[ROM][03.23.13] CyanogenMod 10 on HTC DNA; Audio WIP) - HTC Droid DNA

{
"lightbox_close": "Close",
"lightbox_next": "Next",
"lightbox_previous": "Previous",
"lightbox_error": "The requested content cannot be loaded. Please try again later.",
"lightbox_start_slideshow": "Start slideshow",
"lightbox_stop_slideshow": "Stop slideshow",
"lightbox_full_screen": "Full screen",
"lightbox_thumbnails": "Thumbnails",
"lightbox_download": "Download",
"lightbox_share": "Share",
"lightbox_zoom": "Zoom",
"lightbox_new_window": "New window",
"lightbox_toggle_sidebar": "Toggle sidebar"
}
[ Unofficial CyanogenMod 10 Build on HTC DNA ]
git @ https://github.com/drewx2/android/ (Updated to download all necessary repos)
..CURRENT BUILD NOTES/HISTORY..​
03.24.13: CM10 doesn't include support for csd-client enabled phones (which we need). I have built a work around for it to utilize our csd-client lib; hopefully it brings audio one step closer.
My audio changes can be found at a separate git repo @ https://github.com/drewx2/android_hardware_qcom
03.23.13: Current Audio Status
I have some time to look at things, so here just getting started and thought I would post the first strace for those interested. Looks promising and not too difficult.
Code:
writev(4, [{"\4", 1}, {"AudioHardwareALSA\0", 18}, {"ucm instance opened: 1082445752\0", 32}], 3) = 51
writev(4, [{"\6", 1}, {"AudioFlinger\0", 13}, {"int android::load_audio_interfac"..., 103}], 3) = 117
writev(4, [{"\4", 1}, {"AudioFlinger\0", 13}, {"loadHwModule() error -22 loading"..., 49}], 3) = 63
writev(4, [{"\5", 1}, {"AudioPolicyManagerBase\0", 23}, {"could not open HW module primary"..., 33}], 3) = 57
writev(4, [{"\6", 1}, {"AudioPolicyManagerBase\0", 23}, {"Not output found for attached de"..., 47}], 3) = 71
writev(4, [{"\6", 1}, {"AudioPolicyManagerBase\0", 23}, {"Failed to open primary output\0", 30}], 3) = 54
writev(4, [{"\6", 1}, {"AudioPolicyManagerBase\0", 23}, {"getDeviceForStrategy() speaker d"..., 74}], 3) = 98
clock_gettime(CLOCK_MONOTONIC, {205, 684889413}) = 0
writev(4, [{"\6", 1}, {"AudioPolicyManagerBase\0", 23}, {"getDeviceForStrategy() speaker d"..., 74}], 3) = 98
writev(4, [{"\6", 1}, {"AudioPolicyService\0", 19}, {"couldn't init_check the audio po"..., 54}], 3) = 74
ioctl(3, 0xc0186201, 0xbe961a28) = 0
Complete strace available here: http://bit.ly/Xx3M0v
-------------------------------------------------------------------------------------------------------------------------------------
There won't be any builds posted until one of the "not working" items has been fixed. I mostly likely will focus on audio this weekend.
If you have something meaningful to share, please visit http://webchat.freenode.net/?channels=CM10DNA
03.04.13 @ 03:29AM: Wifi *now* really working
03.04.13 @ 03:29AM: Wifi now working.
Enable 3 dot menu: Settings -> Hardware keys -> Show action overflow [check]
Working:
Radio/RIL (no sound, so can't hear yet), SMS / MMS, 3G / 4G Data, Display, Orientation, Sensors, Lights, NFC, Torch, GPS (not tested), Wifi
Not working:
Camera, Audio, Bluetooth
To Do list
I plan to do bionic/kernel optimizations once all features working.
Click to expand...
Click to collapse
..INSTALL INSTRUCTIONS..​If you do not have S-OFF, you need to unzip the file and flash the boot.img inside via fastboot.
Install Instructions:
1) Download http://bit.ly/W0RAY8
2) Reboot to recovery of your choosing; wipe system/data/dalvik-cache/cache.
3) Install downloaded zip
4) Reboot.
5) Enjoy.
Google Apps @ http://goo.im/gapps/gapps-jb-20121011-signed.zip
Click to expand...
Click to collapse
Click to expand...
Click to collapse
.SPECIAL THANKS FOR DONATING.
RuinedByMTV, drmp3z
kronikings, danahern, karn101, Majik Mushroomz, Matt P., Matt B., pio_masaki, Droidika, Wheelchairmitch, MrIcky, liquidhaste, jamiethemorris, rainabba
.SPECIAL THANKS.
Flyhalf205, x3demond3x[debugging]
dsb9938 [cmdline boot option]
x3demond3x [egl fix]
Jarocks [resized bootanimation/debugging], pio_masaki [debugging],
jcase [unlock], dees_troy [twrp], beaups, jcase, Dr_Drache, dsb9938 [S-OFF]
.... and of course the CyanogenMod team and those who laid the foundation for the DNA....
..WANT TO HELP? HERE ARE SOME DEBUGGING TIPS..​1) Flash boot.img
2) On cmdline do "adb logcat"​
Click to expand...
Click to collapse
If you're ever in the mood to donate, don't forget about the others that have helped advance the HTC DNA to new levels (e.g. dsb9938, jcase, Dr_Drache, beaups, CM team, or XDA for bringing us all together!). Although, I may be working on CM and it may seem like a completely different project, in many ways we all rely on each other to help make the DNA better.
Click to expand...
Click to collapse

