Jump to content

[DEV][ROM][8.1.] CyanogenMod 10 (Android 4.1.2)


Guest KonstaT

Recommended Posts

Guest emc02

Hi Konstat,

I've installed this great ROM yesterday (of course with full wipe) and I know its still (pre)alpha, but there is an issue that is also in your CM9 version:

when starting an App (like facebook, maps,...) that uses GPS (or you could say: when starting GPS), very often my mobile data connection gets lost and needs a few minutes and/or toggling 2G/3G and data on/off to connect again.

This issue is only in your ROMS and was not in tilals coldfusion CM9. Maybe you know what you changed and what could cause that.

All other things are great! Feels much smoother than CM9!

Link to comment
Share on other sites

Guest ..Abel..

Did matt4321 even ask your permission to post it on xda? He could of at least posted a link to the original thread here at modaco...

he wrote this "All credits to KonstaT, Lalit, The CM team and all other blade devs!

Original Modaco Thread: http://www.modaco.co....0-android-411/"

Edited by ..Abel..
Link to comment
Share on other sites

Guest Felsch

To all who think the link is there since the beginning, it wasn't there. It is just as shimizan said, he edited quickly. Also, if you go to page 3 of that topic on XDA then atleast you see they know that there has been some mistake, but the guy who wrote the article didn't answer their call yet... bit of a shame of XDA tho...

Link to comment
Share on other sites

Guest Loren82

KonstaT, there is a change for the patch "0001-don-t-build-Fraunhofer-aac-for-blade.txt"? I saw in your github, you updated the patches, the 25, and when compiling, it gives an error and think it might be by this patch.

Link to comment
Share on other sites

Guest KonstaT

he wrote this "All credits to KonstaT, Lalit, The CM team and all other blade devs!

Original Modaco Thread: http://www.modaco.co....0-android-411/"

Why don't you just bother reading the whole thread - it's only three pages long. There's now some other guy who thinks this ROM is Lalit's work... lol, I can't even post to XDA to clear this up just because I'm under 10. :P

The link wasn't there yesterday and you can still get the impression from the thread that matt4321 is posting his own work. I can't stop people from opening threads for this ROM on other android forums but I would prefer if a link to this thread was always given with acknowledgement whose work your actually posting. There is also no need to give direct links to the files, please post a link to this thread instead. That way everyone can get the latest updates and nowadays I usually remove all old releases/add-ons from my file hosting service anyway.

Major fail on behalf of XDA news team nevertheless. It's not on the sidebar anymore and I'm pretty sure no one saw it anyway. ;)

KonstaT, there is a change for the patch "0001-don-t-build-Fraunhofer-aac-for-blade.txt"? I saw in your github, you updated the patches, the 25, and when compiling, it gives an error and think it might be by this patch.

The patch is perfectly fine. You're doing/not doing something else. I know that you are not using my sources. I really can't help you because I have no idea what code you are using.

Link to comment
Share on other sites

Guest andr0idbeliev3r

Why don't you just bother reading the whole thread - it's only three pages long. There's now some other guy who thinks this ROM is Lalit's work... lol, I can't even post to XDA to clear this up just because I'm under 10. :P

You are under 10?

Link to comment
Share on other sites

Guest KonstaT

You are under 10?

lol, it was just an expression. :D You have to have at least ten posts before you are allowed to post to development forums.

Link to comment
Share on other sites

Guest Loren82

Why don't you just bother reading the whole thread - it's only three pages long. There's now some other guy who thinks this ROM is Lalit's work... lol, I can't even post to XDA to clear this up just because I'm under 10. :P

The link wasn't there yesterday and you can still get the impression from the thread that matt4321 is posting his own work. I can't stop people from opening threads for this ROM on other android forums but I would prefer if a link to this thread was always given with acknowledgement whose work your actually posting. There is also no need to give direct links to the files, please post a link to this thread instead. That way everyone can get the latest updates and nowadays I usually remove all old releases/add-ons from my file hosting service anyway.

Major fail on behalf of XDA news team nevertheless. It's not on the sidebar anymore and I'm pretty sure no one saw it anyway. ;)

The patch is perfectly fine. You're doing/not doing something else. I know that you are not using my sources. I really can't help you because I have no idea what code you are using.

yeah, yeah, I'm using your sources, following your second post, compile for the first time the day 25, but I tried again and do a make clobber, synchronize the repository, I apply the 2 patches, and compile, when takes time shoot me an error. It says something about the folder where the folder where you apply the patch that I mentioned. I'll try and copy the error line that gives me, the only thing that is not your sources is extracted cm.dependencies it Burstlam sources, but the file contains the same as the one I saw that you used and deleted it.

