**** BEGIN LOGGING AT Thu Jul 29 02:59:57 2010 **** BEGIN LOGGING AT Thu Jul 29 05:10:14 2010 **** BEGIN LOGGING AT Thu Jul 29 05:24:18 2010 **** ENDING LOGGING AT Thu Jul 29 05:37:22 2010 **** BEGIN LOGGING AT Thu Jul 29 05:54:14 2010 Jul 29 08:40:31 i can't build dbus-glib-native... Jul 29 08:43:52 dcordes_: sorry, I was away Jul 29 08:44:08 dcordes_: what's up? Jul 29 09:12:12 It fails to build ffmpeg for me.. Can someone help? Here is log http://wklej.org/id/370198/ Jul 29 09:18:43 soltys: maybe that's another one not building in thumb mode... Jul 29 09:18:58 try to put ARM_INSTRUCTION_SET = "arm" into the recipe Jul 29 09:20:11 ok Jul 29 09:37:18 mrmoku: it didn't help Jul 29 09:50:12 soltys: :/ Jul 29 09:50:42 trying too now Jul 29 09:51:21 sometime ago it builded, now it fails.. Jul 29 10:50:03 mrmoku: pong Jul 29 11:13:54 lindi-: yes, license is interesting... and i were ynable to figure out if i can fork it, name 'notbench' and forget about restrictions. i think yes, so license in fact pretty ok. this is benchmark, and such restrictions understandible. may be you know other good microbenchmark? Jul 29 11:15:27 i am using shr-u from a few days ago. on some boots i don't get /dev/mmcblk0px devices, so i can't mount the sd card Jul 29 11:21:47 i also see lots of [ 3953.790000] Alignment trap: dmesg (292) PC=0x4000a7ec Instr=0xe7851008 Address=0x4012f607 FSR 0x813 in dmesg. are these a problem to worry about? Jul 29 11:21:52 b0ef: are you still using ogsmd? Jul 29 11:25:10 ssam: try find right value for kernel parameter glamo_mci.sd_max_clk= between 2500000 and 16000000 Jul 29 11:25:16 mrmoku|lunch: I don't know - I'm not supposed to? Jul 29 11:25:42 Martix, how do i change it? Jul 29 11:26:19 ssam: u-boot or Qi? Jul 29 11:27:06 Martix, not 100% sure. i think Qi Jul 29 11:27:31 ssam: and for Alignment traps fix change in /etc/init.d/alignment.sh value to 2 Jul 29 11:28:49 ssam: for Qi set glamo_mci.sd_max_clk in /boot/append-GTA02 Jul 29 11:29:55 i dont have a /boot, is it on a different partition? Jul 29 11:36:56 i installed to NAND if that makes a difference Jul 29 11:39:42 ~status Jul 29 11:39:43 Since Thu Jul 29 06:58:13 2010, there have been 0 modifications, 37 questions, 0 dunnos, 0 morons and 35 commands. I have been awake for 4h 41m 29s this session, and currently reference 117187 factoids. I'm using about 16472 kB of memory. With 0 active forks. Process time user/system 370.78/28.73 child 0.02/0.02 Jul 29 11:39:47 ssam: I dont use Qi, so I cant help you much Jul 29 11:58:09 i tried creating a /boot/append-GTA02 on the root filesystem (/dev/mtdblock6 according to fstab), but the glamo_mci.sd_max_clk=2500000 does not show in /proc/cmdline so i dont think it worked Jul 29 11:58:34 ssam: Qi does not support kernel parameters if your root is on MTD afaik Jul 29 12:00:09 so i need to install to SD to be able to configure glamo_mci.sd_max_clk. Jul 29 12:01:06 or should i switch to u-boot? (i think it was recomended to switch to Qi to make resume fast) Jul 29 12:01:46 ssam: this is bllody idea to connect resume speed to u-boot. Jul 29 12:02:34 ssam: you may configure u-boot to feed kernel with same parameters to qi, and you'll find no difference in resume or anything else. Jul 29 12:02:59 ssam: when you say "recommended" you better mention who recommends it Jul 29 12:03:40 lindi-, the openmoko wiki Jul 