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

Root [ROM] MTDEV-CM9 build 20130120 [BETA]

I think I've found the cause of the init execve error, I added some extra logging to init:

init: cannot execve('/') for rmt_storage: Permission denied

And it looks like in init.triumph.rc there are some extra spaces around the rmt_storage stuff on line 265. Trying a build with those extra spaces removed.

EDIT: Yep, that seems to fix it. And maybe rmt_storage wasn't running before? Not sure what it does...
 
  • Like
Reactions: BSydz
Upvote 0
I think I've found the cause of the init execve error, I added some extra logging to init:

init: cannot execve('/') for rmt_storage: Permission denied

And it looks like in init.triumph.rc there are some extra spaces around the rmt_storage stuff on line 265. Trying a build with those extra spaces removed.

EDIT: Yep, that seems to fix it. And maybe rmt_storage wasn't running before? Not sure what it does...

That might be what's messing up the files in the /hidden partion....
I think that the rtm_storage was related to the stock ROMs as I am just starting to try and use the /hidden partitions.
 
  • Like
Reactions: bob-st
Upvote 0
I think I've found the cause of the init execve error, I added some extra logging to init:

init: cannot execve('/') for rmt_storage: Permission denied

And it looks like in init.triumph.rc there are some extra spaces around the rmt_storage stuff on line 265. Trying a build with those extra spaces removed.

EDIT: Yep, that seems to fix it. And maybe rmt_storage wasn't running before? Not sure what it does...
I hadn't noticed that the errors were gone in PA, now that I look, I had changed that line. Thanks for pointing that out.

Now we just have to figure out why the wifi is rebooting the phone.

The librasdioif is different in the kernel but we are using prebuilt files. So this leads me to believe that we need to revert the drivers in the kernel to stock, cause I believe that we use our stock WIFI and RIL files but everything else is from the Cherry Magnum. I am leaning toward updating the M410 kernel to work with CM9 but that means a complete revamp of the system. At the very least I think we need to integrate the things from the two kernels according to what system the drivers came from. I wish I could figure out the touch screen drivers, I have tried everything I can to update the drivers but they just don't work.

This is the kernel error that is present on almost all reboots. It is not complete, just to point to the wifi drivers being a culprit with the reboots. Last sysfs file changes so I don't think it is relevant. I also noticed that we are calling the wifi mac address twice, which is noted/fixed in the M410 kernel I believe. I also know that if you change the kernel build from 2.6.32.9-perf to anything else, the wifi will not work, mantera found a work around by adding force module, to make the module load. I ran in to this when I was building my Sharp kernel, it took me a day or so to figure out it was one line in the defconfig, I changed perf to BSydz so it would easily be seen what kernel was installed. Which led me to finding out that it has to be 2.6.32.9-perf or it won't work unless you use force module. I decided I didn't want to force anything so I just left it alone.
[HIGH][87271.129119] Unable to handle kernel NULL pointer dereference at virtual address 00000010
[87271.129272] pgd = acb88000
[87271.129302] [00000010] *pgd=2cd76031, *pte=00000000, *ppte=00000000
[87271.129486] Internal error: Oops: 17 [#1] PREEMPT
[87271.129577] last sysfs file: /sys/devices/virtual/usb_composite/modem/enable
[87271.129699] Modules linked in: libra(F) librasdioif
[87271.129791] CPU: 0 Tainted: GF (2.6.32.59-perf #1)
[87271.129913] PC is at tcp_nuke_addr+0x310/0x328
[87271.130004] LR is at fib4_rule_action+0x40/0x68
[87271.133026] pc : [<8054fac4>] lr : [<80577b58>] psr: 00010013
[87271.133056] sp : a1931df0 ip : 00000000 fp : 809ef42c
[87271.144500] r10: 0000214d r9 : 809f3ba4 r8 : a1930000
[87271.149688] r7 : 00000002 r6 : 00000000 r5 : 6700a8c0 r4 : a2d64a80
[87271.156219] r3 : 00000004 r2 : a1931e64 r1 : 00000000 r0 : ffffff9b
[87271.162719] Flags: nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment user
[87271.169830] Control: 10c57c7d Table: 2cd88059 DAC: 00000015
[87271.175567]
[/HIGH]
 
Upvote 0
Now we just have to figure out why the wifi is rebooting the phone.

I didn't know wifi was rebooting the phone. I had enabled it to download a few apps and it was working, but I didn't have it on for long.

What I have noticed, is some kind of memory leak or something. After being on for several hours, and trying to launch an app, it will usually 'hang' for a longtime before showing the applist, then will reboot the phone when I pick an app to start. Then, it gets hung on the loading screen until I wipe both cache and dalvik from recovery.
 
Upvote 0
I didn't know wifi was rebooting the phone. I had enabled it to download a few apps and it was working, but I didn't have it on for long.