Link to comment
Share on other sites

Guest KonstaT

Here is a new test build. For this build I've enabled few hacks from burstlam. UI is now a bit faster, but there is still some flickering while scrolling. This build also breaks all video playback! I'm not pushing these changes to my github for few reasons (at least for now). I'm hoping that these are only temporary solutions, these rely on hacks on the main CM10 source tree that will never get merged, and builds are currently pretty much broken for our device anyway. I'm also keeping the previous version in the first post as some people might prefer that.

I can't believe I'm saying this but I did some kernel magic. :P I enabled triple frame buffering on our kernel and this should allow project butter to work better on our device. There is a noticeable improvement if you use boot.img from this build with the previous builds. Tearing effect while scrolling e.g. in browser is now completely gone. Also animations are now a bit more fluid.

I also updated the trimmed gapps package yesterday to include a jelly bean version of gmail that has expandable notifications.

cm-10-20120727-KonstaKANG-blade.zip

http://www.mediafire...jactkjje56sa5fi

md5:cc22d9dee8f8b98090a390421ee9b58b

-enabled triple frame buffering in kernel

-improved UI speed

-broke all video playback

-theme manager

-Hardware key custom rebinding (1,2)

-Improve scrolling cache (1)

-several improvements on browser (1,2,3)

Link to comment
Share on other sites

Guest KonstaT

yeah, yeah, I'm using your sources, following your second post, compile for the first time the day 25, but I tried again and do a make clobber, synchronize the repository, I apply the 2 patches, and compile, when takes time shoot me an error. It says something about the folder where the folder where you apply the patch that I mentioned. I'll try and copy the error line that gives me, the only thing that is not your sources is extracted cm.dependencies it Burstlam sources, but the file contains the same as the one I saw that you used and deleted it.

Sorry, I still can't understand. There is no cm.dependencies file in my sources, I removed it. Burstlam hasn't released his device tree for jelly bean.

Compile it without the patch then. It only fixes video playback on youtube/gallery/etc. You can compile without it just fine (or could before hardware/qcom/display got fcked up).

Link to comment
Share on other sites

Guest shmizan

I suggest you contact Tom regarding the triple buffering, I'd hate seeing your work go to waste if it's not merged..

as always your development is too impressive! very very nice :)

Link to comment
Share on other sites

Guest Felsch

Hmm... right I get a fault at the overlayrotator.cpp... I gave up after that... I thinkI am better of using a rom konstat delivers then making my own. I might just switch to Cm7(to compile etc) to play with as it more stable :). (not that cm9 isn't but, on cm7 more works^^)


hardware/qcom/display/liboverlay/overlayRotator.cpp: In member function 'bool overlay::MdpRot::open_i(uint32_t, uint32_t)':

hardware/qcom/display/liboverlay/overlayRotator.cpp:123:32: error: no matching function for call to 'overlay::OvMem::open(uint32_t&, uint32_t&)'

hardware/qcom/display/liboverlay/overlayRotator.cpp:123:32: note: candidate is:

hardware/qcom/display/liboverlay/overlayMem.h:117:13: note: bool overlay::OvMem::open(uint32_t, uint32_t, bool)

hardware/qcom/display/liboverlay/overlayMem.h:117:13: note:   candidate expects 3 arguments, 2 provided

make: *** [out/target/product/blade/obj/SHARED_LIBRARIES/liboverlay_intermediates/overlayRotator.o] Error 1

make: *** Waiting for unfinished jobs....

^is where I get stuck btw. Something really different then what Loren gets tho...

Link to comment
Share on other sites

Guest KonstaT

I suggest you contact Tom regarding the triple buffering, I'd hate seeing your work go to waste if it's not merged..

as always your development is too impressive! very very nice :)

My kernel source goes straight to trash if/when Tom G releases his. ;) I pretty much copied it from here anyway.

