon would even start.

Continue to pull stuff out of the 2.6.35 kernel. There's no guarantee that would work.

Go back and try to figure out what's going on with the hwmod errors without pulling anything hwmod related from the 2.6.35 kernel. That's going to be incredibly difficult without some in-depth knowledge of the hardware.

So once again, I'm at a crossroads with no obvious way forward... :/
 
Ok so I have a question about RAM and killing apps...I have a memory booster and it kills apps but the apps start right back up so I know some core apps can not and should not be killed but my question is stuff like gallery, music player, you tube...etc..they start by them selves and I FORCE CLOSE but my question when I want to Play Final Fantisy III will the phone automatically kill certin apps to free up RAM for the game I want to play??

Its irritating to see all those apps running in the background when I don't need them to run unless I want to use them is this OK? will the phone manage the RAM by its self when running big games that are RAM hogs??
 
tdm said:
So once again, I'm at a crossroads with no obvious way forward... :/
Click to expand...

Would it be worth going back to using the working ICS kernel / CM10 combination and then trying to move forward from there? Sounds like the JB kernel has been leading down a major rathole.
 
sjc11221 said:
Would it be worth going back to using the working ICS kernel / CM10 combination and then trying to move forward from there? Sounds like the JB kernel has been leading down a major rathole.
Click to expand...

This is the ics kernel, from the p970. It all works except the usb port. Cameras and all.

The kernel that we used for our previous ics alpha releases is actually the LS855 gb kernel with some updates. I don't think it will drive the cameras properly.
 
tdm I just wanted to personally thank you for all of your work on this project and I can defiantly understand how frustrating how it can be to have one thing standing in you way. We all want this rom but if you need to move on I for one would completely understand that. Maybe some time away would bring about a fresh perspective or new ideas. Just my two cents.
 
  • Like
Reactions: tdm
tdm said:
The twl4030-usb code calls its interrupt handler twl4030_usb_irq() once in twl4030_usb_probe() immediately after requesting the irq. It's not a "real" interrupt, just an initialization step. This is broken in the 3.0.8 kernel and raises a hardware exception (I think I posted this before). The exception is caused by a gpio call in lge_battery_fet_onoff(), where it attempts to set the charger control, "gpio_set_value(CHAR_CONTROL, 1)" if the usb cable is plugged in. Here's the exception:

In-band Error seen by MPU at address 0
Click to expand...

I fixed this error. It was a simple fix. The LG code did not properly initialize its gpio line before calling gpio_set_value(). This is because twl4030-usb initializes and calls lge_battery_fet_onoff() before hub_charging_ic initializes (and the hub_charging_ic initalizer is where the CHAR_CONTROL gpio is initialized).

That doesn't answer the question of why this (bad) code worked in 2.6.35 but not in 3.0.8. I suspect the newer TI OMAP code is pickier, or perhaps it initializes the hardware differently or more fully.

And, unfortunately, it still does not fix the usb issue. I wonder if that is also due to bad and/or sloppy coding...?
 
JBizzleATL said:
Ok so I have a question about RAM and killing apps...I have a memory booster and it kills apps but the apps start right back up so I know some core apps can not and should not be killed but my question is stuff like gallery, music player, you tube...etc..they start by them selves and I FORCE CLOSE but my question when I want to Play Final Fantisy III will the phone automatically kill certin apps to free up RAM for the game I want to play??

Its irritating to see all those apps running in the background when I don't need them to run unless I want to use them is this OK? will the phone manage the RAM by its self when running big games that are RAM hogs??
Click to expand...

Yes, this is how Android is designed. It automatically kills apps that aren't in use when it needs memory.
 
Thadd Valeika said:
i cant find the ICS/CM9-JB/CM10 ROM Thread. can someone point me in the right direction please because i have a question about flashing the right gapps package:confused:
Click to expand...