29 12:03:45 ssam: but anybody can write to that Jul 29 12:04:51 i might be wrong. i think when i got my phone, and installed shr last autumn, there was a recomentation to switch to a different bootloader than the default on the wiki. i think i followed it. Jul 29 12:05:20 i have also played with beagleboard, so i could be confused Jul 29 12:05:21 ssam: yep different people recommend different things Jul 29 12:05:33 b0ef: well... ogsmd is obsolete... you should use fsogsmd instead... and that would explain why the script does not work for you Jul 29 12:05:41 mrmoku|lunch: it's not obsolete, I use it :) Jul 29 12:05:52 lindi-: ok... unsupported by the author then :P Jul 29 12:05:55 :-) Jul 29 12:06:05 mrmoku: we'll just form a new upstream :) Jul 29 12:06:12 ssam: actually this were openmoko staff, so you understand everything right. Jul 29 12:06:25 lindi-: :) Jul 29 12:06:33 lindi-, please don't Jul 29 12:06:38 xrmx: why not? Jul 29 12:07:12 API is incompatible... Jul 29 12:07:25 lindi-, the idea to substitute python with vala is sound Jul 29 12:07:34 which leads to problems like b0ef's Jul 29 12:07:41 xrmx: but it requires a version of vala that is in debian experimental which is a huge problem Jul 29 12:09:12 lindi-, the problem is why it's not in sid :) Jul 29 12:09:22 so would the consensus be to switch to uboot, so i can fix glamo, and have reliable uSD. Jul 29 12:09:38 i guess this is the right thing to follow http://wiki.openmoko.org/wiki/Flashing_the_Neo_FreeRunner#Flashing_the_boot_loader_to_the_NAND Jul 29 12:11:04 ssam: no consensus on u-boot ;) Jul 29 12:11:20 ssam: i think we have no chance to get consensus. Jul 29 12:12:24 but u-boot would make it easier for me to add a kernel argument? Jul 29 12:12:49 ssam: and it is not worth in fact. is see no problem if different people using different bootloaders. Jul 29 12:12:56 ssam: you're free to switch to u-boot of course Jul 29 12:12:57 xrmx: I believe that is because 'from what I heard was the 0.9.x line declared as development releases by vala- upstream.' Jul 29 12:12:58 yeah Jul 29 12:13:13 we don't need consensus on bootloaders :) Jul 29 12:13:23 ssam: it's hard to say. I'm using Qi only because it does not talk to glamo Jul 29 12:13:39 lindi-: which is of course not true Jul 29 12:13:47 maybe someone should port lilo ;-) Jul 29 12:13:47 lindi-: as it of course talks. Jul 29 12:14:02 lindi-: overwise how do you load image from usd... Jul 29 12:14:04 gena2x: ok, it talks less to glamo than u-boot? ;) Jul 29 12:14:46 lindi-: yeah, 'less' Jul 29 12:15:10 and doesn't talk to jbt Jul 29 12:15:38 but the fact is, I haven't had a single WSOD for at least a year I think Jul 29 12:15:56 lindi-: i have no WSODs on uboot too, now. Jul 29 12:16:08 if u-boot changes were upstream I'd certainly reconsider swithing to it Jul 29 12:16:16 but for now Qi works and changing anything is risky :P Jul 29 12:16:37 so I prefer it now since I am using it now :P Jul 29 12:17:07 heh. Jul 29 12:17:45 the whole bootmenu thing is an Openmoko addition to u-boot, and it's not upstream Jul 29 12:18:35 ThibG: interesting. Jul 29 12:19:15 ThibG: so main problems of uboot 'it talks to glamo' is artificial in fact? Jul 29 12:19:39 gena2x: well the u-boot that works on openmoko talks to glamo? Jul 29 12:19:46 I thougt of it as an optional thing used only by some devices, and in