**** BEGIN LOGGING AT Fri Aug 17 02:59:58 2012 Aug 17 09:23:32 SHR: 03Martin.Jansa 07meta-smartphone * ra437467d6650 10/meta-shr/ (2 files in 2 dirs): meta-shr: use gpsd as gpsd provider also on om-gta02 (instead of fso-gpsd) Aug 17 10:51:27 SHR: 03GNUtoo 07meta-smartphone * r195f4289bf9c 10/meta-samsung/recipes-kernel/linux/ (linux-samsung-tuna/tuna/defconfig linux-samsung-tuna_git.bb): meta-samsung: linux-samsung-tuna: update defconfig and SRCREV. Aug 17 11:34:25 paulk-desktop, Alex[sp3dev] ,hi Aug 17 11:34:40 I want to make tuna's modem work Aug 17 11:34:49 under FSO/SHR Aug 17 11:35:29 hi GNUtoo|laptop Aug 17 11:35:48 GNUtoo|laptop, did it fail? Aug 17 11:35:58 where's the low level in the fso sense in android? Aug 17 11:36:02 like modem init Aug 17 11:36:08 in libsamsung-ipc Aug 17 11:36:17 but I think you just have to include libsamsung-ipc in tuna SHR Aug 17 11:36:19 and it will work Aug 17 11:36:20 like you send efs etc... Aug 17 11:36:28 ah really? Aug 17 11:36:32 I've that: Aug 17 11:36:32 (it does board detection) Aug 17 11:36:37 ah wow Aug 17 11:36:48 so the fso settings are useless Aug 17 11:36:51 maybe you'll have to mount /efs/ Aug 17 11:37:21 oh actually it's called /factory/ Aug 17 11:37:23 on galaxy nexus Aug 17 11:37:26 https://github.com/morphis/libsamsung-ipc/blob/master/samsung-ipc/device/maguro/maguro_ipc.c Aug 17 11:37:29 I mean there are stuff like /dev/modem_fmt or similar for crespo in FSO Aug 17 11:38:23 /dev/umts_boot1 Aug 17 11:38:34 ok Aug 17 11:38:53 and also: Aug 17 11:39:00 /dev/umts_ipc0 Aug 17 11:39:03 /dev/umts_rfs0 Aug 17 11:39:18 and some sysfs for modemctrl Aug 17 11:39:23 do you have these? Aug 17 11:39:45 SHR: 03Martin.Jansa 07shr-chroot * r85ac8bce92f8 10/ (16 files in 8 dirs): bitbake upgrade Aug 17 11:44:30 2012-08-17T11:44:15.786407Z [WARN] libfsotransport : could not open /dev/modem_fmt: No such file or directory Aug 17 11:44:43 which is which? Aug 17 11:45:51 ah wait Aug 17 11:45:53 you're right Aug 17 11:45:59 there is libfsotransport Aug 17 11:46:02 I forgot that Aug 17 11:46:15 so yes, you have to reimplement some of the libsamsung-ipc functions Aug 17 11:46:33 the problem is that morphis is probably still on holliday Aug 17 11:46:49 I guess I can help you if you're ready to do it Aug 17 11:46:55 ok Aug 17 11:47:04 let me sync cornucopia Aug 17 11:47:04 why not use libsamsung-ipc directly? Aug 17 11:47:13 ok Aug 17 11:47:35 GNUtoo|laptop, morphis wanted to use fsotransportd Aug 17 11:47:41 for some reason… Aug 17 11:48:16 ok Aug 17 11:48:50 JaMa, how big is the newer enlightenment? Aug 17 11:49:41 ? you mean disk space? Aug 17 11:49:48 where is the per-device configuration for fsotransport? Aug 17 11:49:55 about the same as few days old alpha Aug 17 11:50:35 JaMa, I mean I'm on 3g with 4GB limit for 2 mohtns Aug 17 11:50:35 *months Aug 17 11:50:45 paulk-desktop, fsogsmd/conf/ Aug 17 11:50:46 found it Aug 17 11:50:49 yes Aug 17 11:51:40 GNUtoo|laptop: ah download? 