Guest Daedric1383 Posted August 28, 2012 Report Share Posted August 28, 2012 Uhm... BTW: D/pppd ( 1143): rcvd [LCP ProtRej id=0x2 80 fd 01 01 00 0f 1a 04 78 00 18 04 78 00 15 03 2f] I/pppd ( 1143): Modem hangup Jan 12 18:56:48 bozo-laptop pppd[26066]: rcvd [LCP ProtRej id=0x2 80 fd 01 01 00 0f 1a 04 78 00 18 04 78 00 15 03 2f] Jan 12 18:56:48 bozo-laptop pppd[26066]: Protocol-Reject for 'Compression Control Protocol' (0x80fd) received Link to comment Share on other sites More sharing options...
Guest ejtagle Posted August 28, 2012 Report Share Posted August 28, 2012 (edited) Uhm... BTW: Ok, let's try disabing compression... ;) edit1: Attached a version to try that disables data compression. I haven't even compiled it, so, maybe it does not compile as it is... I have no access to my development machine right now. Later i will try it myself, but, i post it right now nevertheless, so if someone wants to try it before me, it's welcome :) I have been also working on improving voice call support... Its hard to create a proper resampling algorithm that is simple, but flexible enough to self adjust between sampling rates as provided by the 3g modem, and requested by android audio subsystem... So, please take voice as a work in progress thing, and report if it seems to work or not...huawei-generic-28aug.rar Edited August 28, 2012 by ejtagle Link to comment Share on other sites More sharing options...
Guest univall Posted August 28, 2012 Report Share Posted August 28, 2012 Ok, let's try disabing compression... ;) edit1: Attached a version to try that disables data compression. I haven't even compiled it, so, maybe it does not compile as it is... I have no access to my development machine right now. Later i will try it myself, but, i post it right now nevertheless, so if someone wants to try it before me, it's welcome :) I have been also working on improving voice call support... Its hard to create a proper resampling algorithm that is simple, but flexible enough to self adjust between sampling rates as provided by the 3g modem, and requested by android audio subsystem... So, please take voice as a work in progress thing, and report if it seems to work or not... Does not build. target thumb C++: libhuaweigeneric-ril <= device/nvidia/shuttle/huawei-generic/audiochannel.cpp target thumb C: libhuaweigeneric-ril <= device/nvidia/shuttle/huawei-generic/audioqueue.c target thumb C: libhuaweigeneric-ril <= device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c Export includes file: device/nvidia/shuttle/huawei-generic/Android.mk -- out/target/product/shuttle/obj/SHARED_LIBRARIES/libhuaweigeneric-ril_intermediates/export_includes device/nvidia/shuttle/huawei-generic/audiochannel.cpp: In function 'void* modemAudioIOThread(void*)': device/nvidia/shuttle/huawei-generic/audiochannel.cpp:84:15: error: 'fd' was not declared in this scope make: *** [out/target/product/shuttle/obj/SHARED_LIBRARIES/libhuaweigeneric-ril_intermediates/audiochannel.o] Error 1 make: *** Waiting for unfinished jobs.... device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: missing braces around initializer [-Wmissing-braces] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: (near initialization for 'sAudioChannel.play_q.mtx') [-Wmissing-braces] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: missing initializer [-Wmissing-field-initializers] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: (near initialization for 'sAudioChannel.play_q.rd_cond') [-Wmissing-field-initializers] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: missing initializer [-Wmissing-field-initializers] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: (near initialization for 'sAudioChannel.rec_q.rd_cond') [-Wmissing-field-initializers] Link to comment Share on other sites More sharing options...
Guest Daedric1383 Posted August 28, 2012 Report Share Posted August 28, 2012 If it matters: It can be done :) Link to comment Share on other sites More sharing options...
Guest Scanno Posted August 28, 2012 Report Share Posted August 28, 2012 If it matters: It can be done :) I know... but there are some issues that should be resolved before it can go mainstream. 1. Nvflash: is it possible to flash a data image to the sdcard partition? 2. Wiping and installing from cwm needs a new cwm version that mounts the data partition on the sdcard and not the partition on internal nand. If these issue are not solved it is not useable for the regular user. Link to comment Share on other sites More sharing options...