fact, it's used only on FR, and isn't present upstream Jul 29 12:20:05 that said, there is no FR support at all in upstream's u-boot Jul 29 12:20:56 lindi-: i mean we can do u-boot which not talks to glamo. Jul 29 12:21:33 gena2x: that sounds nice Jul 29 12:22:03 (that said, I find the menu helpful, although it's quite weird in the uboot env) Jul 29 12:22:05 gena2x: you probably want to keep the menu still and operate it with leds and buttons Jul 29 12:23:13 lindi-: but i personally prefer menu. and see what is happening during boot. Jul 29 12:23:45 anyway, bringing u-boot in order for upstream sounds like big job. Jul 29 12:24:01 i'll better to do something more useful. Jul 29 12:24:15 as bootloadrs now are working for everyone. Jul 29 12:24:28 and i'm happy with my u-boot :) Jul 29 12:26:02 may be this were true reason to switching to qi - instead of bringing u-boot (which were written by on e person) in order, someone decided just to rewrite 'all from scratch'. Jul 29 12:26:51 it's common idea. Jul 29 12:27:45 mrmoku: ok, but you mean just replace ogsmd with fsogsmd, in: Jul 29 12:27:55 mrmoku: did ffmpeg build for you ? Jul 29 12:28:21 mrmoku: mdbus2 -s org.freesmartphone.fsogsmd /org/freesmartphone/GSM/Device org.freesmartphone.GSM.PDP.SetCredentials internet internet internet Jul 29 12:28:48 that didn't seem to work.. Jul 29 12:29:26 so, long list of invalid reasons were invented '2mb image limit, boot speed' and nobody really know about glamo. Jul 29 12:30:56 b0ef: no, what distro do you use? and do you have disabled = 1 in /etc/freesmartphone.conf in [ogsmd] section? Jul 29 12:31:14 b0ef: ogsmd is a subsystem of the (old) FSO1 python thing Jul 29 12:31:26 b0ef: fsogsmd is the (new) FSO2 vala rewrite of that subsystem Jul 29 12:31:38 "2MB image limit", were does this thing come from? Jul 29 12:31:40 mrmoku: I'm on shr-t Jul 29 12:31:40 both have the same dbus path (org.freesmartphone.ogsmd) Jul 29 12:31:52 b0ef: current shr-t? Jul 29 12:31:58 yes Jul 29 12:32:11 hmm... I thought shr-t is on fsogsmd as well... Jul 29 12:32:29 maybe ogsmd is just enabled in /etc/freesmartphone.conf... Jul 29 12:32:36 * mrmoku checks shr-t packages Jul 29 12:32:58 mrmoku: I only have a /etc/freesmartphone/ directory;) Jul 29 12:33:05 err Jul 29 12:33:08 /etc/frameworkd.conf Jul 29 12:33:19 too long ago :P Jul 29 12:35:35 right, I don't have disable=1 in the [ogsmd] section Jul 29 12:36:41 ThibG: it were written in the list of reasons to switch to qi Jul 29 12:36:47 b0ef: ok, then it's clear why the gprs script is failing Jul 29 12:37:16 mrmoku: so this script is for shr-u, then? Jul 29 12:37:18 b0ef: you have to merge those two calls (SetCredentials + ActivateContext) into just one call to ActivateContext with the three parameters Jul 29 12:37:20 ThibG: but actually, just u-boot were setup to load image not bigger than 2Mb. Jul 29 12:37:21 yep Jul 29 12:40:03 mrmoku: aiight, I'll try; thanks;) Jul 29 12:40:58 of course, this is just as i am seening things. and i am not pretending to be absolute truth. Jul 29 12:41:00 yw Jul 29 12:43:36 mickey|office, have you seen this? http://pvanhoof.be/blog/index.php/2010/05/13/ipc-performance-the-report Jul 29 13:09:18 xrmx: yep, seen it Jul 29 13:09:54 no surprising results to me though Jul 29 13:22:11 Hello Jul 29 13:22:13 philip nails it with his last sentence Jul 29 13:22:39 "if you