---

I can't help you with specifics but the guys over on the nook color forums have great guides on building and may be amenable to questions if no one here gives you anything useful
sent from my unlocked and rooted droid DNA

---

jamesbra said:
I can't help you with specifics but the guys over on the nook color forums have great guides on building and may be amenable to questions if no one here gives you anything useful
Click to expand...
Click to collapse
Thanks. Brings me back to the days when I was writing some early scripts for the Nook Color... when you had to flash the "harder" way. Anyhow, it reminded me of just mounting each partition separately (/dev/block/mmcblk0pXX). Found kernel/system logs, just what I wanted.

---

What about radio??
Sent from my HTC6435LVW using xda premium

---

typeriz said:
What about radio??
Sent from my HTC6435LVW using xda premium
Click to expand...
Click to collapse
That's a whole separate issue. We can have a Rom booting and usable without the ril working. Granted, this booting will most likely help with ril advancement.
Sent from outer space...

---

drewX2 said:
I've successfully built CM 10 on the DNA, however I'm stuck with debugging after booting in standby (assuming). For those that have built new roms on unsupported devices, how did you debug the boot process? Yes, I understand to use logcat/ddms/etc. I boot into standby with a black screen (no boot loop). Sdcard mounts once I get to black screen, however I built kernel from source as well so no hacks have been applied yet.
Also, when taking any kernel available right now, pulling it apart and remerging with no changes (just verifying process), I go into a boot loop on a working Rom (genome/ukb)
This is what I am doing to recreate boot.img:
Code:
(boot.img taking from DNA cubed kernel, I've renamed kernel and ramdisk)
unbootimg boot.img
mkbootimg --output boot.img --kernel kernel --ramdisk ramdisk.cpio.gz --cmdline 'console=ttyHSL0,115200,n8 androidboot.hardware=dlx user_debug=31' --board '' --base 80600000 --pagesize 2048
Any tips would be great so I can finish up and share. Seems so silly that I can't even remake a working boot.img after spending 20+ hrs working on CM10. It seems I'm a bit slow to get back into the swing of things since taking a hiatus from programming for several years.
Click to expand...
Click to collapse
I have gotten to the same point as you in my AOSP venture with the DNA. From my findings, without the correct kernel drivers being administered from the gate, the USB (debugging) features are not going to work properly. I am in the process of finding a kernel developer who is familiar with Sense (HTC) devices to build the proper kernel to make this possible. With the help of Dr_Drace I was able to put together a working DLX build tree and have successfully compiled FactoryROM (which is 4.2 AOSP based) but am stuck at the end of the boot process. And like you, I am unable to gather any logs due to the lack of USB access. We need a reliable kernel developer to get us past the hurdle. Unfortunately I am not one of them. As far as RIL, it's anyone's guess as to its functionality once we get things booting correctly. We have all of the drivers, it's just making them work together. :fingers-crossed:

---

MyComputerDoctor said:
I have gotten to the same point as you in my AOSP venture with the DNA. From my findings, without the correct kernel drivers being administered from the gate, the USB (debugging) features are not going to work properly. I am in the process of finding a kernel developer who is familiar with Sense (HTC) devices to build the proper kernel to make this possible. With the help of Dr_Drace I was able to put together a working DLX build tree and have successfully compiled FactoryROM (which is 4.2 AOSP based) but am stuck at the end of the boot process. And like you, I am unable to gather any logs due to the lack of USB access. We need a reliable kernel developer to get us past the hurdle. Unfortunately I am not one of them. As far as RIL, it's anyone's guess as to its functionality once we get things booting correctly. We have all of the drivers, it's just making them work together. :fingers-crossed:
Click to expand...
Click to collapse
I'm just going to try stripping as much away as I can to boot and then slowly add things back in. On a side note, I've made a working version of touch clockworkmod recovery. I am going to modify it to add some additional options before sharing (unless someone gets around to it before I do).
What I suggest doing is mounting /dev/block/mmcblk0p24 manually via working recovery. This is the log partition; system.log and kernel.log are written to it.

---

Please keep us updated on your progress. It would be greatly appreciated. And yes you may be part god.

---

Please try to keep this thread focused on development and free of clutter. The Thanks button still works just fine.