Guest ejtagle Posted August 28, 2012 Report Share Posted August 28, 2012 Does not build. target thumb C++: libhuaweigeneric-ril <= device/nvidia/shuttle/huawei-generic/audiochannel.cpp target thumb C: libhuaweigeneric-ril <= device/nvidia/shuttle/huawei-generic/audioqueue.c target thumb C: libhuaweigeneric-ril <= device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c Export includes file: device/nvidia/shuttle/huawei-generic/Android.mk -- out/target/product/shuttle/obj/SHARED_LIBRARIES/libhuaweigeneric-ril_intermediates/export_includes device/nvidia/shuttle/huawei-generic/audiochannel.cpp: In function 'void* modemAudioIOThread(void*)': device/nvidia/shuttle/huawei-generic/audiochannel.cpp:84:15: error: 'fd' was not declared in this scope make: *** [out/target/product/shuttle/obj/SHARED_LIBRARIES/libhuaweigeneric-ril_intermediates/audiochannel.o] Error 1 make: *** Waiting for unfinished jobs.... device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: missing braces around initializer [-Wmissing-braces] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: (near initialization for 'sAudioChannel.play_q.mtx') [-Wmissing-braces] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: missing initializer [-Wmissing-field-initializers] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: (near initialization for 'sAudioChannel.play_q.rd_cond') [-Wmissing-field-initializers] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: missing initializer [-Wmissing-field-initializers] device/nvidia/shuttle/huawei-generic/huaweigeneric-ril.c:247:15: warning: (near initialization for 'sAudioChannel.rec_q.rd_cond') [-Wmissing-field-initializers] This hould fix the compilation error ...huawei-generic-28aug-v2.rar Link to comment Share on other sites More sharing options...
Guest Daedric1383 Posted August 28, 2012 Report Share Posted August 28, 2012 I know... but there are some issues that should be resolved before it can go mainstream. 1. Nvflash: is it possible to flash a data image to the sdcard partition? 2. Wiping and installing from cwm needs a new cwm version that mounts the data partition on the sdcard and not the partition on internal nand. If these issue are not solved it is not useable for the regular user. 1. With so much space for system, why would you write to the SDcard / Data ? 2. CWM sees the second partition on the sdcard has a sd-ext, and backups it as such. On a restore, you get your /data restored. Link to comment Share on other sites More sharing options...
Guest ejtagle Posted August 28, 2012 Report Share Posted August 28, 2012 I know... but there are some issues that should be resolved before it can go mainstream. 1. Nvflash: is it possible to flash a data image to the sdcard partition? 2. Wiping and installing from cwm needs a new cwm version that mounts the data partition on the sdcard and not the partition on internal nand. If these issue are not solved it is not useable for the regular user. 1) No, it is not possible... NVflash can only flash to one device at a time, and it must be the boot device. In our case, nand... But, it could be done with a 2 step procedure... After all, the android image can simply be created as a ext4fs image, and then just be copied to a properly partitioned SD card... 2) Yes, but probably CWM can be configured to do it... Are we absolutely sure the problems are caused by bad NAND blocks ? ... We could force nand verification after writes, but that costs time and also write performance suffers... Link to comment Share on other sites More sharing options...
Guest univall Posted August 28, 2012 Report Share Posted August 28, 2012 Yes, compiles OK. Will revert on results soon. Link to comment Share on other sites More sharing options...
Guest univall Posted August 28, 2012 Report Share Posted August 28, 2012 This hould fix the compilation error ... Ok. No 3g data. Issues with making calls and hanging up. Similar to earlier situations. I have tried to capture data between a few reboots, making call, etc. Hopefully will reveal important clues. Sorry for bringing back negative news. Hope it will be a better report next time :-) Cheers!full-28812.txtradio-28812.txt Link to comment Share on other sites More sharing options...
Guest Scanno Posted August 28, 2012 Report Share Posted August 28, 2012 1) No, it is not possible... NVflash can only flash to one device at a time, and it must be the boot device. In our case, nand... But, it could be done with a 2 step procedure... After all, the android image can simply be created as a ext4fs image, and then just be copied to a properly partitioned SD card... 2) Yes, but probably CWM can be configured to do it... Are we absolutely sure the problems are caused by bad NAND blocks ? ... We could force nand verification after writes, but that costs time and also write performance suffers... The CWM recovery can be done.... i made the touch recovery, i guess you could do the same kind of thing to the ramdisk of the recovery kernel. Or make some kind of choice before flashing (mount /data to second sd partition). Link to comment Share on other sites More sharing options...