have to pass a lot of data between two programs and don’t need a lot of flexibility, then DBus is not the right answer, and never intended to be." Jul 29 13:24:53 How Can I activate eth0 on debian with shr kms kernel without mdbus ? Jul 29 13:26:14 sunr: try omhacks Jul 29 13:26:34 thanks i check now Jul 29 13:31:17 mickey|office, i was wondering if fso uses the dbus + fd trick or not Jul 29 13:32:17 xrmx: no, we don't, since we don't have high volumes to transfer Jul 29 13:32:37 dbus + fd makes sense if you have huge blobs, i.e. audio, video, etc. Jul 29 13:35:40 mickey|office, but the save on latency is pretty interesting, the report-1.odt has more details Jul 29 13:36:12 hmm Jul 29 13:36:14 i don't think so Jul 29 13:36:32 1ms vs. 100ms does hardly matter when your modem takes 2 seconds to respond Jul 29 13:36:52 considering that some functions even are in the minute-space rather than the seconds-space :) Jul 29 13:45:04 om wifi power 1 : setting wifi power : no such file or directory / any ideas ? Jul 29 13:47:27 sunr: module loaded? Jul 29 13:47:43 sunr: g_ether or something similar (sorry no access to device atm) Jul 29 13:50:03 I think ar6000 is in kernel because .ko Jul 29 13:50:33 will see around g_ether Jul 29 13:50:39 sunr: 2.6.32? Jul 29 13:55:39 one more hint about wifi: Jul 29 13:55:51 lindi : 2.6.32.13 Jul 29 13:56:02 sunr: that has no power_on node for wifi Jul 29 13:56:06 echo s3c2440-sdi > $P/bind; dmesg | tail -n 20; echo s3c2440-sdi > $P/unbind Jul 29 13:56:12 P=/sys/bus/platform/drivers/s3c2440-sdi Jul 29 13:56:12 echo s3c2440-sdi > $P/bind; dmesg | tail -n 20; echo s3c2440-sdi > $P/unbind Jul 29 13:56:33 will reload in-kernel ar6000 stuff. Jul 29 13:56:36 I think I did try to find bind/unbind node for wlan under 2.6.32 but failed Jul 29 13:56:43 gena2x: what's $P? Jul 29 13:56:49 ah Jul 29 13:56:53 yeah :) Jul 29 13:56:59 if that works I'll add it to omhacks Jul 29 13:57:09 ops Jul 29 13:57:09 I'm quite sure I was unable to find the node, hmm Jul 29 13:57:10 wait. Jul 29 13:57:17 isn't this already added to omhacks? Jul 29 13:57:25 this is plaing bind/unbind. Jul 29 13:57:30 this is plain bind/unbind. Jul 29 13:57:32 Gabrys: no since I was unable to find it under 2.6.32 Jul 29 13:57:49 find what Jul 29 13:57:56 the dmesg and tail is of course for other purpose. Jul 29 13:58:04 Gabrys: node for turning wlan on and off Jul 29 13:58:20 gena2x: there is no sys/bus/platform/drivers/s3c2440-sdi under 2.6.32 here Jul 29 13:58:27 strange Jul 29 13:58:42 maybe you need to load a module or something Jul 29 13:58:56 Gabrys: well wlan works so the module is working :) Jul 29 13:59:17 debian-gta02:/home/gena/theoraarm# uname -a Jul 29 13:59:17 Linux debian-gta02 2.6.34b45snddebug #92 Wed Jul 28 19:17:39 MSD 2010 armv4tl GNU/Linux Jul 29 13:59:17 debian-gta02:/home/gena/theoraarm# find /sys/bus/platform/|grep s3c2440-sdi Jul 29 13:59:35 /sys/bus/platform/devices/s3c2440-sdi Jul 29 13:59:37 /sys/bus/platform/drivers/s3c-sdi/s3c2440-sdi Jul 29 13:59:49 but last time i checked this were in .29 Jul 29 14:02:27 gena2x: yes works with 2.6.29 Jul 29 14:02:30 gena2x: but not with 2.6.32 Jul 29 14:03:02 Gabrys: if you can figure out how to turn wifi off in 2.6.32 then please tell me :) Jul 29 14:03:50 lindi-: I'm currently using 2.6.29 and I'm not eager to switch before qtmoko v25 is released :-) Jul 29 