What I have noticed, is some kind of memory leak or something. After being on for several hours, and trying to launch an app, it will usually 'hang' for a longtime before showing the applist, then will reboot the phone when I pick an app to start. Then, it gets hung on the loading screen until I wipe both cache and dalvik from recovery.
WIFI works fine, but the kernel oops/random reboot is always the tainted libra/librasdioif. Even when it is an expected reboot like Shadowgun: DeadZone, it gives the tainted libra/librasdioif error then the kernel debug info then kernel panic. The debug info does change depending on what is happening at the time. I am trying to learn more about how to debug the info from the last_kmsg but it is pretty complex. It could very well be a memory leak, I figure the more heads the merrier. I want to see a solid ICS ROM that is stable above all other things.

That is why I have been trying to get the touchscreen drivers from the M410 kernel working, as that made a huge difference with Sharp, U9000 and CM7, but the changes needed just aren't working with ICS. Plus the update fixed the lines in the screen that is seen with some phones. I can update all the related files with no issues, like atmel_mxt165 and synaptics_i2c_t1320 also the corresponding files in include/linux. The only one I can't get to work is bi041p_ts, I have read all the touchscreen porting threads and have compared the files between CM7 and CM9, which you probably saw was just the pressure lines. Which is how tjstyle got it working on the M410, but he did it in the bu21018mwv file. I don't think it has to do with the .idc cause we use the same one from CM7 with CM9.

Any way, I'm rambling, so any info or bouncing ideas is definitely welcome.
 
Upvote 0
WIFI works fine, but the kernel oops/random reboot is always the tainted libra/librasdioif. Even when it is an expected reboot like Shadowgun: DeadZone, it gives the tainted libra/librasdioif error then the kernel debug info then kernel panic. The debug info does change depending on what is happening at the time. I am trying to learn more about how to debug the info from the last_kmsg but it is pretty complex. It could very well be a memory leak, I figure the more heads the merrier. I want to see a solid ICS ROM that is stable above all other things.

That is why I have been trying to get the touchscreen drivers from the M410 kernel working, as that made a huge difference with Sharp, U9000 and CM7, but the changes needed just aren't working with ICS. Plus the update fixed the lines in the screen that is seen with some phones. I can update all the related files with no issues, like atmel_mxt165 and synaptics_i2c_t1320 also the corresponding files in include/linux. The only one I can't get to work is bi041p_ts, I have read all the touchscreen porting threads and have compared the files between CM7 and CM9, which you probably saw was just the pressure lines. Which is how tjstyle got it working on the M410, but he did it in the bu21018mwv file. I don't think it has to do with the .idc cause we use the same one from CM7 with CM9.

Any way, I'm rambling, so any info or bouncing ideas is definitely welcome.

Temple Run 2 and Balloon TD 5 also reboot the phone while playing. I think it is a RAM issue because by installing this app: https://play.google.com/store/apps/...sImNvbS5yb290dW5pbnN0YWxsZXIucmFtYm9vc3RlciJd it puts a button at the top of the screen that is visible at all times that you can press to kill Apps and free RAM, and by continually pressing the button, I was able to keep the phone alive longer than usual while those two games were open.
 
Upvote 0
WIFI works fine, but the kernel oops/random reboot is always the tainted libra/librasdioif. Even when it is an expected reboot like Shadowgun: DeadZone, it gives the tainted libra/librasdioif error then the kernel debug info then kernel panic. The debug info does change depending on what is happening at the time. I am trying to learn more about how to debug the info from the last_kmsg but it is pretty complex. It could very well be a memory leak, I figure the more heads the merrier. I want to see a solid ICS ROM that is stable above all other things.

Using force module load is probably a pretty bad idea... if any internal kernel data structures change, it can cause crashes and other problems.

When you talk about the m410 kernel, which one is that? I see you talk about it in your kernel thread, but I can't tell where it's from.

I think the goal should be to have a set of blobs (libcamera.so, libra.so, whatever else depends on the kernel and certainly any kernel modules) matching a vendor kernel version ("stock" kernel). Maybe that is what you are trying to do with m410.
 