Guest ejtagle Posted August 28, 2012 Report Share Posted August 28, 2012 Ok. No 3g data. Issues with making calls and hanging up. Similar to earlier situations. I have tried to capture data between a few reboots, making call, etc. Hopefully will reveal important clues. Sorry for bringing back negative news. Hope it will be a better report next time :-) Cheers! According to those logs, pppd has successfully negotiated a connection. The RIL is expecting pppd to run the ip-up script and properly set the following properties to recognize a connection was successfully set up: net.ril1.if net.ril1.local-ip net.ril1.dns1 net.ril1.dns2 net.ril1.gw Please, check the ip-up script to make sure it is executable, and to make sure it is suid root. If this scripto is not running, the RIL will never know a connection was successfull... Link to comment Share on other sites More sharing options...
Guest Daedric1383 Posted August 28, 2012 Report Share Posted August 28, 2012 (edited) 1) No, it is not possible... NVflash can only flash to one device at a time, and it must be the boot device. In our case, nand... But, it could be done with a 2 step procedure... After all, the android image can simply be created as a ext4fs image, and then just be copied to a properly partitioned SD card... 2) Yes, but probably CWM can be configured to do it... Are we absolutely sure the problems are caused by bad NAND blocks ? ... We could force nand verification after writes, but that costs time and also write performance suffers... The CWM recovery can be done.... i made the touch recovery, i guess you could do the same kind of thing to the ramdisk of the recovery kernel. Or make some kind of choice before flashing (mount /data to second sd partition). CWM does not need tampering. If it behaves exactly has on other devices, the second partition of the sdcard will be backed up and restored. CWM sees it as sdext, Android will see it as /data. According to those logs, pppd has successfully negotiated a connection. The RIL is expecting pppd to run the ip-up script and properly set the following properties to recognize a connection was successfully set up: net.ril1.if net.ril1.local-ip net.ril1.dns1 net.ril1.dns2 net.ril1.gw Please, check the ip-up script to make sure it is executable, and to make sure it is suid root. If this scripto is not running, the RIL will never know a connection was successfull... Could compiled files be posted please? I would like to test ... Edited August 28, 2012 by Daedric1383 Link to comment Share on other sites More sharing options...
Guest univall Posted August 29, 2012 Report Share Posted August 29, 2012 (edited) CWM does not need tampering. If it behaves exactly has on other devices, the second partition of the sdcard will be backed up and restored. CWM sees it as sdext, Android will see it as /data. Could compiled files be posted please? I would like to test ... Edited August 29, 2012 by univall Link to comment Share on other sites More sharing options...
Guest univall Posted August 29, 2012 Report Share Posted August 29, 2012 (edited) CWM does not need tampering. If it behaves exactly has on other devices, the second partition of the sdcard will be backed up and restored. CWM sees it as sdext, Android will see it as /data. Could compiled files be posted please? I would like to test ... ejtagle: Could you please post exact permissions/suid on above files. I will check in the terminal if it changes the situation. daedric123: attached compiled lib EDIT: added a few more logs/info dumps. Hope this helps.libhuawei.tar.gznetcfg.txtpppd_meminfo.txtpppd_ps.txtpppd_dmesg.txt Edited August 29, 2012 by univall Link to comment Share on other sites More sharing options...
Guest ejtagle Posted August 29, 2012 Report Share Posted August 29, 2012 ejtagle: Could you please post exact permissions/suid on above files. I will check in the terminal if it changes the situation. daedric123: attached compiled lib EDIT: added a few more logs/info dumps. Hope this helps. netcfg gives the answer... ppp0 interface is UP and properly configured :) ... It must be something related to the ip-up script. Use getprop to get the system properties... We need to know if the ip-up script is being executed, The reason is that ip-up uses setprop to set some android properties with critical data required by the RIL to detect that the ppp0 connection is UP, and the DNS to be used. And the setprop requires root access to work.... Link to comment Share on other sites More sharing options...
Guest univall Posted August 29, 2012 Report Share Posted August 29, 2012 netcfg gives the answer... ppp0 interface is UP and properly configured :) ... It must be something related to the ip-up script. Use getprop to get the system properties... We need to know if the ip-up script is being executed, The reason is that ip-up uses setprop to set some android properties with critical data required by the RIL to detect that the ppp0 connection is UP, and the DNS to be used. And the setprop requires root access to work.... Here it is: getprop.txtgetprop.txt Link to comment Share on other sites More sharing options...