14:04:43 I'm staying with 2.6.29 too since 2.6.32 does not seem to be usable Jul 29 14:04:50 -bash: /sys/bus/platform/drivers/s3c2440-sdi/unbind: No such file or directory Jul 29 14:04:57 2.6.32.13 Jul 29 14:05:02 lindi-: export P=/sys/bus/platform/drivers/s3c-sdi/ ? Jul 29 14:06:19 now it's different :) Jul 29 14:07:08 bash: echo: write error: No such device Jul 29 14:07:12 shit Jul 29 14:08:01 ok, that were just hint. and really for .29, as i last time fixed problem in ar6000 in .29. Jul 29 14:08:20 thanks anyway Jul 29 14:08:24 gena2x: not currently running 2.6.32 since I need sound for my video test :) Jul 29 14:08:43 lindi-: i am working on sound. promise it soon :) Jul 29 14:10:41 lindi-: isn't rfkill the way in 2.6.32? Jul 29 14:11:01 mrmoku: I am not sure Jul 29 14:11:14 mrmoku: it kills my usb wlan too? Jul 29 14:11:33 usb dongle? Jul 29 14:11:36 yep Jul 29 14:11:50 why should it? Jul 29 14:14:27 root@om-gta02 ~ # cat /sys/class/rfkill/*/name Jul 29 14:14:27 gta02-pm-bt Jul 29 14:14:27 ar6000 Jul 29 14:14:35 mrmoku: cause fso does fd = Posix.open( Path.build_filename( devfs_root, "rfkill" ), Posix.O_RDWR ); and I don't see them mentioning ar6000 anywhere Jul 29 14:15:42 mrmoku: I can try that next time I boot 2.6.32 Jul 29 14:15:52 k Jul 29 14:16:04 mrmoku: have you tested that it works? Jul 29 14:16:30 I nhave no usb dongle Jul 29 14:16:46 but I can test to turn on/off wifi and bt independently Jul 29 14:19:12 i have written script which automatically downloads contact photos from facebook, gravatar, twitter, blip or gadu-gadu and adds them to contact list :D Jul 29 14:19:36 rfkill list ar6000 can it turn it on ? Jul 29 14:20:35 mrmoku: does the normal rfkill tool work? Jul 29 14:23:16 lindi-: what is the normal rfkill tool? Jul 29 14:24:35 with Request/ReleaseResource all works fine Jul 29 14:26:57 mrmoku: http://packages.debian.org/sid/rfkill Jul 29 14:34:46 lindi-: building... Jul 29 16:18:49 gena2x: FCSE patch works well Jul 29 16:19:01 dos1: i guess that's the way to do it since the FR has no camera Jul 29 16:19:50 gena2x: do you have .opk for lmbench? Jul 29 16:25:01 Martix: hi. no. Jul 29 16:25:23 Martix: lmbench is intended to run with 'make results' Jul 29 16:25:44 Martix: so packaging it seem a bit pointless for me. Jul 29 16:27:41 Martix: and i'm on debian/qtmoko, don't forget :) Jul 29 16:35:22 ok Jul 29 16:59:21 noWS kernel + FCSE support: http://downloads.openmoko.cz/martix/shr/kernel/ Jul 29 17:04:44 FCSE? Jul 29 17:05:47 ThibG: http://lwn.net/images/conf/rtlws11/papers/proc/p01.pdf Jul 29 17:06:59 Martix: need reliable tests :) Jul 29 17:07:27 hm ok Jul 29 17:07:29 thanks :) Jul 29 17:07:32 sounds interesting Jul 29 17:07:59 this is also interesting http://lwn.net/images/conf/rtlws11/papers/proc/p11.pdf Jul 29 17:10:26 xrmx: preemption is other beast. our most important realtime event source (touchscreen) has 1 hz latency minimum with .34 driver. Jul 29 17:10:46 i mean HZ Jul 29 17:11:46 HZ=1/200 sec Jul 29 17:13:17 this is independent of preemption. Jul 29 17:18:16 gena2x, i haven't read the papers in depth but look at the cyclictest results, the preempt-rt worst case is half of the other with fcse for arm9 Jul 29 17:27:44 xrmx: i think where is no need of _realtime_ things in fr. we simply have no such problem. do we have problem of too high latency of input events? Jul 29 17:29:00 depend, you may have a loaded system and want a good reactivty, no ? Jul 29 17:29:56 xrmx: instead, we have general speed problem. this means we should do general optimization and not add complexity of for example sheduling irq handlers. Jul 29 17:30:46 misc: i think we should have a _problem_ to solve. do you have testcase with problem? Jul 29 17:31:02 misc: which kind of load? Jul 29 17:31:39 misc: cpu, io? reaction to what? to touchscreen? to serial? to wifi data? Jul 29 17:31:47 misc: to usb? Jul 29 17:32:26 gena2x: technically, I think the example of having wifi and openvpn are not really relevent for most user Jul 29 17:33:08 gena2x: regarding my experience, i had the impression that qtmoko was sometimes sluggish when I received some call, but as 1) had to answer the phone 2) a old version of qtmoko on the phone, i didn't looked much Jul 29 17:33:40 so of course, it will not help much for this, but maybe I had a missbehaving application in the background, or something like that Jul 29 17:34:01 would be good to understand that better. Jul 29 17:34:24 may be it just loads some images from disk (sd or flash) for example Jul 29 17:34:33 and this cause hight latency. Jul 29 17:34:51 JesusMcCloud-lt: thanks for taking that bug :-) Jul 29 17:35:03 gena2x, that's oprofile job no? Jul 29 17:35:06 what was troubling me was that I "clicked" answer, and I had to wait to get the voice Jul 29 17:36:18 xrmx: i think, oprofile can help much with initial analysis. Jul 29 17:37:08 misc: too many possibilites. but oprofile can really help. Jul 29 17:37:32 gena2x: well, i assume that I am the only one to face the problem ? Jul 29 17:38:45 misc: i rarely recieve call to openmoko, most time i am calling back people then i load my qtmoko. Jul 29 17:39:15 misc: also seem i have some problem with gsm on resume. Jul 29 17:39:32 misc: problem is that sometimes it stops recieving call. Jul 29 17:41:25 mrmoku, nP i just stumbeled upon it, but it will still have to wait till middle of august, since i won't be home any sooner Jul 29 17:43:32 JesusMcCloud-lt: np Jul 29 18:10:59 gena2x: FCSE has been tested it with LTP, but I'll test it also with lmbench Jul 29 18:12:57 freesmartphone.org: 03mickey 07cornucopia * rba807ab0165c 10/fsogsmd/src/ (4 files in 3 dirs): fsogsmd: more stub implementation for GetSimUnlockCounters Jul 29 18:12:59 freesmartphone.org: 03mickey 07cornucopia * r87c0ae97ec33 10/fsogsmd/src/plugins/modem_ti_calypso/ (commands.vala mediators.vala): fsogsmd: modem_ti_calypso: implement org.freesmartphone.GSM.SIM.GetUnlockCounters() Jul 29 18:13:25 mrmoku: ready for use Jul 29 18:14:08 >>> gsmsim.GetUnlockCounters() Jul 29 18:14:08 { 'SIM PIN': 3, 'SIM PUK': 2} Jul 29 18:22:08 boo Jul 29 18:23:24 ~seen playya Jul 29 18:23:26 playya <~playya@unaffiliated/playya> was last seen on IRC in channel #openmoko-cdevel, 2d 6h 50m 30s ago, saying: 'leviathan, http://downorisitjustme.com/res.php?url=cgit.openembedded.org'. Jul 29 18:28:52 freesmartphone.org: 03mickey 07cornucopia * rcf6197d84edc 10/fsogsmd/src/plugins/modem_ti_calypso/ (commands.vala mediators.vala): fsogsmd: modem_ti_calypso: GetSimUnlockCounters: include SIM PIN2 and SIM PUK2 in output Jul 29 18:38:28 mickeyl: nice :) Jul 29 18:38:50 mickeyl: PUK 2 ? Jul 29 18:38:53 is that normal? Jul 29 18:39:07 pretty much yes Jul 29 18:39:11 very rarely used though Jul 29 