40MB tarballs + something for svn upgrade (depends on how old is your checkout) Aug 17 11:51:49 ok Aug 17 11:51:51 thanks Aug 17 11:52:19 so the functions in libsamsung-ipc are: https://github.com/morphis/libsamsung-ipc/blob/master/samsung-ipc/device/maguro/maguro_ipc.c#L45 Aug 17 11:53:07 ok Aug 17 11:53:14 which are at: https://github.com/morphis/libsamsung-ipc/blob/master/samsung-ipc/device/xmm6260/xmm6260_ipc.c#L215 Aug 17 11:53:14 I reimplement that in vala? Aug 17 11:53:19 in the lowlevel plugin Aug 17 11:53:23 yes I saw that Aug 17 11:53:38 it's function pointers to more commons functions Aug 17 11:53:50 I looked some time ago for mirage* Aug 17 11:53:56 ah ok Aug 17 11:54:28 but I'm not sure replacing modem_access in the config file will work Aug 17 11:54:35 indeed Aug 17 11:55:14 my question was that there were stuff duplicated between libsamsung-ipc and lowlevel-crespo Aug 17 11:55:20 so I wonder what to do Aug 17 11:55:40 but I guess I'll have to implement a lowlevel-tuna Aug 17 11:56:53 yes you must reimplement all in lowlevel-tuna Aug 17 11:57:10 because it's designed so that the functions in libsamsung-ipc are overwritten by fsogsmd's functions Aug 17 11:57:22 ok Aug 17 11:57:39 but why is there copyright morphis on the file you pointed me to then? Aug 17 11:57:42 he used android too? Aug 17 11:58:24 with libsamsung-ipc? Aug 17 11:59:02 it says it's because it's from crespo file, and he did write code in crespo ipc file Aug 17 11:59:11 because there are also shared functions (for android and SHR) there Aug 17 11:59:18 so he has written some of these Aug 17 11:59:23 but no the I/O handlers Aug 17 11:59:26 not* Aug 17 11:59:29 ok Aug 17 11:59:44 ok Aug 17 12:00:13 so apparently open/close/read/write is in modem_samsung Aug 17 12:00:20 while it should be in lowlevel_ Aug 17 12:00:29 because these are device-specific too Aug 17 12:02:01 maguro uses read() and not ioctl Aug 17 12:02:31 so I suggest that you move samsungtransport.vala to lowlevel_samsung_crespo Aug 17 12:02:38 and create another one for tuna Aug 17 12:03:03 hmmm Aug 17 12:03:46 morphis, ping Aug 17 12:04:00 I'm not sure it's the right thing tough Aug 17 12:04:11 morphis, should know what's the best Aug 17 12:04:14 but he's not there Aug 17 12:04:49 indeed Aug 17 12:05:52 maybe I should do what you advise....but in a branch... Aug 17 12:06:12 yes, sure Aug 17 12:06:20 morphis will probably want to review a change that big anyway Aug 17 12:07:09 indeed Aug 17 12:07:09 that's why I was not confortable with it Aug 17 12:07:50 maybe I hack crespo's stuff to make it work on tuna right now Aug 17 12:07:50 and try to integrate later? Aug 17 12:07:50 :q Aug 17 12:07:52 oops Aug 17 12:08:45 that's what I would do sincerely Aug 17 12:09:28 ok Aug 17 12:10:15 btw tuna's gps may need your gps program too Aug 17 12:10:47 I tried wakelock + echo sysfs and nothing appeared on gpspipe -w Aug 17 12:11:17 how is it? a serial? Aug 17 12:12:24 yes Aug 17 12:13:14 RESET_GPIO=/sys/devices/virtual/sec/gps/GPS_nRST/value Aug 17 12:13:17 ONOFF_GPIO=/sys/devices/virtual/sec/gps/GPS_PWR_EN/value Aug 17 12:13:17 and: Aug 17 12:13:29 UART_DRIVER=/dev/ttyO0 Aug 17 12:14:20 did you see what kind of data is coming out of the serial? Aug 17 12:15:52 no Aug 17 12:15:55 I'll look Aug 17 12:19:32 nothing.... Aug 17 12:20:38 okay Aug 17 12:20:48 so I guess we'll have to trace the non-free binary, etc Aug 17 12:20:59 yes Aug 17 12:24:26 what program should I trace? Aug 17 12:24:39 is there a daemon? Aug 17 12:25:01 let me look Aug 17 12:26:42 ./lib/hw/gps.omap4.so Aug 17 12:27:34 in /system/vendor Aug 17 12:27:34 I guess I use replicant 2.3 gps_test ? Aug 17 12:28:00 can you cat: /system/vendor/etc/sirfgps.conf Aug 17 12:28:13 let me check the one to use Aug 17 12:28:58 use bcm4751_hal Aug 17 12:29:00 and starce