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

Root Hroark's Prevail Kernel Factory

Would we be able to remove and maybe patch this?

its funny because the Samsung Admire source has that too

I thought metro pcs did not use CIQ

we can probably remove it, I really dont think it is doing to much since the services are not loading and there are no calls coming from the framework

who knows
 
its funny because the Samsung Admire source has that too

I thought metro pcs did not use CIQ

we can probably remove it, I really dont think it is doing to much since the services are not loading and there are no calls coming from the framework

who knows

That looks like attempted communications with the tty device for CIQ.
look for #ifdef and #ifndef blocks around those lines to see if it;s simply switched by a compile flag, it's much easier to just tell the compiler to not use those sections than to edit them out and hope nothing breaks
 
I tested your zImage and it is fine

can you try to download the the Kouma 1.2 that he has in his thread

extract the ramdisk by following the instructions in my OP

and then building a new kernel with his unmodified 1.2 ramdisk and your zImage

and by the way who is Chubby?

I successfully extracted the Ramdisk from Kouma1.2. Compiled the boot.img. I'm going to flash it now and try it with the kouma 1.2 base
 
Any progress making a oc kernel for the prevail. Would love to see one for cm7


im not gonna make one. Im really not concerned about OC, I am more interested in ICS

but now that we have working source, someone may be able to do it
 
Hroark: I was going to mod your packing scripts for my own use/protection from stupid. I was wondering if you'd like me to forward you my edits.

Here is what I'm planning to change in psudocode:
if zImage does not exist where it's supposed to {
if .config does not exist where it should {
copy prevailconfig to .config }
Run make to build zImage }

Continue as normal

That way if I do something silly, like try to pack a kernel on a clean system, it compiles automagically
 
Hroark: I was going to mod your packing scripts for my own use/protection from stupid. I was wondering if you'd like me to forward you my edits.

Here is what I'm planning to change in psudocode:
if zImage does not exist where it's supposed to {
if .config does not exist where it should {
copy prevailconfig to .config }
Run make to build zImage }

Continue as normal

That way if I do something silly, like try to pack a kernel on a clean system, it compiles automagically

well first of all, are you messing around with this at all, it did not seem like you were interested

I have patched the source up to 2.6.35.14 and released a test kernel

do you want that updated source code and the needed fsr module

i dont think we need to have it copy prevaconfig because i make different configs to test and just do stuff like this


cp newsettingsconfig .config

or

cp myconfig .config

you get my point
 
well first of all, are you messing around with this at all, it did not seem like you were interested

I have patched the source up to 2.6.35.14 and released a test kernel

do you want that updated source code and the needed fsr module

i dont think we need to have it copy prevaconfig because i make different configs to test and just do stuff like this


cp newsettingsconfig .config

or

cp myconfig .config

you get my point

Yeah, I'm totally interested. I just don't have much time to dedicate to putting in effort at the moment. I chose to use prevailconfig as what would serve as a default in case I forgot to copy my own config. This way I ensured myself a kernel that would work. I figured a few extra lines would keep me from an occasional case of the stupids, and make something to give back to you so that you can give others something that they could use for custom ramdisks without having to worry about the specifics of kernel compiling.

Anyways, here's teh code I added so far to cwmboot.pl

Code:
my $kerndir = "$dir/../kernel";
in the section you defined strings in.
and:
Code:
chdir $kerndir or die "$ARGV[0] $!";

unless (-e "./arch/arm/boot/zImage") {
	unless (-e "./.config") {
		system("cp ./prevailconfig ./.config");
	}
	system("make");
}
after the ramdisk gets gziped and before it changes directories to pack the boot.img
 
Yeah, I'm totally interested.

well if you want to make a new script that is cool


lets call it something other than cwmboot.pl

maybe makeall.pl or something like that

I am thinking of posting on github my latest .14 code and all steps I have done to make it

I will upload the broken Samsung code first, than add a commit for the fix to make it working source, than any other commits

what do you think, or would I just be wasting my time
 
well if you want to make a new script that is cool


lets call it something other than cwmboot.pl

maybe makeall.pl or something like that

I am thinking of posting on github my latest .14 code and all steps I have done to make it

I will upload the broken Samsung code first, than add a commit for the fix to make it working source, than any other commits

what do you think, or would I just be wasting my time

I am definatly interested in the new source and learning more about how to patch it and so on. I have been doing a lot of reading on kernel modding and tinkering with the hpkf you put together. Could you send a link for the patched source and any other good reading material?