18:39:28 PUK2 you need when you fscked up with PIN2 Jul 29 18:39:36 and if I fail to enter that long PUK correctly two times I need a new SIM? Jul 29 18:39:42 no Jul 29 18:39:46 that was my false output Jul 29 18:39:47 it's like: Jul 29 18:39:55 >>> gsmsim.GetUnlockCounters() Jul 29 18:39:55 { 'SIM PIN': 3, 'SIM PIN2': 2, 'SIM PUK': 10, 'SIM PUK2': 10} Jul 29 18:40:02 ahh, ok :) Jul 29 18:59:14 mickeyl: what about call forwarding? it's seems to be still unimplemented in fsogsmd Jul 29 18:59:21 mickeyl: registering with selected network too Jul 29 19:00:00 true, never got around to do that, since i'm not sure whether it's being used at all Jul 29 19:00:13 i never managed to forward a call Jul 29 19:00:14 did anyone? Jul 29 19:00:20 ah, wait Jul 29 19:00:27 i'm talking about deflection Jul 29 19:00:31 not regular forwarding settings Jul 29 19:01:14 mickeyl: i wanted to do module in shr-settings for forwarding settings, but i ended up with discovering that it's not implemented yet :) Jul 29 19:01:35 yeah, will fix that missing spot when i have a chance Jul 29 19:01:42 but i'm not using call forwarding though, so i'm not pushing ;) Jul 29 19:02:40 wrt. selecting a specific provider... is anyone using that? I find myself not even using it when roaming. Jul 29 19:02:56 there are some modems who can't even do that :) Jul 29 19:03:10 I used that with my old nokia from time to time... Jul 29 19:03:27 can live without though :) Jul 29 19:03:45 well, i'll certainly add it Jul 29 19:03:50 just has very low priority Jul 29 19:04:06 yup, agree Jul 29 19:04:19 that SetFunctionality thing would be more interesting :) Jul 29 19:04:45 1 -> 4 -> 1? Jul 29 19:05:04 yup Jul 29 19:05:21 *nod* probably just a one-liner somewhere Jul 29 19:06:02 yeah Jul 29 19:06:25 hmm, let me try something Jul 29 19:08:42 please try Jul 29 19:08:44 freesmartphone.org: 03mickey 07cornucopia * re93d61565b90 10/fsogsmd/src/lib/atmediators.vala: Jul 29 19:08:44 freesmartphone.org: fsogsmd: set modem status back to ALIVE_SIM_READY when we are no Jul 29 19:08:44 freesmartphone.org: longer registered. Jul 29 19:09:23 mickeyl: ok, thanks Jul 29 19:16:50 hmm, no Jul 29 19:17:04 there's a security check missing Jul 29 19:19:25 hi mickeyl Jul 29 19:20:05 I talked to kvalo today, he wants me to send a mail Jul 29 19:20:17 GNUtoo|laptop: tried to add sound Jul 29 19:20:23 GNUtoo|laptop: doesn't work in htc-msm-2.6.32 Jul 29 19:20:38 what does doens't work mean? Jul 29 19:20:51 GNUtoo|laptop: I tried to copy from your tree (rebase branch) and from the original codeaurora original branch Jul 29 19:21:25 GNUtoo|laptop: arch/arm/mach-msm/qdsp6 in our android-msm-2.6.32 based branch lacks required stuff Jul 29 19:21:32 and? Jul 29 19:21:33 config QSD:SOUND Jul 29 19:21:37 _ Jul 29 19:21:43 GNUtoo|laptop: doesn't work without it it looks Jul 29 19:21:51 ah ok Jul 29 19:22:03 GNUtoo|laptop: I am wondering why you use 2.6.29 kernel as base. Jul 29 19:22:14 GNUtoo|laptop: there is no codeaurora tree with 2.6.32+ that has alsa, right ? Jul 29 19:22:28 dcordes_, I use 2.6.32 Jul 29 19:22:56 GNUtoo|laptop: I mean where you copy ALSA from. Jul 29 19:23:15 dcordes_, what kernel do you use? Jul 29 19:23:20 why not using my kernel Jul 29 19:23:24 it's not very modified Jul 29 19:24:44 dcordes_, you've got hd2? Jul 29 19:24:52 do you want me to build a kernel for you? Jul 29 