it Aug 17 12:29:04 strace* Aug 17 12:30:22 http://pastie.org/private/a4jduhrxvnc0vstcprdjg Aug 17 12:30:34 ok thanks Aug 17 12:31:14 strange, baud rate is 460800 Aug 17 12:32:38 lol ok Aug 17 12:33:04 where's that Aug 17 12:33:09 I looked for that and missed it Aug 17 12:33:23 UART_BAUD_RATE=2 Aug 17 12:34:23 ahh Aug 17 12:34:25 #UART_BAUD_RATE : 0 - 115200bps 1 - 230400, 2 - 460800, 3 - 57600 Aug 17 12:34:26 yep and: #UART_BAUD_RATE : 0 - 115200bps 1 - 230400, 2 - 460800, 3 - 57600 Aug 17 12:36:46 I've stuff on it Aug 17 12:37:01 00001090 78 00 00 f8 00 f8 80 00 78 f8 00 00 f8 00 00 78 |x.......x......x| Aug 17 12:37:31 wow Aug 17 12:38:18 gpsd:UNK: ISGPS syncing at byte 1751: 0x30c3aeb0 Aug 17 12:38:25 but nothing on gpspipe -w Aug 17 12:38:28 nor -R Aug 17 12:38:30 ok Aug 17 12:38:38 *-r Aug 17 12:38:46 then strace if you can Aug 17 12:38:56 ok Aug 17 12:39:16 or I'll try reseting it Aug 17 12:47:24 am I right that fsogsmd (as usual) uses some bikeshed and tries to do ipc and power itself instead of reusing libsamsung-ipc? NIH in everything? Aug 17 12:52:52 Alex[sp3dev], definitely, that's what it does :) Aug 17 12:53:27 well actually the only thing it does by itself instead of reusing libsamsung-ipc is I/O handlers Aug 17 12:59:36 paulk-desktop, what was the strace option for printing everything (I already used it with the serprog-duino project but I can't remember nor find it in the man) Aug 17 13:00:14 I mean for full strings Aug 17 13:00:15 -s ? Aug 17 13:00:39 GNUtoo|laptop, usually, I use that line: strace -f -v -x -s 512 bin Aug 17 13:00:47 yes -s with a alarge value Aug 17 13:00:53 ok Aug 17 13:25:10 ~nf Aug 17 13:25:10 The #openmoko-cdevel Newsflash Bulletin Board. (continued at ~NF2. For help see ~NF-help) - - - Recommended URLs and channels(chanlogs etc):see ~RL - - - NEWSFLASH [2012-03-28] server "agni.openmoko.org" got new hw, so no more downtime expected Aug 17 13:25:43 JesusMcCloud, hi Aug 17 13:26:03 hmmm I can't reproduce Aug 17 13:30:53 SHR: 03shr-devel 07buildhistory * r474e2bbc0d6e 10/ (400 files in 400 dirs): packages: Build 201208171350 of shr 20120817 for machine om-gta02 on opmbuild Aug 17 14:33:58 paulk-desktop, I cannot get the logs anymore Aug 17 14:34:35 which logs? Aug 17 14:37:10 oops Aug 17 14:37:11 I mean output Aug 17 14:37:13 I had huge ouptut on /dev/ttyO0 Aug 17 14:37:31 ah ok Aug 17 14:37:47 now I cannot reproduce that Aug 17 14:38:21 data from strace would be more useful anyway Aug 17 14:38:40 http://pastie.org/private/0biju86iwvvjd3pnlyhfbw Aug 17 14:38:57 so I did Aug 17 14:39:07 echo 1 > /sys/devices/virtual/sec/gps/GPS_nRST/value Aug 17 14:39:13 echo 0 > /sys/devices/virtual/sec/gps/GPS_PWR_EN/value Aug 17 14:39:15 etc... Aug 17 14:39:32 ok I see Aug 17 14:39:44 does it open the serial on the trace? Aug 17 14:39:49 apparently not Aug 17 14:40:21 ahh maybe that's it? Aug 17 14:40:24 because I keep the serial port open Aug 17 14:40:29 strace.958:open("/dev/ttyO0", O_RDWR|O_NOCTTY) = 12 Aug 17 14:40:55 ok then what's next on the trace? Aug 17 14:41:17 or could you paste/upload the whole trace please? Aug 17 14:41:44 note the trace number Aug 17 14:41:55 it's in another thread Aug 17 14:42:06 yes, I saw that happening on crespo's gpsd Aug 17 14:42:13 I don't know why it does that Aug 17 14:42:14 yes Aug 17 14:42:15 let me upload the tarball Aug 17 14:42:19 thanks Aug 17 14:45:57 