Thanks in advance Hroark.
 
well if you want to make a new script that is cool


lets call it something other than cwmboot.pl

maybe makeall.pl or something like that

I am thinking of posting on github my latest .14 code and all steps I have done to make it

I will upload the broken Samsung code first, than add a commit for the fix to make it working source, than any other commits

what do you think, or would I just be wasting my time

I'd say it's worth it. it shows how you came from point A and arrived at point B. Like showing your work in math class
 
I am definatly interested in the new source and learning more about how to patch it and so on. I have been doing a lot of reading on kernel modding and tinkering with the hpkf you put together. Could you send a link for the patched source and any other good reading material?

Thanks in advance Hroark.

ok

when i get a little time I will put it on git hub and try to make it easy to follow how i did everything
 
ok

when i get a little time I will put it on git hub and try to make it easy to follow how i did everything

Your the man Hroark. I'm actually really getting into this. :D
I have been playing around with a bunch of new configs, thanks to your spiffy terminal menu in hpkf. Not trying anything major, just seeing what does what and trying to get it all into my brain. :cool:
 
Your the man Hroark. I'm actually really getting into this. :D
I have been playing around with a bunch of new configs, thanks to your spiffy terminal menu in hpkf. Not trying anything major, just seeing what does what and trying to get it all into my brain. :cool:

if your referring to the

make menuconfig


that is in all Linux kernels
 
if your referring to the

make menuconfig


that is in all Linux kernels

Oh. See I learned something new today. I didn't know that.
I'm currently tinkering with the I/O scheduling, with some luck. I got faster read writes, but horrible over all performance in the quadrant. I got a 1128 over all, but my read/write speeds are around 7.0 mb/s.
 
ok here is the .14 code

I applied the commits in the same steps I used to get to this point

it may or may not be easy for you to follow and understand, but this is the best I can do to make it easy

https://github.com/hroark13/prevail_kernel_fixed

be sure to look at the history

So I'm on the github link, where should I start?
Do I need to sync hpkf to your github repo?
Or if I download everything as a zip, can I just extract it or add it to hpkf somehow?
 
well if you want to download it, there is a download link


if you want to look at was done


start at history and browse around

I'm reading through as we speak. I started with the folders posted the furthest back, and I'm going to work my way to the newest folders and files.
I also downloaded it as a zip.
 
Is this correct:

*
* Generic Driver Options
*
path to uevent helper (UEVENT_HELPER_PATH) []
Maintain a devtmpfs filesystem to mount at /dev (DEVTMPFS) [N/y/?] n
Select only drivers that don't need compile-time external firmware (STANDALONE) [Y/n/?] y
Prevent firmware from being built (PREVENT_FIRMWARE_BUILD) [Y/n/?] y
Userspace firmware loading support (FW_LOADER) [Y/?] y
Include in-kernel firmware blobs in kernel binary (FIRMWARE_IN_KERNEL) [Y/n/?] y
External firmware blobs to build into the kernel binary (EXTRA_FIRMWARE) []
Driver Core verbose debug messages (DEBUG_DRIVER) [N/y/?] n
Managed device resources verbose debug messages (DEBUG_DEVRES) [N/y/?] n
Enable a generic cross-process locking mechanism (GENLOCK) [N/y/?] (NEW) y
Enable a misc-device for userspace to access the genlock engine (GENLOCK_MISCDEVICE) [N/y/?] (NEW) y

I chose yes for all those options
 
Back
Top Bottom