Aggiornamenti sul Kernel 3.x

[chievo93]

Aggiornamento da Adi:

[DEV]

So I have started playing around with stuff here..

Thought of posting here just to keep you guys updated (But don't bug me for ETA's! )

I just thought of porting the existing 2.6.35 code into 3.0.. and I have made a bit of progress upto now(I guess!)..

I set up printk flags everywhere in the board-code.. and the Kernel dies 'just' before the board_init function..

Here is the log: Uncompressing Linux... done, booting the kernel. <6>Initializing cgroup subsys - Pastebin.com

My git: https://github.com/AdiPat/android_ke...atona-3.0-port

My latest commit

Look at Line 359, I commented omap2_init_common_hw() .. and as per 2.6.35 , it initializes the clockdomains..
When I compared the LG Kernel and omapzoom kernel, I found that LG has removed some clockdomain code.

So I will put it all back in there and try again..

As for these warnings..

Code:
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for l4_core
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for l4_per
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for l4_wkup
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for usbhs_ohci
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for usbhs_ehci
<4>omap_hwmod: gpt12_fck: missing clockdomain for gpt12_fck.
They are surely not killing the Kernel since they appear in the LG OB Kernel too(Working, bootable one) ..

And Big thanks to hillbeast and dhiru for their experience and help..!

[amo93]

Aggiornamento da Adi:

[DEV]

So I have started playing around with stuff here..

Thought of posting here just to keep you guys updated (But don't bug me for ETA's! )

I just thought of porting the existing 2.6.35 code into 3.0.. and I have made a bit of progress upto now(I guess!)..

I set up printk flags everywhere in the board-code.. and the Kernel dies 'just' before the board_init function..

Here is the log: Uncompressing Linux... done, booting the kernel. <6>Initializing cgroup subsys - Pastebin.com

My git: https://github.com/AdiPat/android_ke...atona-3.0-port

My latest commit

Look at Line 359, I commented omap2_init_common_hw() .. and as per 2.6.35 , it initializes the clockdomains..
When I compared the LG Kernel and omapzoom kernel, I found that LG has removed some clockdomain code.

So I will put it all back in there and try again..

As for these warnings..

Code:
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for l4_core
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for l4_per
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for l4_wkup
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for usbhs_ohci
<4>omap_hwmod: _populate_mpu_rt_base found no _mpu_rt_va for usbhs_ehci
<4>omap_hwmod: gpt12_fck: missing clockdomain for gpt12_fck.
They are surely not killing the Kernel since they appear in the LG OB Kernel too(Working, bootable one) ..

And Big thanks to hillbeast and dhiru for their experience and help..!

Per adesso funziona...piú o meno

Inviato dal mio GT-I9003 con Tapatalk 2

[SuPeRvB92]

qualcuno che possa tradurlo?XD intendo non solo una traduzione inglese italiano ma più una traduzione tecnica XD

[JkDroid]

Molto semplice, Adi ha intuito una possibile soluzione al crash copiando da LG. Sono i primi passi ci vorrà del tempo per qualcosa di concreto!!

[chievo93]

Un aggiornamento anche da Dhiru:

An update from my side too.

I have been trying to get MTD working on 2.6.35 before we use it on 3.0 since it would be better to fix issues with 2.6.35 and then port it to 3.0

So far I was able to initialize the Onenand and create partitions. We will have to include yaffs and make changes to the ramdisk.

Code:
<6>OneNAND driver initializing
<6>omap2-onenand omap2-onenand: initializing on CS0, phys base 0x08000000, virtual base e0880000
<7>OneNAND Manufacturer: Samsung (0xec)
<6>Muxed OneNAND 512MB 1.8V 16-bit (0x50)
<6>OneNAND version = 0x0241
<7>Chip support all block unlock
<7>Chip has 4KiB pagesize
<6>Scanning device for bad blocks
<4>Bad eraseblock 2047 at 0x1ffc0000
<5>Creating 7 MTD partitions on "omap2-onenand":
<5>0x000001480000-0x000001c80000 : "boot"
<5>0x000001c80000-0x000002480000 : "recovery"
<5>0x000002480000-0x00001ca80000 : "datadata"
<5>0x00001ca80000-0x00001dc00000 : "cache"
<5>0x00001dc00000-0x00001e880000 : "efs"
<5>0x00001e880000-0x00001f880000 : "radio"
<5>0x00001f880000-0x000020000000 : "reservoir"

[JkDroid]

Lui si sta occupando di un'altra cosa! Dai dai che qualcosa si muove!!!

[Dandelion]

Stanno lavorando da poco, e già mi piace! rispetto a hillbeast rendono più partecipi gli utenti dei loro progressi, così non c'è bisogno di chiedere quanto manca ancora
E magari pubblicando i vari log qualche altro utente potrebbe avere un'illuminazione e contribuire ^^

[JkDroid]

Mi sa che hillbeast è un pezzo che è fermo, non accorgersi di come LG ha lavorato sul Kernel mi sembra grossa!!

[chievo93]

Mi sa che hillbeast è un pezzo che è fermo, non accorgersi di come LG ha lavorato sul Kernel mi sembra grossa!!

Basta vedere le correzioni che Adi ha fatto sul suo Github rispetto al kernel di hillbeast!

Fix some silly typos
Delete shit files

[Dandelion]

Adi si che ha tempo da perdere! credo abbia fatto più lui in due giorni che HB in 6 mesi....
E a quanto pare LG non si smentisce mai... la qualità del codice è la stessa dei loro materiali...