Hmm... right I get a fault at the overlayrotator.cpp... I gave up after that... I thinkI am better of using a rom konstat delivers then making my own. I might just switch to Cm7(to compile etc) to play with as it more stable :). (not that cm9 isn't but, on cm7 more works^^)


hardware/qcom/display/liboverlay/overlayRotator.cpp: In member function 'bool overlay::MdpRot::open_i(uint32_t, uint32_t)':
hardware/qcom/display/liboverlay/overlayRotator.cpp:123:32: error: no matching function for call to 'overlay::OvMem::open(uint32_t&, uint32_t&)'
hardware/qcom/display/liboverlay/overlayRotator.cpp:123:32: note: candidate is:
hardware/qcom/display/liboverlay/overlayMem.h:117:13: note: bool overlay::OvMem::open(uint32_t, uint32_t, bool)
hardware/qcom/display/liboverlay/overlayMem.h:117:13: note: candidate expects 3 arguments, 2 provided
make: *** [out/target/product/blade/obj/SHARED_LIBRARIES/liboverlay_intermediates/overlayRotator.o] Error 1
make: *** Waiting for unfinished jobs....
[/CODE]

^is where I get stuck btw. Something really different then what Loren gets tho...

Yeah, harware/qcom/display stuff is currently broken for us. You can fix the compile error by adding this one line to msm_rotator header (but you should also add it to the kernel). It will compile but it still won't work. Edit. looking it again that might be a different error...

I've reseted my HEAD in harware/qcom/display to this commit.

Link to comment
Share on other sites

Guest TheWhisp

@ above: You should update your /include/linux/msm_rotator.h (I had a similar error)

@ KonstaT: Are you also getting this?


/msm7x27.hwcomposer(  122): drawLayerUsingCopybit: copybit stretch failed

E/copybit (  122): copyBits failed (Invalid argument)

E/copybit (  122): 0: src={w=256, h=320, f=10, rect={0,0,240,301}}

E/copybit (  122):	 dst={w=256, h=320, f=10, rect={0,19,240,301}}

E/copybit (  122):	 flags=00021000

The screen is kind of yellow on lockscreen / home screen. I couldn't take screenshot, because they appear correct..

Link to comment
Share on other sites

Guest KonstaT

@ above: You should update your /include/linux/msm_rotator.h (I had a similar error)

@ KonstaT: Are you also getting this?


/msm7x27.hwcomposer( 122): drawLayerUsingCopybit: copybit stretch failed
E/copybit ( 122): copyBits failed (Invalid argument)
E/copybit ( 122): 0: src={w=256, h=320, f=10, rect={0,0,240,301}}
E/copybit ( 122): dst={w=256, h=320, f=10, rect={0,19,240,301}}
E/copybit ( 122): flags=00021000[/CODE]

The screen is kind of yellow on lockscreen / home screen. I couldn't take screenshot, because they appear correct..

Yeah, that's what I'm getting too (with bigger screen resolution though ;)). Bootanimation is just black screen and I can't even get past the lockscreen because display just sort of hangs.

Link to comment
Share on other sites

Guest matt4321

I want to clear up the XDA issue (i'm matt4321 on XDA), since I started the thread I gave credits to you and tilal considering it's you guys doing this. When the article was written on XDA saying it was my work I immediately contacted the person who wrote it to say it's not my work but mainly KonstaTs. I haven't heard anything back since. To make it more clear I since added the modaco thread in order for people to see fully that this isn't my rom but yours. I will go over now and edit it to say in writing that this is your rom. I thought at first I had made myself clear on the thread that it wasn't my rom but I have added more and more to clarify that it isn't. I absolutely do not intend to be stealing or taking credit for peoples work, that it is completely against what I stand for. I'm very sorry for any misunderstanding about this. KonstaT I am actually big fan and appreciate all the work you do! I totally understand if you want me to take the thread down, just ask and I will remove it immediately. Once again, sorry!

Link to comment
Share on other sites

Guest ufizo

I've installed this great ROM yesterday (of course with full wipe) and I know its still (pre)alpha, but there is an issue that is also in your CM9 version:

when starting an App (like facebook, maps,...) that uses GPS (or you could say: when starting GPS), very often my mobile data connection gets lost and needs a few minutes and/or toggling 2G/3G and data on/off to connect again.

+1

I have the exact same issue, and believed that it is some problem at my ISP's end, but it seems that I am not alone.

I never get to open gmaps and latitude without my connection dropping off, in a few minutes it works,.. but it becomes useless while travelling. With WiFi, things are fine.

Edited by ufizo
Link to comment
Share on other sites

Guest emc02

+1

I have the exact same issue, and believed that it is some problem at my ISP's end, but it seems that I am not alone.

I dont think its a problem of the ISP, because in tilals builds there is no problem with that (and how could the ISP be affected by your GPS Module?)

when disabling GPS manually there is no problem with disconnecting mobile data and everything works fine...