---

Do either of you have the device/vendor posted to Github by chance? I actually started putting one together but haven't had much time to actually make progress.
Which kernels have you used with source built Rom? Tried the stock kernel by chance? I always feel its a good starting point. Have either of you uploaded the zip? I can flash and take a look... They say two heads are better then one so if we're all working on the same thing I'm sure progress will be made! Lol
Sent from my HTC6435LVW using Tapatalk 2

---

Ill try to get it up or post a zip tonight (not workable, but see if you can troubleshoot ). I was off for a bit bit but now working four long shifts in a row.

---

I contacted cayniarb.. most familiar will remember him from all the tiamat htc kernels.. not sure if he will help but i sent him a link to the forum for him to take a look..
---------- Post added at 08:35 PM ---------- Previous post was at 07:53 PM ----------
since i am not versed in helping you guys ill will post cayniarbs comments in hopes something might give you a idea.. we talk breifly and when i say breifly it was only a few lines... " ok this isnt a unexpected problem the devs should try to straight up disabling problem drivers--- turning off usb means no debugging , but it might also make it boot---- as far the actual kernel changes for that it will consist of of rippin gour the sense usb drivers and replacing them with google driver.. by disable i mean just cut the init references out of the ramdisk"
hope this gives someone and idea that maybe they didnt have before.. sorry i can be no help- my skill dont lend to being this early in the project.

---

leech2082 said:
I contacted cayniarb.. most familiar will remember him from all the tiamat htc kernels.. not sure if he will help but i sent him a link to the forum for him to take a look..
---------- Post added at 08:35 PM ---------- Previous post was at 07:53 PM ----------
since i am not versed in helping you guys ill will post cayniarbs comments in hopes something might give you a idea.. we talk breifly and when i say breifly it was only a few lines... " ok this isnt a unexpected problem the devs should try to straight up disabling problem drivers--- turning off usb means no debugging , but it might also make it boot---- as far the actual kernel changes for that it will consist of of rippin gour the sense usb drivers and replacing them with google driver.. by disable i mean just cut the init references out of the ramdisk"
hope this gives someone and idea that maybe they didnt have before.. sorry i can be no help- my skill dont lend to being this early in the project.
Click to expand...
Click to collapse
Thanks for helping out anyway you can. Now that I've completed my work stretch, I can go back to working on this a bit. What he is saying is essentially what I am doing, however, the simple fact of taking a working boot.img, splitting it up, then remaking with zero changes and it it not working anymore has eluded me.

---

drewX2 said:
Thanks for helping out anyway you can. Now that I've completed my work stretch, I can go back to working on this a bit. What he is statins is essentially what I am doing, however, the simple fact of taking a working boot.img, splitting it up, then remaking with zero changes and it it not working anymore has eluded me.
Click to expand...
Click to collapse
Wink, wink!
D
.

---

dsb9938 said:
Wink, wink!
D
.
Click to expand...
Click to collapse
Haha judging by the amount of thanks I think we are all hoping that aosp is coming really soon for xmas and you guys are just playing us, haha. Thank you for your amazing work bringing the world of real android closer to reality

---

leech2082 said:
I contacted cayniarb.. most familiar will remember him from all the tiamat htc kernels.. not sure if he will help but i sent him a link to the forum for him to take a look..
---------- Post added at 08:35 PM ---------- Previous post was at 07:53 PM ----------
since i am not versed in helping you guys ill will post cayniarbs comments in hopes something might give you a idea.. we talk breifly and when i say breifly it was only a few lines... " ok this isnt a unexpected problem the devs should try to straight up disabling problem drivers--- turning off usb means no debugging , but it might also make it boot---- as far the actual kernel changes for that it will consist of of rippin gour the sense usb drivers and replacing them with google driver.. by disable i mean just cut the init references out of the ramdisk"
hope this gives someone and idea that maybe they didnt have before.. sorry i can be no help- my skill dont lend to being this early in the project.
Click to expand...
Click to collapse
Leech I actually worked with him two weeks ago to build an Tiamat AOSP kernel for the DNA. Since then I've had my birthday and 3 family members birthdays and now Christmas. The zip is still on my system waiting to be flashed We'll get it to work. The dudes built kernels for every single HTC device to date practically. I have faith.

---

Kernel building no longer issue. I just had one command left out which kept giving me a PITA. Now to find out which things can be eliminated from HTC and still boot.

---

And now to get the hardware to work properly lol. Battle has just begun. Great work guys. Will be downloading and attempting to fix some things after Xmas.
Sent from my DNA.

---

When did they add xxhdpi to aosp?
Sent from my SGH-I747 using Tapatalk 2

---

idkwhothatis123 said:
And now to get the hardware to work properly lol. Battle has just begun. Great work guys. Will be downloading and attempting to fix some things after Xmas.
Sent from my DNA.
Click to expand...
Click to collapse
Do you really have to wait until after Christmas? Come on.
:victory: