**** BEGIN LOGGING AT Fri Jul 06 02:59:58 2012 Jul 06 08:27:19 hmm. where are the damn socket permissions set in android Jul 06 08:32:33 mickey_office: ping Jul 06 08:39:52 freesmartphone.org: 03morphis 07cornucopia * r94731e1f328c 10/fsogsmd/src/plugins/dbus_service/plugin.vala: fsogsmd: dbus_service: don't make private variable static Jul 06 08:39:54 freesmartphone.org: 03morphis 07cornucopia * rbd16a3f28aa1 10/fsogsmd/src/plugins/dbus_service/ (device_servicemanager.vala plugin.vala): fsogsmd: dbus_service: construct modem outside of our service manager Jul 06 08:39:55 freesmartphone.org: 03morphis 07cornucopia * r27e9fd373b12 10/fsogsmd/src/ (23 files in 7 dirs): fsogsmd: don't use theModem in any mediator anymore but use protected modem property Jul 06 08:39:56 freesmartphone.org: 03morphis 07cornucopia * r1aa3d988052b 10/fsogsmd/src/plugins/dbus_service/ (Makefile.am device_servicemanager.vala plugin.vala): fsogsmd: dbus_service: move device service manager implementation to own source file Jul 06 08:39:58 freesmartphone.org: 03morphis 07cornucopia * rdf377321398c 10/fsogsmd/src/ (3 files in 2 dirs): fsogsmd: lib: remove access of theModem variable from call class Jul 06 08:39:59 freesmartphone.org: 03morphis 07cornucopia * r2f6e5608930e 10/fsogsmd/src/lib/modem.vala: fsogsmd: lib: completely remove theModem variable Jul 06 08:40:02 freesmartphone.org: 03morphis 07cornucopia * r6b4b3708eaf5 10/fsogsmd/src/lib/ (at/atcommand.vala modem.vala): Jul 06 08:40:02 freesmartphone.org: fsogsmd: lib: configure AT commands after registration with the modem to use Jul 06 08:40:02 freesmartphone.org: I don't really like this solution to access the modem in the AT command implementation. Jul 06 08:40:03 freesmartphone.org: When we're refactoring this the AT commands should be only configured with things they Jul 06 08:40:03 freesmartphone.org: really need like the used charset. Logging should happen in the context of the channel the Jul 06 08:40:03 freesmartphone.org: commands are executed on. Jul 06 08:40:04 freesmartphone.org: 03morphis 07cornucopia * r4ef9971c1c22 10/fsogsmd/src/plugins/dbus_service/device_servicemanager.vala: fsogsmd: dbus_service: don't forget to save reference to our modem Jul 06 08:40:04 freesmartphone.org: 03morphis 07cornucopia * rce16d1a5543c 10/fsogsmd/src/ (5 files in 2 dirs): fsogsmd: adjust for the removal of the global theModem variable Jul 06 08:40:05 freesmartphone.org: 03morphis 07cornucopia * r98779929a6cf 10/fsogsmd/tests/ (testatcommand.vala testsmsstorage.vala): fsogsmd: test: adjust for recent changes to get tests working again Jul 06 08:40:09 freesmartphone.org: 03morphis 07cornucopia * r1ad91967f3e6 10/libfsoframework/fsoframework/subsystem.vala: libfsoframework: support deregistration of service objects Jul 06 08:40:09 freesmartphone.org: 03morphis 07cornucopia * rcb6f25a0aff4 10/fsogsmd/src/plugins/dbus_service/device_servicemanager.vala: fsogsmd: dbus_service: print out modem type when configuration is done Jul 06 08:40:09 freesmartphone.org: 03morphis 07cornucopia * r4e2cf7dd81bb 10/fsogsmd/ (54 files in 20 dirs): fsogsmd: don't use global modem reference anymore but a local one Jul 06 08:40:10 freesmartphone.org: 03morphis 07cornucopia * rf03fc96bd944 10/fsogsmd/src/lib/modem.vala: fsogsmd: lib: invert status update sending for NetworkTimeReport class Jul 06 08:40:11 freesmartphone.org: 03morphis 07cornucopia * r22c928ffd81a 10/fsogsmd/src/lib/ (call.vala mediator.vala modem.vala): Jul 06 08:40:35 freesmartphone.org: fsogsmd: lib: assign modem to use on mediator creation Jul 06 08:40:35 freesmartphone.org: In the future every mediator should use the protected member of the abstract base class to Jul 06 08:40:35 freesmartphone.org: acces the modem instead of the global variable theModem. Jul 06 08:55:07 freesmartphone.org: 03morphis 07morphis/multi-device * r3af9ad1b505d 10cornucopia/fsogsmd/src/lib/ (Makefile.am modemmanager.vala): fsogsmd: lib: add abstract modem manager class as base for a real one Jul 06 08:55:08 freesmartphone.org: 03morphis 07morphis/multi-device * r15cca27ad257 10cornucopia/fsogsmd/src/ (4 files in 2 dirs): fsogsmd: dbus_service: rework to add modem management support Jul 06 09:06:55 freesmartphone.org: 03morphis 07morphis/hfp * rfe4b3799c64e 10cornucopia/fsogsmd/ (8 files in 6 dirs): fsogsmd: modem_hfp_hf: add initial support for a HFP HF modem type Jul 06 09:06:56 freesmartphone.org: 03morphis 07morphis/hfp * rd886303bb492 10cornucopia/fsogsmd/src/lib/ (Makefile.am bluetooth.vala): Jul 06 09:06:56 freesmartphone.org: fsogsmd: lib: add a bluetooth manager class for bluez Jul 06 09:06:56 freesmartphone.org: With this management class you can register different bluetooth profile handlers which will Jul 06 09:06:56 freesmartphone.org: be automatically associated with bluetooth devices handled by the bluez daemon if they Jul 06 09:06:57 freesmartphone.org: supports the registered profile. Jul 06 09:06:57 freesmartphone.org: 03morphis 07morphis/hfp * r2a1febc444d1 10cornucopia/fsogsmd/src/lib/bluetooth.vala: fsogsmd: lib: add interface definition for bluez handsfree related APIs Jul 06 09:06:58 freesmartphone.org: 03morphis 07morphis/hfp * r4ea6bb497707 10cornucopia/fsogsmd/src/ (7 files in 2 dirs): fsogsmd: dbus_service: expose service registration method of our manager to the public Jul 06 09:06:58 freesmartphone.org: 03morphis 07morphis/hfp * rc4d6837ada58 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/ (Makefile.am agent.vala plugin.vala profile.vala): fsogsmd: modem_hfp_hf: register bluez handsfree agent when device with HFP HF profile is found Jul 06 09:06:59 freesmartphone.org: 03morphis 07morphis/hfp * r399940058ac5 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/agent.vala: fsogsmd: modem_hfp_hf: register handfree dbus agent with bluez Jul 06 09:07:01 freesmartphone.org: 03morphis 07morphis/hfp * r02804ea98d44 10cornucopia/fsogsmd/ (4 files in 3 dirs): fsogsmd: modem_hfp_hf: uses dbus file descriptor passing feature for org.blue.HandsfreeAgent interface Jul 06 09:07:01 freesmartphone.org: 03morphis 07morphis/hfp * r33722802e363 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/ (Makefile.am agent.vala modem.vala plugin.vala profile.vala): fsogsmd: modem_hfp_hf: rework to support new multi-device style core Jul 06 09:07:01 freesmartphone.org: 03morphis 07morphis/hfp * r7c6d1ee6da47 10cornucopia/fsogsmd/src/plugins/dbus_service/device_servicemanager.vala: fsogsmd: dbus_service: accessing method via base identifier causes compile-time errors Jul 06 09:07:01 freesmartphone.org: 03morphis 07morphis/hfp * r0f3ba182fe34 10cornucopia/fsogsmd/src/ (2 files in 2 dirs): fsogsmd: lib: let the childs access the service path of our provider Jul 06 09:07:06 freesmartphone.org: 03morphis 07morphis/hfp * rd377b3ef0646 10cornucopia/fsogsmd/src/ (2 files in 2 dirs): fsogsmd: dbus_service: make unregisterService method available for the public Jul 06 09:07:06 freesmartphone.org: 03morphis 07morphis/hfp * r3ac0eab1f546 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/plugin.vala: fsogsmd: modem_hfp_hf: don't setup a modem for device a second time Jul 06 09:07:06 freesmartphone.org: 03morphis 07morphis/hfp * rbbf6a1d899e1 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/plugin.vala: fsogsmd: modem_hfp_hf: remove device from list of controlled devices on removal Jul 06 09:07:06 freesmartphone.org: 03morphis 07morphis/hfp * r3dda19c1bdf3 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/modem.vala: fsogsmd: modem_hfp_hf: rework bluez agent registration/deregistration Jul 06 09:07:07 freesmartphone.org: 03morphis 07morphis/hfp * ra865fa3a52fa 10cornucopia/fsogsmd/src/lib/bluetooth.vala: fsogsmd: lib: bluetooth: handle service connect and disconnect Jul 06 09:07:11 freesmartphone.org: 03morphis 07morphis/hfp * rcdbdec33bd5e 10cornucopia/fsogsmd/src/plugins/dbus_service/plugin.vala: fsogsmd: dbus_service: be verbose when registering/unregistering a modem Jul 06 09:07:11 freesmartphone.org: 03morphis 07morphis/hfp * r18dfcc950478 10cornucopia/libfsotransport/fsotransport/ (Makefile.am unix.vala): libfsotransport: a new unix filedescriptor based transport type Jul 06 09:07:11 freesmartphone.org: 03morphis 07morphis/hfp * r2ce89cbc8fe2 10cornucopia/fsogsmd/src/lib/modem.vala: fsogsmd: lib: modem: add async modifier for powerOn/powerOff methods Jul 06 09:07:11 freesmartphone.org: 03morphis 07morphis/hfp * r0d3a2ce5fdc3 10cornucopia/libfsotransport/fsotransport/commandqueue.vala: libfsotransport: don't assert when transport is already open but warn the user Jul 06 09:07:12 freesmartphone.org: 03morphis 07morphis/hfp * recdf35d25b3f 10cornucopia/libfsotransport/fsotransport/unix.vala: libfsotransport: unix: we have a different open behavior for this type of transport Jul 06 09:07:17 freesmartphone.org: 03morphis 07morphis/hfp * r434a550ee9bf 10cornucopia/fsogsmd/src/lib/at/atcommandqueue.vala: fsogsmd: lib: accept an AtCommandQueue instead of a AtChannel for a AtCommandSequence Jul 06 09:07:17 freesmartphone.org: 03morphis 07morphis/hfp * r52b2a7c9c564 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/ (Makefile.am channel.vala modem.vala): fsogsmd: modem_hfp_hf: implement transport and channel setup procedure Jul 06 09:07:17 freesmartphone.org: 03morphis 07morphis/hfp * rf1bde742e386 10cornucopia/libfsotransport/fsotransport/basetransport.vala: libfsotransport: basetransport: make class members available for child classes Jul 06 09:07:17 freesmartphone.org: 03morphis 07morphis/hfp * rf179f0996ad1 10cornucopia/libfsotransport/fsotransport/unix.vala: libfsotransport: unix: some minor improvements for correct file descriptor handling Jul 06 09:07:18 freesmartphone.org: 03morphis 07morphis/hfp * r33880e99ca17 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/modem.vala: fsogsmd: modem_hfp_hf: keep a reference to the socket otherwise our file descriptor will be closed Jul 06 09:07:36 freesmartphone.org: 03morphis 07morphis/hfp * r32c54848d92a 10cornucopia/fsogsmd/ (3 files in 3 dirs): fsogsmd: lib: at: implement AT+CIND command Jul 06 09:07:36 freesmartphone.org: 03morphis 07morphis/hfp * r3c2c233cde79 10cornucopia/fsogsmd/ (src/lib/at/atcommands.vala tests/testatcommand.vala): fsogsmd: lib: at: implement AT+CMER command Jul 06 09:07:36 freesmartphone.org: 03morphis 07morphis/hfp * rc80c0c0aaa09 10cornucopia/fsogsmd/src/lib/at/atcommands.vala: fsogsmd: lib: at: implement AT+CMEE command Jul 06 09:07:36 freesmartphone.org: 03morphis 07morphis/hfp * r03c989398e52 10cornucopia/fsogsmd/ (src/lib/at/atcommands.vala tests/testatcommand.vala): fsogsmd: lib: at: implement test functionality for AT+CHLD command Jul 06 09:07:37 freesmartphone.org: 03morphis 07morphis/hfp * r4d9c5a8ae17f 10cornucopia/fsogsmd/src/ (7 files in 2 dirs): fsogsmd: modem_hfp_hf: implement service level connection establishment procedure Jul 06 09:08:36 freesmartphone.org: 03morphis 07morphis/hfp * r96a683628330 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/ (Makefile.am modem.vala unsolicited.vala): fsogsmd: modem_hfp_hf: implement handler for unsolicited response messages Jul 06 09:08:36 freesmartphone.org: 03morphis 07morphis/hfp * r2958cb58a504 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/ (channel.vala modem.vala slc.vala): fsogsmd: modem_hfp_hf: move SLC handling to modem class and keep our channel unresponsible Jul 06 09:08:36 freesmartphone.org: 03morphis 07morphis/hfp * re8ed160181cb 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/modem.vala: fsogsmd: modem_hfp_hf: don't use any SMS, phonebook or call handler yet Jul 06 09:08:36 freesmartphone.org: 03morphis 07morphis/hfp * r5d5010c07ccf 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/modem.vala: fsogsmd: modem_hfp_hf: register a unsolicited response since we're not a "real" AT channel Jul 06 09:09:36 freesmartphone.org: 03morphis 07morphis/hfp * r14dd776f940f 10cornucopia/fsogsmd/ (8 files in 4 dirs): fsogsmd: modem_hfp_hf: implement own handler class for the indicators Jul 06 09:09:36 freesmartphone.org: 03morphis 07morphis/hfp * r3b450a73bc8d 10cornucopia/fsogsmd/src/lib/bluetooth.vala: fsogsmd: lib: bluetooth: add org.bluez error domain Jul 06 09:09:36 freesmartphone.org: 03morphis 07morphis/hfp * rbb08df2d65f6 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/modem.vala: fsogsmd: modem_hfp_hf: throw errors from the org.bluez error domain and nothing else Jul 06 09:09:36 freesmartphone.org: 03morphis 07morphis/hfp * r6ec7240fa96e 10cornucopia/fsogsmd/src/ (4 files in 2 dirs): fsogsmd: modem_hfp_hf: move indicator type definitions to core library as they are pretty std. 27.007 Jul 06 09:09:37 freesmartphone.org: 03morphis 07morphis/hfp * r469eeeeb285a 10cornucopia/fsogsmd/src/plugins/modem_hfp_hf/ (indicators.vala modem.vala unsolicited.vala): fsogsmd: modem_hfp_hf: update indicator status handling Jul 06 09:40:36 paulk-desktop: hi Jul 06 09:40:41 hi Jul 06 09:40:55 paulk-desktop: thank you, data is working fine Jul 06 09:41:07 ah, I'm glad to hear it! Jul 06 09:41:47 so, only gps, ussd and socket permissions are left. I'm a little bit busy with bureaucracy at work today, but I plan to make some ipc dumps with the blob and disassemble it Jul 06 09:42:03 ok, nice! Jul 06 09:42:07 any ideas how to fix socket group/perms? Jul 06 09:42:15 I'd love to have USSD logs from the non-free RIL btw Jul 06 09:42:35 Alex[sp3dev], get it back to the groub /dev/socket was in GB Jul 06 09:42:41 group* Jul 06 09:42:48 the clean way to do this is via ueventd.rc Jul 06 09:43:07 paulk-desktop: does ueventd.rc change permissions each time the file is created? Jul 06 09:44:02 I think changing /dev/socket/ perms just once should work Jul 06 09:44:23 paulk-desktop: I'm afraid we need to fix /dev/socket/samsung-ril-socket each time ril is restarted Jul 06 09:44:32 ouch Jul 06 09:45:26 you can try, maybe it'll fix its perms every time it is created Jul 06 09:45:42 ueventd sounds like it deals with events daemonized, so that could work Jul 06 09:46:34 gotta make blob dumps now Jul 06 09:46:45 probably need to reflash my custom cm9 build first Jul 06 09:48:01 I advise you to rebuild a kernel with -eng and ro.secure=0 Jul 06 09:48:12 you'll get all the logs from the RIL then Jul 06 10:57:54 paulk-desktop: ok, I've nearly completed my bureaucratic quest. So now I will make some proprietary ril tracing Jul 06 11:13:50 Alex[sp3dev], ok, good! Jul 06 11:19:45 paulk-desktop: it looks to me RSSI is broken Jul 06 11:20:00 proprietary ril shows 4 bars and our 2 Jul 06 11:20:14 I had that impression too :) Jul 06 11:20:18 I'll make two ussd calls and enable gps and make the dumps Jul 06 11:20:32 they must have changed the return structure in ICS Jul 06 11:27:24 paulk-desktop: try downloading http://rghost.ru/download/39068531/720d2e1971c820d8f767ed614174303aa47301d3/dump_radio_ussd_and_gps Jul 06 11:27:39 it's a very verbose log Jul 06 11:27:50 nice Jul 06 11:28:00 for reference, ussd numbers were #102# and *102# Jul 06 11:28:13 first one is latin, second one is cyrillic Jul 06 11:28:18 wow, how do you get it to be that verbose? Jul 06 11:28:57 paulk-desktop: don't remember. either it's insecure kernel or I added some samsung config file to enable debuggins Jul 06 11:29:34 ok Jul 06 11:29:53 the logs I get look different, but I only have the ipc data printed, not the whole structure + data Jul 06 11:30:21 paulk-desktop: well, it's galaxy nexus. a 'developer' and 'google approved' sort of thingie. maybe that changes something Jul 06 11:30:36 ah, I thought it was galaxy s2 Jul 06 11:30:39 we're good then Jul 06 11:30:43 ipc should be the same Jul 06 11:30:59 nexus s ril was also more verbose than any other samsung-branded phone Jul 06 11:31:13 maybe google asked that or something… Jul 06 11:32:23 D/RILJ ( 400): [0125]> SEND_USSD #102# Jul 06 11:32:30 nice Jul 06 11:33:01 so data is "01 0F 05 AA 18 4C 36 02 00 00 00 00 00 00 00 00" Jul 06 11:33:22 it's obviously not ASCII Jul 06 11:33:35 should I make an ussd call with our ril? Jul 06 11:33:49 yes please Jul 06 11:34:00 I think it'll send the same data Jul 06 11:35:37 http://pastebin.com/Q6fr7rnj Jul 06 11:35:54 a bit different Jul 06 11:36:49 original: "01 0F 05 A3 18 4C 36 02" Jul 06 11:36:59 a3 18 4c is good Jul 06 11:37:20 3rd byte is different (0x4), last one differs and our lacks one byte. right? Jul 06 11:37:31 3rd is length Jul 06 11:38:00 obviously our encoding is correct, just some flag bits are wrong Jul 06 11:38:27 ok Jul 06 11:38:31 we can assume the last byte is always 2 and add the 0x30 mask.. but that *is* going to crash one day Jul 06 11:38:44 I'll take more logs from other phones Jul 06 11:39:01 well, we could consult the disasm as well Jul 06 11:39:31 ok but I'm not good at it Jul 06 11:40:23 paulk-desktop: could you push your latest commits (including makefile hack) and morphis libsamsung-ipc to replicant main repo so that we have it in one place? Jul 06 11:40:57 paulk-desktop: I also have made some copy-pasta magic Jul 06 11:40:59 https://gitorious.org/~ast/replicant/asts-hardware_ril_samsung-ril/commit/9f3c5150ab0db61899bf5a8839b854c436e707bf Jul 06 11:41:14 well replicant libsamsung-ipc should be way outdated: I stopped pushing there months ago Jul 06 11:42:05 and it's not used in replicant tree Jul 06 11:42:20 do you think we can or should replicate the gps? afaiu, there's a gps firmware inside the .so file which gets loaded to it on start Jul 06 11:43:23 I mean, as a separate projects. I'll implement gps callbacks for ril anyway because open gps is not as security critical as an open ril and we can control power in kernel Jul 06 11:43:45 ok Jul 06 11:43:52 it does load a firmware on galaxy s, nexus s, etc (that's bcm4751 chip, that is in Galaxy 7 tab too)… Jul 06 11:43:58 so maybe we should do a fwcutter Jul 06 11:44:13 as replicant won't distribute the firmware Jul 06 11:44:26 but I'm interested in having a free CPU-side GPS implementation anyway Jul 06 11:45:17 honestly, I'm more interested in getting gnu/linux running and I don't care about openness as long as it works.. because I'm not going to keep sensitive data on a phone anyway Jul 06 11:45:40 ok, I understand Jul 06 11:46:33 like, I would like kernel side to be open, and, ideally, upstreamed. but userland is less problematic. unfortunately unless we make all hardware it is not possible to put pressure on vendors Jul 06 11:47:00 anyway, gps needs to be hacked. this stupid MEIF has been bothering me for long Jul 06 11:47:32 btw, do we have any docs for that? it should be possible to obtain the specs from nokia but I'm not stepping on the NDA land Jul 06 11:48:12 IIRC it was possible to ask NOKIA under NDA Jul 06 11:48:19 but it was somehow bought by microsoft Jul 06 11:48:25 and now they don't allow anyone to access the docs Jul 06 11:48:42 I have MEIF logs and traces Jul 06 11:48:48 from nexus s Jul 06 11:49:24 btw. is microsoft going to make money with WP7/WP8? why don't they port MS Office to linux? Or is their religion to make everything proprietary and broken even if they lose money on it? Jul 06 11:50:26 haha. At least they're making money with Android… Jul 06 12:00:45 paulk-desktop: ok, I've made a successful USSD call but you won't like what I've done Jul 06 12:02:09 paulk-desktop: basically I think you've made a mistake in arithmetic but I've not checked the formulae but hacked it Jul 06 12:03:02 paulk-desktop: MAGIC https://gitorious.org/~ast/replicant/asts-hardware_ril_samsung-ril/commit/92b6f145f3379bc2c5262e5f977856d06bbfa121 Jul 06 12:04:27 ok Jul 06 12:04:31 I'll look at it Jul 06 12:04:38 but what I did is standard GSM7 Jul 06 12:04:58 are the formulae given somewhere? if not, we need to recheck Jul 06 12:18:27 about n900 docs, is there something usefull inside? http://maemo.jacekowski.org/docs/ Jul 06 12:19:23 paulk-desktop: ^ Jul 06 12:21:13 mhh I don't think there'll be anything about MEIF Jul 06 12:21:19 n900 GPS isn't MEIF… Jul 06 12:21:49 ok Jul 06 12:28:58 paulk-desktop: I'm sure GSM7 <-> ASCII conversion is wrong. Or non-latin encoding is broken. I'll try to analyze and fix that later Jul 06 12:29:20 most ASCII requests, including interactive ones, work though Jul 06 12:33:09 ok Jul 06 12:41:32 paulk-desktop: what happens when you define a socket in init.rc? does it get opened by zygote and the daemon gets it on a socket() call? Jul 06 12:42:11 I don't quite understand how proprietary blobs set permissions Jul 06 12:43:00 I don't know how it works in init.rc. Though if there is something about sockets there, it's probably worth trying to make use of it Jul 06 13:05:07 that's so stupid. I'm implementing dummy {return 0} functions for ril client to make GPS driver happy. hope it still works without AGPS Jul 06 15:46:35 hmmm I've a theory(without any try) for gta04 Jul 06 15:47:01 alsaloop.conf uses plughw:0 Jul 06 15:47:04 for playback Jul 06 15:47:25 so in my theory it rings Jul 06 15:47:38 that occupy the sound card Jul 06 15:47:45 then alsaoop try to open it Jul 06 15:47:47 it fails Jul 06 15:48:01 what does the ringtone uses? Jul 06 15:48:05 -Dmedia? Jul 06 15:48:29 maybe I should look in fsodeviced Jul 06 15:48:49 paulk-desktop: I've compared the strace of proprietary ril and ours and haven't found anything interesting about sockets. I think I'll have some supper and then try to play with init.rc Jul 06 15:49:34 ok Jul 06 15:50:19 I think it works like this. We define a socket in init.rc. Zygote opens a domain socket and keeps it until the daemon requests it. Besides, in init.rc, we can set the groups of users (multiple) who can use the socket. so it should be easy to add audio and gps Jul 06 15:51:43 btw cm9 is full of shit. why the hell are proprietary rommanager and terminal required for a full build (you can touch empty files of course) Jul 06 15:52:29 that was for GB too IIRC Jul 06 15:52:44 I remember I had to remove that rommanager dependency on replicant Jul 06 15:53:30 what's the point in making closed-source zero-pay software? Jul 06 15:54:58 Alex[sp3dev]: less care factor Jul 06 17:19:07 mrmoku: can you please delete the trac users iwdy23 and absscam Jul 06 17:22:45 mrmoku: what is the vala equivalent to int fd = open(file, O_RDWR | O_NOCTTY | O_NONBLOCK);? Jul 06 17:27:41 mrmoku: is the equivalent of i"nt fd " a vala inputstream and outputstream ? Jul 06 17:32:38 nschle85: if you need noctty/nonblock, why not use posix functions (inb4 I know no vala but I know it has wrappers for everything) Jul 06 17:33:30 Alex[sp3dev]: I am porting a c program to vala, to learn vala. Jul 06 17:33:59 Alex[sp3dev]: so i have only an educational problem :-) Jul 06 17:35:22 Alex[sp3dev]: the c program does not work, because its stack gets corrupted during regexec but so i decided to port to vala because this stuff may be integrated into fso Jul 06 17:36:49 and fso=vala Jul 06 17:58:10 nschle85: maybe, you should use async gio? Jul 06 17:59:52 Alex[sp3dev]: i already asked in the vala channel on different server, the closest implementation is ti use Posix.open , but you are right if someone wants to use it in java style its gio Jul 06 18:00:26 the question is what gio uses inside - select, epoll or whatever Jul 06 18:00:36 actually shouldn't matter since it's kernel's job Jul 06 18:00:56 Alex[sp3dev]: but using files in posix and java/vala is very different, in sight of API Jul 06 18:01:09 s/sight/view Jul 06 18:01:35 nschle85: I would say you should use what most people are using if you don't want to support your bikeshed yourself Jul 06 18:02:52 I mean that I've not found a ready to use posix vapi Jul 06 18:03:36 Alex[sp3dev]: i ll use the posix vapi, but i did not know that it exist... Jul 06 18:04:09 nschle85: I think it does not exist except for some proposed patches on the mailing list Jul 06 18:04:18 async seems nice. like C# 5.0 Jul 06 18:04:34 and yes, I like .NET and hate java Jul 06 18:04:45 Alex[sp3dev]: [19:52:14] nschle85, well Posix.open will let you do a direct port of the C. other options would be GLib.File (in gio-2.0), GLib.IOChannel (in glib-2.0), libftdi, etc. Jul 06 18:04:57 nice Jul 06 18:05:57 ok, thank you for the dicussion... Jul 06 18:06:53 thank you for reminding me of vala. now I'm trying to write a program utilizing async according to the tutorial Jul 06 18:08:18 Alex[sp3dev]: i am not amused about vala but it seems the only way to rapid development in C Jul 06 18:09:07 Alex[sp3dev]: and if you want to extend fso you have to learn it Jul 06 18:09:15 nschle85: from a brief glance it seems like a good language. still allowing pointer magic and mutable data (unlike haskell) while having type inference, strict typing and an overall good oop model (better than c++) Jul 06 18:10:18 Alex[sp3dev]: hmm. may be someone is missing overloading methods and constructors Jul 06 18:10:54 nschle85: overloading methods is generally a bad idea except for overloading the subscript operator for hash maps Jul 06 18:11:26 Alex[sp3dev]: ???????? why ? Jul 06 18:11:28 I like the fact that vala has lambda expressions like C#. I like that it's like C# Jul 06 18:11:49 for example, var a = (a, b) => { a + b; } //quite nice Jul 06 18:13:06 Alex[sp3dev]: i was not woundering about lamdas it was about overloading Jul 06 18:13:09 nschle85: because overloaded methods may cause confusion. especially when someone overloads operators. it's an argueable topic, though. lambda expressions and var/auto keyword are more important - the lack of them makes java one huge pain. Jul 06 18:14:35 Alex[sp3dev]: in java you cannot overload operators Jul 06 18:14:53 nschle85: in c++ you can and that leads to a lot of trouble when used carelessly Jul 06 18:15:15 also, C++ has a lot of nontrivial cases like RVO Jul 06 18:15:26 Alex[sp3dev]: i know but nobody will forbid it :-( Jul 06 18:16:21 but overloading may reduce redundant information (method name and parameter list) Jul 06 18:17:53 yes, it is generally a good feature. maybe it was not included because vala authors wanted to translate to pure C without using a lot of vtbl? Jul 06 18:19:31 Alex[sp3dev]: yes this may be the answer Jul 06 18:21:48 Alex[sp3dev]: next difference is: java has a clear api for most things, as i learned today vala offers a lot off classes , so vala has no own api vala collects existing apis Jul 06 18:23:25 that's the idea behind all G* software today - to write core code in C, then generate XML description and use gobject-introspection to hook any language without manually implementing bindings. vala is a bit different because most of its libs call C directly Jul 06 18:24:49 Alex[sp3dev]: in java you are somehow guided thru the api but in vala the same jungle like in c/c++ still exists Jul 06 18:26:09 nschle85: java has a consistent API but terrible syntax and anything involving opengl/DSP means writing tons of jni stubs (as opposed to importing native code from high level language. jna is too slow) and copying memory back and forth. java is not the best choice if you need to interact with C code Jul 06 18:26:15 Alex[sp3dev]: so in my point of view why not integrate vala language elements in c standard ? Jul 06 18:27:09 nschle85: let alone C. it is already getting complex. but advocating vala among non-geeks might be a good idea Jul 06 18:28:22 actually there were a lot of projects like vala aimed to improve stability by introducing strict typing and pointer checks. vala's strong point is community support Jul 06 18:28:49 Alex[sp3dev]: i am not talking about using java but vala is always compared to java... and in my oppinion vala is far away from java so they should not be compared Jul 06 18:29:51 nschle85: in my opinion java as a platform is good because of tight integration, introspection (reflection) and things like maven. but the language is so terrble that I can't stand it. jruby or scala are more fun Jul 06 18:30:41 Alex[sp3dev]: why is the language so terribble ? Jul 06 18:31:05 Alex[sp3dev]: i think c++ language is terrible Jul 06 18:31:37 nschle85: c++ is beyond evil although if you don't have glib and posix, it's the only way in many cases Jul 06 18:32:39 Alex[sp3dev]: but what does you disturb in java language ? Jul 06 18:33:05 nschle85: as for java: no operator overloading, no lambdas, no type inference, generics type erasure (wtf? can't make a new T [ ]), poor integration with native code (jna is slow while C#/.NET offers P/invoke where you can even set structure alignment and don't need to write code in C to utilize VM classes) Jul 06 18:34:31 ok, I like vala async capabilities Jul 06 18:34:39 really simple compared to AIO in C Jul 06 18:35:09 and doesn't cause cpu load so it means it's implemented properly Jul 06 18:40:34 Alex[sp3dev]: operator overloading i would never see in java !!! lamdas will come in Java8 this year, type inferiance is in java 7 : List names = new ArrayList<>(); , for native access code you should read http://en.wikipedia.org/wiki/Java_Native_Access Jul 06 18:41:39 List names = new ArrayList<>(); //cool code, bro. I want this: var names = new ArrayList(); Jul 06 18:42:00 Alex[sp3dev]: but why is java language so bad ? Jul 06 18:42:26 nschle85: because it is backwards compatible Jul 06 18:43:04 Alex[sp3dev]: no its not Jul 06 18:43:47 for example, in .NET, they chose to ship old .net 1.1 and new 2.0+ instead of keeping everything binary compatible. so they throwed away some old code and implemented many things from scratch. that's why there are no type-erased generics, for example Jul 06 18:44:31 either way, I just don't like java, web and enterprise things. I find it hard to troll a java expert Jul 06 18:45:42 Alex[sp3dev]: you told me one argument: you dont like it :-) Jul 06 18:45:54 that's a very strong argument for me Jul 06 20:05:59 GNUtoo, do you have any objections agains fsoaudiod: 61b2068fb6fe073ebee8f0a3ab8ea4d69509bb9f Jul 06 20:52:22 GNUtoo, hi. do you have any objections agains fsoaudiod: 61b2068fb6fe073ebee8f0a3ab8ea4d69509bb9f Jul 06 20:52:30 GNUtoo-hplaptop, ^^^ Jul 06 20:54:37 slyon, do you have an url? Jul 06 20:54:51 GNUtoo-hplaptop, sure: http://git.freesmartphone.org/?p=cornucopia.git;a=commit;h=61b2068fb6fe073ebee8f0a3ab8ea4d69509bb9f Jul 06 20:54:52 slyon, I've a fix for the ringing+silence btw Jul 06 20:55:00 slyon: using Posix; leads to the namespace name Posix cannot be foud, (vala) do you have any idea ? i am using ubuntu vala 0.14 Jul 06 20:55:23 GNUtoo-hplaptop, wow. cool! Jul 06 20:55:36 slyon, it's not tested yet Jul 06 20:55:47 but I'll paste here Jul 06 20:56:16 let me boot it first Jul 06 20:56:21 because the fix is on the phone Jul 06 20:56:30 nschle85, are u using --pkg posix? Jul 06 20:56:44 i am using monodevelop Jul 06 20:57:15 i've no idea about monodevelop. but this here looks pretty much like your problem, right? https://mail.gnome.org/archives/vala-list/2009-May/msg00001.html Jul 06 20:57:36 can you try to compile manually and adding --pkg posix? Jul 06 20:57:47 or search if you can add some compiler flags in monodevelop Jul 06 21:00:38 slyon, try that: http://pastie.org/private/pseoz47myasgje0a25lriw Jul 06 21:00:50 in /etc/freesmartphone/conf/GTA04/alsaloop.conf Jul 06 21:01:02 basically it used plughw:0 Jul 06 21:01:13 wen you ring that uses default Jul 06 21:01:19 slyon: now it gives error: main.vala:40.3-40.7: error: The symbol `POSIX' could not be found POSIX.Termios term_attr; Jul 06 21:01:38 but plughw:0 requires the total control of the sound card Jul 06 21:01:42 no multiplexing Jul 06 21:01:48 nschle85, do you have the posix vapi installed? Jul 06 21:01:50 default uses multiplexing Jul 06 21:03:07 slyon: i guess its part of valac package Jul 06 21:04:54 nschle85, ahh. i guess you have to use Posix instead of POSIX Jul 06 21:05:41 slyon: schle85@dual-wohnzimmer:/usr/share/vala-0.14/vapi$ ls -al | grep posix -rw-r--r-- 1 root root 86747 Feb 7 16:02 posix.vapi Jul 06 21:05:43 slyon, can you test it? Jul 06 21:05:49 at that hour I cannot test Jul 06 21:06:03 slyon: but changing case did not help Jul 06 21:06:41 slyon: http://pastebin.com/k4GsB3Re this i found where i can copy some posix stuff Jul 06 21:07:37 GNUtoo-hplaptop, yes. I will test. just have to look if i've ringtones installed at all Jul 06 21:07:57 GNUtoo-hplaptop, do you know where the ringtones are in the filesystem? Jul 06 21:08:30 no but I know they are played by fsoaudiod Jul 06 21:08:35 and that shr-settings have them Jul 06 21:09:05 s/have/access/ Jul 06 21:09:06 GNUtoo-hplaptop meant: and that shr-settings access them Jul 06 21:09:18 ah right Jul 06 21:09:19 sorry Jul 06 21:09:24 I remember now Jul 06 21:09:29 /usr/share/sound/ Jul 06 21:09:34 female_ringtone.wav Jul 06 21:09:52 and female_message.wav Jul 06 21:10:10 it's strange that I didn't remember it Jul 06 21:10:13 I use it all the time Jul 06 21:11:38 GNUtoo-hplaptop: you should make some holidays :-) Jul 06 21:11:48 GNUtoo-hplaptop, well. i had to scp them :) Jul 06 21:12:02 GNUtoo-hplaptop, what's the expected result? Jul 06 21:12:19 i hear the ringtone. once the ringtone is finished i can talk Jul 06 21:12:24 slyon, yes Jul 06 21:12:34 the issue was: Jul 06 21:12:40 you ear the ringtone Jul 06 21:12:42 but as long as the ringtone is still playing i can neither hear nor speak Jul 06 21:12:57 then you accept the call Jul 06 21:13:09 then you don't ear the remote Jul 06 21:13:13 but the remote ear you Jul 06 21:13:29 does it work better? Jul 06 21:13:51 strange that you have no ringtones Jul 06 21:14:05 can you describe exactly what you did Jul 06 21:14:09 and the result too Jul 06 21:14:20 GNUtoo-hplaptop, I'm using a self build shr on debian at the moment, thats the reason for no ringtones Jul 06 21:14:33 GNUtoo-hplaptop, 1) call my gta04 Jul 06 21:14:36 2) gta04 rings Jul 06 21:14:46 3) accept the call on gta04 Jul 06 21:15:10 4) start talking Jul 06 21:15:19 5) remote hears my talking Jul 06 21:15:27 6) talk at remote side Jul 06 21:15:37 7) gta04 hears talking from the remote side Jul 06 21:15:44 so i guess everything is as it should be Jul 06 21:15:44 wow Jul 06 21:15:47 yes Jul 06 21:15:49 I'll push Jul 06 21:15:54 yes Jul 06 21:15:55 can I have a tested-by line Jul 06 21:16:04 yes sure Jul 06 21:16:17 good job! Jul 06 21:16:21 what name should I put in tested-by, pm me the line Jul 06 21:16:36 thanks Jul 06 21:17:15 thanks a lot for the test Jul 06 21:17:23 no problem Jul 06 21:19:11 nschle85, what's your problem exactly? which function? Jul 06 21:19:54 slyon iam porting an aswwering for huawei modem from c to vala Jul 06 21:20:15 and for which function call do you have the problem? Jul 06 21:21:15 http://pastebin.com/9rWBsGVJ last line Jul 06 21:22:57 slyon, ok was missing static in methiod declaration Jul 06 21:23:56 nschle85, allright :) Jul 06 21:24:09 slyon but it did not help Jul 06 21:24:15 root@dual-wohnzimmer:/home/nschle85/GIT-HUAWEI-ANSWERING/vala-answering-machine/vala-huawei-answering# valac --pkg=posix main.vala main.vala:40.3-40.7: error: The symbol `POSIX' could not be found POSIX.Termios term_attr; Jul 06 21:24:39 POSIX.Termios term_attr; --> Posix.termios Jul 06 21:25:26 nschle85, the namespace is called Posix not POSIX. and the Termios is called termios (thats the way it is defined in the posix.vapi file) Jul 06 21:29:49 slyon: ok that problem is soved, now i have ambigous stdout (GLib and Posix) which one should i take ? is it used like GLib.stdout.printf()? Jul 06 21:31:31 nschle85, i'd use the default (glib): stdout.printf("Hello, World\n"); Jul 06 21:32:12 the "GLib." is optional, as the GLib namespace is implicit in Vala Jul 06 21:32:37 slyon: can i declare stdout.printf as default ? so it conflicts with Posix namespace Jul 06 21:33:02 slyon, what's your use case of your gta04? Jul 06 21:33:13 personally I need the following features: Jul 06 21:33:18 * phone Jul 06 21:33:23 * headphones for music Jul 06 21:33:33 * ocasional browsing Jul 06 21:33:49 nschle85, i dont know... if it conflicts in the namespace, just add the GLib. to make clear which one you want to use Jul 06 21:34:08 the phone is now fixed Jul 06 21:34:14 rest the 2 other uses cases Jul 06 21:34:29 GNUtoo-hplaptop, i want it to replace my HP Pre 2. My main usecases are: Basic telephony, sms, contacts and calendar Jul 06 21:34:53 slyon, ok so your uses cases are easier than mines Jul 06 21:35:00 nice to have would be browsing and email Jul 06 21:35:13 slyon, do you know that yocto revived the pimlico project Jul 06 21:35:37 so their tasks application, calendar application etc.... should work again Jul 06 21:35:49 GNUtoo-hplaptop, wow i didn't know that Jul 06 21:36:06 I'll look into it. maybe it's useable Jul 06 21:36:17 at work I build the denzil release and I did an image using sato Jul 06 21:36:25 and I think there was calendar etc... Jul 06 21:36:28 from pimlico Jul 06 21:36:33 cool Jul 06 21:36:42 just check it still work with master Jul 06 21:36:47 I loved the tasks app Jul 06 21:37:10 at the moment i'm using debian, though. as OE pissed me of somehow... it just breaks very often and moves on very fast Jul 06 21:37:37 slyon, ok, when did you start using debian? Jul 06 21:38:10 GNUtoo-hplaptop, i worked on it for about 3 days ... started about 2 weeks ago Jul 06 21:38:26 ah ok, so after the stabilizing SHR thing Jul 06 21:38:44 now i have a basic "SHR on Debian" with illume2 FSO and the SHR apps + firefox and wicd for wifi Jul 06 21:38:49 ok Jul 06 21:39:02 personally I will stay with SHR but do a meta-freedom I think Jul 06 21:39:13 so I would remove any possible non-free stuff Jul 06 21:39:27 I'm too much used to oe Jul 06 21:39:41 slyon, can you do screenshots? Jul 06 21:39:48 with gpe-scap Jul 06 21:39:54 yeah... i'll stay with shr as well. just on another base :) Jul 06 21:40:31 I'm interested in debian for the freedom it has Jul 06 21:41:37 slyon is there any logging in vala ? i would like to declare a log line in my code Jul 06 21:41:49 slyon, what was the calendar name of pimlico? Jul 06 21:41:50 dates? Jul 06 21:41:57 yes dates Jul 06 21:42:09 nschle85, hmm.. like logging to a file? Jul 06 21:42:16 I'll try to compile it Jul 06 21:42:38 slyon: logging to stdout is also suffivcient Jul 06 21:43:00 slyon: including source code line aso.. Jul 06 21:45:29 GNUtoo-hplaptop, http://slyon.de/files/ the icons on the homescreen are not yet working somehow. so i'm using the menu for now Jul 06 21:46:08 ok Jul 06 21:46:24 nschle85, hmm.. i guess there are some glib logging mechanisms... but i dont know if they include the line, etc. Jul 06 21:46:49 slyon: ok ill goggle it later :-) Jul 06 21:47:06 nschle85, http://stackoverflow.com/questions/10400842/logging-with-vala Jul 06 21:48:18 ok dates finished....time for screenshots Jul 06 21:53:05 http://scap.linuxtogo.org/index.php?page=1 Jul 06 21:53:29 so it works again.... Jul 06 21:53:30 nice Jul 06 21:53:42 GNUtoo-hplaptop, wow. this looks pretty useable :) Jul 06 21:53:55 has it some sort of synchronizing? Jul 06 21:54:01 maybe trough pisi Jul 06 21:54:06 ok Jul 06 21:54:11 since it uses evolution data server Jul 06 21:54:13 maybe it works Jul 06 21:54:15 maybe not Jul 06 21:54:25 i see. Jul 06 21:54:30 I didn't test that Jul 06 21:54:32 I can test it Jul 06 21:54:55 what should I do exactly for the sync? Jul 06 21:55:07 GNUtoo-hplaptop: dates has had dead upstream for a long time? Jul 06 21:55:19 seem so Jul 06 21:55:24 it was part of openhand Jul 06 21:55:30 openhand was bought by intel Jul 06 21:55:49 yeah and that's why it might very well be removed from debian soon too Jul 06 21:55:58 yes but it's working again Jul 06 21:56:01 GNUtoo-hplaptop, i'm interessted in syncing my gmail calendar... but it doesn't have a high priority. so you don't need to test it now Jul 06 21:56:04 because now intel does yocto Jul 06 21:56:30 slyon, gmail is supported by pisi it seems Jul 06 21:56:55 ## Example for Google Calendar Jul 06 21:57:20 and it support pimlico dates Jul 06 21:57:21 pimlicodates Jul 06 21:57:29 [pimlicodates] Jul 06 21:57:40 cool. Jul 06 21:57:43 it's written in python Jul 06 21:58:06 do you need any of theses features: Jul 06 21:58:09 * slyon apt-get install dates Jul 06 21:58:15 ok Jul 06 21:58:16 do you need any of theses features: Jul 06 21:58:21 * bluetooth Jul 06 21:58:27 * better sound scenarios? Jul 06 21:58:47 because sound scenarios are very partial right now Jul 06 21:58:55 it's the strict minimum Jul 06 21:59:43 bluethooth: no. i dont have any bluethooth peripherals Jul 06 22:00:08 ok Jul 06 22:00:10 better sond scenarios: not now. for now the bare minimum is enough for me. but for the long term this would be nice Jul 06 22:00:18 ok Jul 06 22:00:49 ok so you're pretty much done Jul 06 22:00:54 note that pisi is in python Jul 06 22:01:01 and that it seem still developped Jul 06 22:01:07 not totally sure tough Jul 06 22:01:14 just that the version seem recent enough Jul 06 22:01:15 yes. i've used pisi to sync my opimd contacts already Jul 06 22:01:24 I use it too Jul 06 22:01:29 for phone->sim syncing Jul 06 22:03:41 lindi-, do you have any idea why i get: echo: write error: Operation not permitted when i try to suspend my gta04 (debian) with echo "mem" > /sys/power/state Jul 06 22:05:46 slyon, for me it works Jul 06 22:05:59 but it says the same thing Jul 06 22:06:03 basically it suspends Jul 06 22:06:07 but also says that Jul 06 22:06:30 for me it seems to suspend, but awakes directly after Jul 06 22:06:43 like usbnetworking is disconnected and directly connected again Jul 06 22:07:26 slyon, try shutting the gps down Jul 06 22:07:40 slyon, in SHR we handle the GPS Jul 06 22:07:47 I made a program for that Jul 06 22:07:52 quick and dirty tough Jul 06 22:07:56 for instance no help Jul 06 22:08:01 but it works reliabily Jul 06 22:08:41 GNUtoo-hplaptop, ahhh right Jul 06 22:08:51 the gps. i didn't remember that :) Jul 06 22:09:12 ok, what will you do? Jul 06 22:09:15 integrate it Jul 06 22:09:19 or just shut it down? Jul 06 22:12:15 integrate it Jul 06 22:12:34 ok nice Jul 06 22:12:42 the program that I did is here: Jul 06 22:13:13 http://git.freesmartphone.org/?p=utilities.git;a=tree;f=gta04/gps-handler;h=1458b9b6d4ac0ca46e5b89a78b6f3626a6b92731;hb=HEAD Jul 06 22:13:19 it depends on gpsd Jul 06 22:13:21 i plan to do a shr-config-gta04 deb package, which includes all the stuff like e config, gps quirks, etc. Jul 06 22:13:30 ok Jul 06 22:13:38 GNUtoo-hplaptop, thanks for the link Jul 06 22:13:45 np Jul 06 22:13:50 I'm interested in debian too Jul 06 22:13:55 it's more freedom friendly Jul 06 22:15:08 yes, today i've read, that they are even trying to become a FSF recommended distribution Jul 06 22:15:57 wow Jul 06 22:16:00 nice Jul 06 22:16:27 i'll release my debian work once it becomes ready and useable. i did the first part of it today: http://lists.alioth.debian.org/pipermail/pkg-e-devel/2012-July/001555.html Jul 06 22:16:29 note that I use trisquel, an fsf recommanded distribution Jul 06 22:16:36 i se Jul 06 22:16:38 ok let me read it Jul 06 22:16:50 and I've even done the port of coreboot on my desktop Jul 06 22:16:52 it's just a request to update elementary in debian Jul 06 22:16:56 ok Jul 06 22:16:56 cool Jul 06 22:17:15 because we need elementary-1.0 to use libphone-ui-shr Jul 06 22:18:10 ok Jul 06 22:18:30 then we need fsoaudiod where i've a (very hacky) package now to test. we need shr-e-gadgets (i've a test package for that as well) and we need some configs Jul 06 22:18:43 ok Jul 06 22:19:25 in my opinion it is much easier to maintain some 5-6 deb packages than maintaining the whole shr/OE distro and are about all the toolchain and core changes Jul 06 22:19:53 with debian the most is done by other people and we can concentrate on our own stuff. Jul 06 22:20:41 s/are about// Jul 06 22:21:01 I like debian for freedom reasons and for stability, however it's easier to customize SHR Jul 06 22:21:10 you just do a bbappend Jul 06 22:21:13 and you're done Jul 06 22:21:15 yes that's true Jul 06 22:21:37 however it's plagued with bugs.... Jul 06 22:21:57 i'll see which results my debian experiment will yield Jul 06 22:22:04 ok good Jul 06 22:22:48 GNUtoo-hplaptop, btw next wednesday RMS is giving a talk at my university Jul 06 22:23:00 wow nice Jul 06 22:23:05 I meet RMS many times Jul 06 22:23:14 and we often talk by mail Jul 06 22:23:27 (for the replicant project) Jul 06 22:23:30 i'll show him the gta04 and talk about the powervr reverse engineering efforts Jul 06 22:23:40 cool. it's the first time i'll meet him Jul 06 22:23:41 ok Jul 06 22:26:25 slyon, maybe you could meet him before or after the talk? Jul 06 22:26:49 GNUtoo-hplaptop, yes. that's what i'm planing to do Jul 06 22:27:06 slyon, then write him a mail Jul 06 22:27:14 or ask to johnsu01 Jul 06 22:27:19 hi johnsu01 Jul 06 22:28:23 well i dont think i need to write him a mail. i can just go to the front of the room after his talk and talk to him Jul 06 22:28:58 yes but if he has to leave right after you will lack time Jul 06 22:32:33 hmm i'll see Jul 06 22:32:45 ok Jul 06 22:33:18 i'll leave now. i'm gettin pretty tired. Jul 06 22:33:26 slyon, also replicant is beeing ported to the gta04 Jul 06 22:33:31 by paulk-desktop Jul 06 22:33:32 ok Jul 06 22:33:33 GNUtoo-hplaptop, i know. Jul 06 22:33:36 see you later Jul 06 22:33:40 see you! **** ENDING LOGGING AT Sat Jul 07 02:59:58 2012