Link to comment
Share on other sites

Guest KonstaT

I want to clear up the XDA issue (i'm matt4321 on XDA), since I started the thread I gave credits to you and tilal considering it's you guys doing this. When the article was written on XDA saying it was my work I immediately contacted the person who wrote it to say it's not my work but mainly KonstaTs. I haven't heard anything back since. To make it more clear I since added the modaco thread in order for people to see fully that this isn't my rom but yours. I will go over now and edit it to say in writing that this is your rom. I thought at first I had made myself clear on the thread that it wasn't my rom but I have added more and more to clarify that it isn't. I absolutely do not intend to be stealing or taking credit for peoples work, that it is completely against what I stand for. I'm very sorry for any misunderstanding about this. KonstaT I am actually big fan and appreciate all the work you do! I totally understand if you want me to take the thread down, just ask and I will remove it immediately. Once again, sorry!

Thanks, it's all good man. I really don't care that much what goes on in XDA and I was more amused than offended by it the whole time. ;) It was the fault of the XDA news team anyway. I'm sure there's threads like that on almost every Blade forum so you don't have to worry about that.

Link to comment
Share on other sites

Guest KonstaT

Hi Konstat,

I've installed this great ROM yesterday (of course with full wipe) and I know its still (pre)alpha, but there is an issue that is also in your CM9 version:

when starting an App (like facebook, maps,...) that uses GPS (or you could say: when starting GPS), very often my mobile data connection gets lost and needs a few minutes and/or toggling 2G/3G and data on/off to connect again.

This issue is only in your ROMS and was not in tilals coldfusion CM9. Maybe you know what you changed and what could cause that.

All other things are great! Feels much smoother than CM9!

+1

I have the exact same issue, and believed that it is some problem at my ISP's end, but it seems that I am not alone.

I never get to open gmaps and latitude without my connection dropping off, in a few minutes it works,.. but it becomes useless while travelling. With WiFi, things are fine.

Sorry, I can't reproduce this. I use GPS mainly for tracking and I haven't noticed this. Can you get a logcat? How does the connection get cut off? Connection gets dropped or you're still connected but data doesn't move?

I don't think there is any difference in mine and CFX device trees that could explain this. Maybe some issue in upstream CM9/10 code that has appeared after the latest CFX builds?

Link to comment
Share on other sites

Guest reload99

The battery duration on this ROM is better than battery duration of Cyanogenmod 7.2? What's the difference between GAAPs medium and Full? The latest version of this ROM is the one which is in the first page?

Link to comment
Share on other sites

Guest Loren82

Hmm... right I get a fault at the overlayrotator.cpp... I gave up after that... I thinkI am better of using a rom konstat delivers then making my own. I might just switch to Cm7(to compile etc) to play with as it more stable :). (not that cm9 isn't but, on cm7 more works^^)


hardware/qcom/display/liboverlay/overlayRotator.cpp: In member function 'bool overlay::MdpRot::open_i(uint32_t, uint32_t)':

hardware/qcom/display/liboverlay/overlayRotator.cpp:123:32: error: no matching function for call to 'overlay::OvMem::open(uint32_t&, uint32_t&)'

hardware/qcom/display/liboverlay/overlayRotator.cpp:123:32: note: candidate is:

hardware/qcom/display/liboverlay/overlayMem.h:117:13: note: bool overlay::OvMem::open(uint32_t, uint32_t, bool)

hardware/qcom/display/liboverlay/overlayMem.h:117:13: note:   candidate expects 3 arguments, 2 provided

make: *** [out/target/product/blade/obj/SHARED_LIBRARIES/liboverlay_intermediates/overlayRotator.o] Error 1

make: *** Waiting for unfinished jobs....

^is where I get stuck btw. Something really different then what Loren gets tho...

It solves the above problem was my fault to make a false step, now I get stuck at the same point you are getting compile by adding the line "unsigned int secure;" to "include / linux / msm_rotator.h"?

Link to comment
Share on other sites

Guest emc02

Sorry, I can't reproduce this. I use GPS mainly for tracking and I haven't noticed this. Can you get a logcat? How does the connection get cut off? Connection gets dropped or you're still connected but data doesn't move?

the signalbar is getting grey but sometimes shows an up arrow, but no down arrow

I opened facebook-app with gps enabled, then connection got lost and about a minute later connection is back (should all be in the logcat)

maybe this helps

thanks a lot!

alogcat.2012-07-27-19-11-58+0200.txt

Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...

Important Information

By using this site, you agree to our Terms of Use.