http://gnutoo.homelinux.org/downloads/people/paulk/strace.tar.bz2 Aug 17 14:48:54 so yeah, I assume it's sending a firmware first Aug 17 14:49:20 ok Aug 17 14:50:39 yes that's what I tought at the very beginning when I got nothing on serial, I tough that it was similar to crespo Aug 17 14:51:04 is the firmware on the filesystem somewhere Aug 17 14:51:10 or is it in the lib? Aug 17 14:51:45 I don't know Aug 17 14:52:53 btw did you kill the process or did it die? Aug 17 14:53:19 no I ctrl+c it Aug 17 14:53:23 under adb Aug 17 14:53:34 under cm with android 4.x Aug 17 14:53:37 mhh, maybe let it finish Aug 17 14:53:45 the hal thing should end it after a minute or so Aug 17 14:53:47 ok Aug 17 14:53:58 so that we can see a bit of the protocol Aug 17 14:59:46 ok Aug 17 14:59:56 should I put it near the window? Aug 17 15:00:40 because I'm in a room without window right now Aug 17 15:01:29 for instance some stuff could appear only when the GPS had data? Aug 17 15:01:40 like a fix Aug 17 15:04:36 I don't know but when Alex traced his galaxy s2, there was a lot of data coming Aug 17 15:06:38 SHR: 03shr-devel 07buildhistory * r7c32ad75eb48 10/ (393 files in 393 dirs): packages: Build 201208171540 of shr 20120817 for machine nokia900 on opmbuild Aug 17 15:08:57 ok Aug 17 15:09:00 I'll try Aug 17 15:09:06 *retry Aug 17 15:09:12 let's move the laptop Aug 17 15:09:54 GNUtoo|laptop: the gps seems to be an arm system Aug 17 15:10:55 GNUtoo|laptop: not too many strings: http://paste.debian.net/184207/ Aug 17 15:15:53 ok Aug 17 15:16:26 lindi-, are you able to extract the firmware? Aug 17 15:16:39 GNUtoo|laptop: yeah Aug 17 15:17:42 how? Aug 17 15:17:50 what dd command? Aug 17 15:17:58 it's on tuna Aug 17 15:20:36 GNUtoo|laptop: tuna? Aug 17 15:21:02 maguro Aug 17 15:21:07 galaxy nexus Aug 17 15:21:12 grep write.12 strace.958 | sed 1,5d|cut -d'"' -f2|sed 's@\\@\\\\@g'|(while read i; do bash -c "echo -e -n \"$i\""; done) > firmware.img Aug 17 15:21:13 I do many things at the same time.... Aug 17 15:21:28 ok Aug 17 15:21:37 and then you can look at it with e.g. arm-linux-gnueabi-objdump -b binary -m armv5 --endian=little -axD firmware.img Aug 17 15:21:57 okthanks Aug 17 15:23:38 18dc8: f0169034 ; instruction: 0xf0169034 Aug 17 15:23:41 what's that? Aug 17 15:24:31 probably data Aug 17 15:24:36 ok Aug 17 15:25:55 GNUtoo|laptop: is the strace complete? Aug 17 15:26:06 no idea Aug 17 15:26:12 I'll try to make a better one Aug 17 15:27:15 should I be close to a windows or far? Aug 17 15:27:54 GNUtoo|laptop: I guess it's just sending the firmware Aug 17 15:28:04 ok Aug 17 15:28:11 GNUtoo|laptop: if you want to reverse engineer the protocol then you might want to be outside Aug 17 15:34:21 ok Aug 17 15:34:30 but I guess once it's loaded it would be SIRF Aug 17 15:38:53 http://pastie.org/private/8dkjcg37knuww55rjva Aug 17 15:39:37 GNUtoo|laptop: you might want to include timestamps in your strace too Aug 17 15:39:47 the strange thing is that it takes long before Stopping GPS! Aug 17 15:39:50 and then it doesn't display the icon Aug 17 15:39:51 ok Aug 17 15:40:07 GNUtoo|laptop: since if you get some actual gps data then it is important to have the timing Aug 17 15:40:24 I don't right now Aug 17 15:40:24 GNUtoo|laptop: so -ttt Aug 17 15:40:33 I'm only interested in the load stuff right now Aug 17 15:40:42 because I suspect that it's plain SIRF Aug 17 15:40:55 yep it's 32-bit little-endian arm code Aug 17 15:41:57 ok Aug 17 15:42:40 sGpsInterface->init: -1 doesn't seem ok Aug 17 15:42:42 right paulk-desktop ? Aug 17 15:43:34 agreed Aug 17 15:43:56 maybe things were changed in ICS Aug 17 15:45:07 indeed and I've only ICS to trace Aug 17 15:45:16 and I've not ICS sources Aug 17 15:45:19 and I can't get them right now nor soon Aug 17 15:45:54 I'm looking Aug 17 15:46:02 ok thanks a lot Aug 17 15:48:24 SHR: 03shr-devel 07buildhistory * r75eb6c22408b 10/packages/om_gta04-oe-linux-gnueabi/ (6 files in 6 dirs): packages: Build 201208171715 of shr 20120817 for machine om-gta04 on opmbuild Aug 17 15:49:05 they added once callback in ICS Aug 17 15:50:57 ok Aug 17 15:51:05 can you make an ICS branch? Aug 17 15:52:54 done: git://gitorious.org/replicant/crespo-gps-utils.git -b ics Aug 17 15:53:14 ok thanks a lot Aug 17 15:57:32 sGpsInterface->init: -1 Aug 17 15:57:34 still Aug 17 15:57:35 hmmm Aug 17 15:58:36 hmm something's wrong Aug 17 15:58:44 it should print the time callback Aug 17 15:58:47 it doesn't Aug 17 15:58:48 any log on logcat? Aug 17 15:59:09 http://pastie.org/private/jjujaz1jps383youmuqedw Aug 17 15:59:26 it doesn't print that: Aug 17 15:59:27 request_utc_time_callback Aug 17 15:59:46 I think it's not supposed to print it Aug 17 16:00:09 ah ok Aug 17 16:00:30 so, anything on logcat -b main? Aug 17 16:00:41 let me look Aug 17 16:01:25 D/gps_sirf( 202): GSD4t Firmware Upload Done Aug 17 16:01:26 D/gps_sirf( 202): ### Samsung Proxima-HSPA, CSR SiRF Software 2.0, B13 E19 2011,10,24 ### Aug 17 16:02:18 it does it at the same time than starting GPS! Aug 17 16:02:49 does this appear with the bcm4751_hal too? Aug 17 16:03:19 yes it does Aug 17 16:03:23 that's the whole point Aug 17 16:03:37 I never used something else than bcm4751_hal Aug 17 16:03:50 strangely I see no GPS request on the top bar either Aug 17 16:03:59 oh ok Aug 17 16:04:29 so "D/gps_sirf( 1395): GSD4t Firmware Upload Done" appears right during "Starting GPS!" Aug 17 16:04:53 the second time the samsung string didn't appear tough Aug 17 16:04:54 I see Aug 17 16:05:28 ah wait Aug 17 16:05:36 you're building with 2.3, right? Aug 17 16:06:40 yes Aug 17 16:06:47 I've no other choice right now Aug 17 16:07:30 then maybe there is a header issue when doing sizeof(GpsCallbacks), Aug 17 16:07:59 try removing the line with #include Aug 17 16:08:04 and instead downlaod CM9's gps.h Aug 17 16:08:08 and include it Aug 17 16:08:50 that one: https://raw.github.com/CyanogenMod/android_hardware_libhardware/ics-release/include/hardware/gps.h Aug 17 16:11:39 ok Aug 17 16:11:53 SHR: 03shr-devel 07buildhistory * r68a32d58e0eb 10/packages/crespo-oe-linux-gnueabi/ (6 files in 6 dirs): packages: Build 201208171757 of shr 20120817 for machine crespo on opmbuild Aug 17 16:19:48 sGpsInterface->init: -1 Aug 17 16:19:50 still..... Aug 17 16:19:52 hmmm Aug 17 16:20:24 GNUtoo|laptop: it fails under strace? Aug 17 16:20:33 yes Aug 17 16:20:46 but there is that in the log in another hand: Aug 17 16:20:48 D/gps_sirf( 1025): GSD4t Firmware Upload Done Aug 17 16:21:03 GNUtoo|laptop: but works without strace? Aug 17 16:21:15 no idea Aug 17 16:21:19 I'll try Aug 17 16:21:56 same: Aug 17 16:21:58 sGpsInterface->init: -1 Aug 17 16:39:10 http://gnutoo.homelinux.org/downloads/people/paulk/strace5-ics2.tar.bz2 Aug 17 16:48:07 paulk-desktop, ^^^ Aug 17 16:48:25 ok I'll look at it