Guest ejtagle Posted August 29, 2012 Report Share Posted August 29, 2012 Here it is: getprop.txt The ip-up script is not executing, or has insufficient privileges to set properties... Link to comment Share on other sites More sharing options...
Guest ejtagle Posted August 29, 2012 Report Share Posted August 29, 2012 Posting the ril i am working on... voice is half working... No other changes.. There seems to be a race condition on uninitialization of the audio bridge that sometimes locks the ril... Still trying to figure the exact cause, though...huawei-generic-29aug-v3.rar Link to comment Share on other sites More sharing options...
Guest univall Posted August 29, 2012 Report Share Posted August 29, 2012 The ip-up script is not executing, or has insufficient privileges to set properties... Maybe Scanno could look in to this? Link to comment Share on other sites More sharing options...
Guest DerArtem Posted August 29, 2012 Report Share Posted August 29, 2012 Hi, have finally ported jellybean to the Folio 100. Many thanks for the nice work - especially the audio drivers ;) I have still a few issue to work out but all in all it work pretty fine right now. Now I am wondering about the source of the nvidia blobs you are using in https://github.com/s.../jb/proprietary. I have extracted the nvidia blobs on my github - with the rotation bug on jellybean - from the nvidia ventana ics image on http://developer.nvi...ent-kit-updates. The were released on 19.04.2012. Now there is a new image available which is dated with 21.05.2012. Has somebody already extracted and tried using them? Where are your nvidia blobs from? Link to comment Share on other sites More sharing options...
Guest univall Posted August 29, 2012 Report Share Posted August 29, 2012 (edited) Posting the ril i am working on... voice is half working... No other changes.. There seems to be a race condition on uninitialization of the audio bridge that sometimes locks the ril... Still trying to figure the exact cause, though... @ejtagle: I understand that you are still working on this. Yet, compiled this version - in the practical sense, nothing seems to have changed, in the small no. of tests (making and receiving calls) that I did. Attaching the usual radio and full logs, hoping these could be of some use to you to debug.log298.txtradio298.txt Edited August 29, 2012 by univall Link to comment Share on other sites More sharing options...
Guest ejtagle Posted August 29, 2012 Report Share Posted August 29, 2012 Hi, have finally ported jellybean to the Folio 100. Many thanks for the nice work - especially the audio drivers ;) I have still a few issue to work out but all in all it work pretty fine right now. Now I am wondering about the source of the nvidia blobs you are using in https://github.com/s.../jb/proprietary. I have extracted the nvidia blobs on my github - with the rotation bug on jellybean - from the nvidia ventana ics image on http://developer.nvi...ent-kit-updates. The were released on 19.04.2012. Now there is a new image available which is dated with 21.05.2012. Has somebody already extracted and tried using them? Where are your nvidia blobs from? Not tried them yet. We were using, as far as i know, blobs extracted, probably from the picasso ROM... The ones in the ICS image have a rotation bug (in fact, seems they don't properly implement rotation of big textures in opengles, and that is causing the rotation bug)... With a bit of luck, the new ICS image will fix that opengles bug and will be usable on JB... Link to comment Share on other sites More sharing options...
Guest ejtagle Posted August 29, 2012 Report Share Posted August 29, 2012 Maybe Scanno could look in to this? not the best solution right now, but try doing chmod 4755 /system/etc/ppp/ip-up chmod 4755 /system/etc/ppp/ip-down chmod 4755 /system/bin/setprop Link to comment Share on other sites More sharing options...
Guest univall Posted August 29, 2012 Report Share Posted August 29, 2012 (edited) not the best solution right now, but try doing chmod 4755 /system/etc/ppp/ip-up chmod 4755 /system/etc/ppp/ip-down chmod 4755 /system/bin/setprop Attached logs captured as I do the above. And getprop after it. No difference after the above commands. What should we be doing next?log298-cmd.txtn-getprop.txtradio298-cmd.txt Edited August 30, 2012 by univall Link to comment Share on other sites More sharing options...
Recommended Posts
Please sign in to comment
You will be able to leave a comment after signing in
Sign In Now