Upvote 0
Seeing how much progress has been made and that I haven't donated since Christmas, I sent in $20 to cover MTDEV's costs. I cannot do this all the time, but I encourage others to spare $5 or so when they can. Together we can help maintain the costs of g60's file server, the development rigs, and also MTDEV dev phones. This will surely ease the costs. I don't even use my Triumph anymore except for helping out around here, but let's face it - with the Nexus 4 I have, anyone else who can afford that thing not only will get updates from Google, but also will be able to get a variety of custom ROMs. The Triumph may not be nearly as powerful, but I'm just sick of seeing big companies screw users with low end and mid end handsets. After all the effort put into MTDEV, and having a perfect experience (nearly, minus charging issues) on the Nexus 4, I want people without deep pockets to get stable and modern software. I know you guys aren't rolling in cash, and neither am I, but if I can spare a little to a dev team I know works their asses off and will likely not have enough funding without core supporters (unlike mainstream official CM, where there's not a lot of, but way more cash) I will continue to make this like an internet charity. You guys probably are more ethical than the CM team. Even they take beer and coffee funds, but MTDEV has never spent a dime with donations. In short, my money will go where it is needed the most, to the best devs I can find, and to the people that need an upgrade and deserve one. Viva la comunidad!
 
Upvote 0
Using force module load is probably a pretty bad idea... if any internal kernel data structures change, it can cause crashes and other problems.

When you talk about the m410 kernel, which one is that? I see you talk about it in your kernel thread, but I can't tell where it's from.

I think the goal should be to have a set of blobs (libcamera.so, libra.so, whatever else depends on the kernel and certainly any kernel modules) matching a vendor kernel version ("stock" kernel). Maybe that is what you are trying to do with m410.
Check out this thread, third post http://androidforums.com/triumph-all-things-root/655402-dev-bsydz-triumph-x6-port-encylopedia.html
 
  • Like
Reactions: adamto
Upvote 0
...I sent in $20 to cover MTDEV's costs. I cannot do this all the time, but I encourage others to spare $5 or so when they can.
Long time lurker here finally decided to register.

Finally progress has been made with the camera and that we might see the end of the tunnel :D

I have $25 or so in my paypal that hasn't cleared yet but once it does I will be more than happy to donate for the cause.

And please hold me to this! I don't want to forget and not donate :eek:
 
  • Like
Reactions: MikeRL and dsmryder
Upvote 0
Hi, first I would like to start saying thank you for all the great work the MTDEV is doing. I have been using CM7 for quite a while and and have been loving it.

This weekend I have decided to update my Triumph to CM9 and for some reason, the phone says "no service". I came back to CM7 and everything is working fine. I have also installed CM9-ALPHA (multiple developers) and it worked too.

However, I really want to use your version and I am wondering if this no service thing is a known issue. While tweaking on my phone to find a solution, I have noticed the phone cannot go out of airplane mode.

I have re-installed CM9 5 times but had the same problem every time. I have wiped the Dalvik Cache on the last two install as I thought this could help, but it did not.

Do you have any idea, on where does the problem come from?

Thank you for your help, your great work and your time.
 
Upvote 0
Hi, first I would like to start saying thank you for all the great work the MTDEV is doing. I have been using CM7 for quite a while and and have been loving it.

This weekend I have decided to update my Triumph to CM9 and for some reason, the phone says "no service". I came back to CM7 and everything is working fine. I have also installed CM9-ALPHA (multiple developers) and it worked too.

However, I really want to use your version and I am wondering if this no service thing is a known issue. While tweaking on my phone to find a solution, I have noticed the phone cannot go out of airplane mode.

I have re-installed CM9 5 times but had the same problem every time. I have wiped the Dalvik Cache on the last two install as I thought this could help, but it did not.

Do you have any idea, on where does the problem come from?

Thank you for your help, your great work and your time.

The 1/20 build in the OP has no data.
 
  • Like
Reactions: Harchers
Upvote 0
Hi, first I would like to start saying thank you for all the great work the MTDEV is doing. I have been using CM7 for quite a while and and have been loving it.

This weekend I have decided to update my Triumph to CM9 and for some reason, the phone says "no service". I came back to CM7 and everything is working fine. I have also installed CM9-ALPHA (multiple developers) and it worked too.

However, I really want to use your version and I am wondering if this no service thing is a known issue. While tweaking on my phone to find a solution, I have noticed the phone cannot go out of airplane mode.

I have re-installed CM9 5 times but had the same problem every time. I have wiped the Dalvik Cache on the last two install as I thought this could help, but it did not.

Do you have any idea, on where does the problem come from?

Thank you for your help, your great work and your time.

Download and flash the 2/23 build, the link is a page or two back
 
  • Like
Reactions: Harchers
Upvote 0
Download and flash the 2/23 build, the link is a page or two back

I'm looking for this one so I can check it out. I went back to all the posts from 2/23 until today and can't find it :thinking:

I thought I might have skipped over it but I checked twice...

Edit: I can't find it on the MTDEV Rom Browser either.

Edit #2: Sorry for the double post. Browser went wonky on me LOL
 
Upvote 0
  • Like
Reactions: GKNByNW
Upvote 0

BEST TECH IN 2023

We've been tracking upcoming products and ranking the best tech since 2007. Thanks for trusting our opinion: we get rewarded through affiliate links that earn us a commission and we invite you to learn more about us.

Smartphones