ICS threads are here:

http://androidforums.com/marquee-all-things-root/598581-rom-alpha-quattrimus-ics.html

http://androidforums.com/marquee-al...rom-ics-miui-v4-2-9-14-alpha-4-9-16-12-a.html

If you are not a developer, you really shouldn't be installing anything else. And if you do, please don't expect us to answer support questions.
 
So in my earlier message I noted these messages:

omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for l4_core
omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for l4_per
omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for l4_wkup
omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for usbhs_ohci
omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for usbhs_ehci
omap_hwmod: gpt12_fck: missing clockdomain for gpt12_fck.
print_constraints: dummy:
Click to expand...

I think the messages for l4_core, l4_per, and l4_wkup are probably okay. The definitions look pretty similar to the ones in the 2.6.35 kernel, and it's not clear to me how these would be related to usb anyway.

I removed the messages for usbhs_ohci and usbhs_ehci by removing the corresponding definitions. I *think* these are for host side usb, but I'm not sure. These definitions did not even exist in the 2.6.35 kernel omap3xxx code, they omap44xx only. So I guess they added support for omap3xxx ohci/ehci drivers in the 3.0.8 kernel.

Don't know about gpt12_fck, but I'm assuming that's benign.

The print_constraints message comes from the regulator code. There's a facility to use a "dummy", or virtual, regulator in certain circumstances. This isn't used for the sniper board but for some reason it's still compiled into the 3.0.8 kernel. I just removed that to fix the message.

So that brings me up to the current issue, which is, when unplugging the usb cable, I see the following messages about i2c errors:

twl4030_phy_power + : PHY_PWR_CTRL=0 ====^^==== (twl4030-usb.c)
twl4030_phy_power + : PHY_CLK_CTRL=6 ====^^==== (twl4030-usb.c)
twl4030_phy_power + : PHY_CLK_CTRL_STS=1 ====^^==== (twl4030-usb.c)
twl: i2c_read failed to transfer all messages
twl4030_usb twl4030_usb: Write1[0,0xfd] wrote 00 but read 01
[charging_msg] charging_ic_work_func
[Battery] Charging IC - EOC
omap_i2c omap_i2c.1: controller timed out
twl: i2c_write failed to transfer all messages
twl4030_usb twl4030_usb: Write2[0,0xfd] wrote 00 but read 01

That is just the start, there are more errors following.

So far I've tried to take the 2.6.35 i2c-omap.c code and put it into 3.0.8 but that's not booting. I'll be digging into this further...
 
Hey Blooddawn or TDM, I haven't been around for a while, did you guys ever get the camera's working on any of the Android 4.0 ROMS?
 
bfldworker said:
I have been trying to keep up with everything. But between work, life and the fact that I also have been fixing up computers on the side I haven't been able to keep up to speed on this thread.

Has any other Test, Alpha, Beta, RC or flashable fixes been released since this post?

http://androidforums.com/5005193-post980.html
Click to expand...

I thinks there's a test3 from bloodawm, but i don't know what is working /not working on that test or even the differences with tdm's ROM .
 
Tito23p said:
I thinks there's a test3 from bloodawm, but i don't know what is working /not working on that test or even the differences with tdm's ROM .
Click to expand...

I believe the test3 from bloodawan you are referring to is actually for his hyper kernel (I'm running it now) and is designed (currently) for his ctmod and other gb roms.
 
I think I'm going to have to throw in the towel on the 3.0.8 kernel.

I'm just about as persistent and stubborn as they get, and I've got pretty good kernel skills, but this just isn't happening for some reason.

So, starting next week, I'm going to be building jb with the 2.6.35 kernel. It should be on par with ics, at least.
 
tdm said:
I think I'm going to have to throw in the towel on the 3.0.8 kernel.

I'm just about as persistent and stubborn as they get, and I've got pretty good kernel skills, but this just isn't happening for some reason.

So, starting next week, I'm going to be building jb with the 2.6.35 kernel. It should be on par with ics, at least.
Click to expand...

Does the 3.0.8 boot with the bloodawn test ROM?
 
You must log in or register to reply here.
Back
Top Bottom