Aug 17 16:52:14 paulk-desktop, I must have done a bad manupulation..... Aug 17 16:52:16 now it works better Aug 17 16:52:27 got a nmea callback: $GPGSA,A,1,,,,,,,,,,,,,,,*1E Aug 17 16:52:39 oh great! Aug 17 16:52:56 I must have pushed the wrong update binary I guess Aug 17 16:53:05 for instance I looked at the last binary in the mm log Aug 17 16:53:07 pushed that Aug 17 16:53:15 but it was not necessarly the right one Aug 17 16:53:18 I removed the stop Aug 17 16:53:21 I'll try gpsd Aug 17 17:15:54 http://gnutoo.homelinux.org/downloads/people/paulk/strace6.tar.bz2 Aug 17 17:16:06 I commented the deactivation in that Aug 17 17:16:28 http://pastie.org/private/ltgihutrxywivdilhamc4a Aug 17 17:20:12 So we have MEIF in there Aug 17 17:20:57 SIRF* Aug 17 17:20:58 not meif Aug 17 17:23:07 GNUtoo|laptop, from line 1262 of strace.1031 from http://gnutoo.homelinux.org/downloads/people/paulk/strace5-ics2.tar.bz2 Aug 17 17:24:00 all of what's is before is unknown protocol Aug 17 17:26:11 though it should be easy to figure it out Aug 17 17:42:50 SHR: 03shr-devel 07buildhistory * r1cd90b5718f6 10/packages/tuna-oe-linux-gnueabi/ (20 files in 20 dirs): packages: Build 201208171820 of shr 20120817 for machine tuna on opmbuild Aug 17 17:43:00 SHR: 03shr-devel 07buildhistory * rf24d09b609db 10/images/tuna/eglibc/chroot-image/ (build-id files-in-image.txt image-info.txt): images: Build 201208171820 of shr 20120817 for machine tuna on opmbuild Aug 17 19:28:32 paulk-desktop, yes but.... Aug 17 19:28:43 basically here's what happens Aug 17 19:28:54 it tries SIRF Aug 17 19:29:00 it nearly recognize it Aug 17 19:29:08 but then comes an unknown frame Aug 17 19:29:20 and then it seem unknown stuff Aug 17 20:01:54 paulk-desktop, I don't find strace.1031 Aug 17 20:02:08 ahh strace5 Aug 17 20:02:11 not 6 Aug 17 20:02:32 yes Aug 17 20:03:48 what's the best strategy between: Aug 17 20:03:55 1) decode the init Aug 17 20:04:05 *decode the init first Aug 17 20:04:19 or try to make gpsd work first in a chroot or trough reboot Aug 17 20:04:36 by trough reboot I mean activate in CM and then reboot Aug 17 20:05:59 basically should I do that: Aug 17 20:06:04 1) echo 0,1 to sysfs Aug 17 20:06:30 2) stty -F /dev/ttyO0 the_speed Aug 17 20:06:40 3) cat firmware.img > /dev/ttyO0 Aug 17 20:06:46 4) launch gpsd ? Aug 17 20:07:41 I don't think it'll work Aug 17 20:07:49 because it does a bit more than just writing the firmware Aug 17 20:07:54 there is a small protocol Aug 17 20:08:11 but we could implement it Aug 17 20:11:37 ok Aug 17 20:11:52 how should reverse it? Aug 17 20:12:11 the problem here is that there aren't any logs Aug 17 20:12:19 ah? Aug 17 20:12:23 so we can't really know what stuff means Aug 17 20:12:27 you mean the app log Aug 17 20:12:28 ah ok Aug 17 20:12:41 I mean that the lib doesn't write in human language what it's doing Aug 17 20:12:48 we can force the logs with the /system/vendor/etc maybe Aug 17 20:12:55 oh, great idea Aug 17 20:13:02 what about crespo's gps? did it have logs? Aug 17 20:13:09 oh yes Aug 17 20:13:11 a lot and a lot Aug 17 20:13:17 on stdout Aug 17 20:13:28 so basically no one would ever see them Aug 17 20:13:35 as it's running as a daemon Aug 17 20:13:40 ok Aug 17 20:13:49 maybe I should try another image? Aug 17 20:13:49 but once you start it from shell, there is a very verbose output Aug 17 20:13:52 like official image? Aug 17 20:14:13