19:24:57 GNUtoo|laptop: I will wait for mainline msm alsa. Jul 29 19:24:59 freesmartphone.org: 03mickey 07cornucopia * rb163e36f9c96 10/fsogsmd/src/lib/modem.vala: Jul 29 19:25:00 freesmartphone.org: fsogsmd: improve modem state logic in order to fix #580 Jul 29 19:25:00 freesmartphone.org: - do nothing, if we are already in the requested state (not even with 'force') Jul 29 19:25:00 freesmartphone.org: - go 'back' only if forced Jul 29 19:25:09 dcordes_, you'll wait a very long time then Jul 29 19:25:12 GNUtoo|laptop: yes go for it. our kernel is here http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=shortlog;h=refs/heads/htc-msm-2.6.32 Jul 29 19:25:25 mickeyl: ok, will try that one... and bump fso-glib too Jul 29 19:26:13 mrmoku: yep. and specs. Jul 29 19:26:54 dcordes_, I didn't understand...it didn't compile? Jul 29 19:28:19 dcordes_, I can help since I'll need to do it for nexusone Jul 29 19:28:32 GNUtoo|laptop: not following you now Jul 29 19:29:29 dcordes_, what phone do you have? Jul 29 19:29:36 hd2? Jul 29 19:29:46 htcleo yes Jul 29 19:29:47 I bet it won't work with an android kernel Jul 29 19:29:51 ok Jul 29 19:29:58 sorry need to check something else now Jul 29 19:30:07 dcordes_, do you want me to rebase alsa for you? Jul 29 19:30:16 I can't do it now tough Jul 29 19:30:19 I'm on 3g Jul 29 19:37:33 mickeyl, did you see that: http://git.linuxtogo.org/?p=groups/mobile-linux/kernel.git;a=shortlog;h=refs/heads/htc-msm-2.6.32 it seem highly developped Jul 29 19:37:33 GNUtoo|laptop: did you read that vodafone took the nexus one out of the program? Jul 29 19:37:38 that pretty much kills it IMO for us Jul 29 19:37:57 hmmm Jul 29 19:38:03 * mrmoku read that too Jul 29 19:38:08 yeah, seen that. but i think that's an android-style kernel Jul 29 19:38:12 and that google stops selling in us Jul 29 19:38:13 graziano told me that dream is not sold anymore Jul 29 19:38:16 so I'm not sure the commits are any good for us Jul 29 19:38:19 mrmoku: yep Jul 29 19:38:33 I knew google doesn't sell nexusone anymore Jul 29 19:38:39 but vodaphone... Jul 29 19:39:59 mickeyl: hmm, http://git.freesmartphone.org/?p=specs.git;a=blob_plain;f=html/org.freesmartphone.GSM.SIM.html;hb=HEAD#StoreMessage still points to this "For a list of valid properties see SendMessage" -- can you point me what property I should use to get delivery notification(?)? Jul 29 19:42:42 lindi-: for ogsmd, I'm afraid i don't know, since alphaone implemented the SMS stuff. for fsogsmd, right now only SendTextMessage is supporting delivery notifications. The StoreMessage interface is not supposed to be used any longer for text messages. I'll probably change it to allow sending binary SMS soon. Jul 29 19:42:55 mickeyl: ok so this api too has changed? Jul 29 19:43:21 not yet, no Jul 29 19:43:40 but iirc the properties are ignored in fsogsmd Jul 29 19:43:54 since I decided to come up with the more user oriented SendTextMessage Jul 29 19:44:04 and rather provide another interface for binary sms Jul 29 19:44:17 are there any existing clients that use deliver notifications? I could read source Jul 29 19:44:44 perhaps an older version of SHR, back when it used ogsmd Jul 29 19:45:16 mickeyl: opimd-utils Jul 29 19:45:22 s/mickeyl/lindi-/ Jul 29 19:45:24 dos1 meant: lindi-: opimd-utils Jul 29 19:45:32 lindi-: "status-report-indicator" might be it Jul 29 19:45:43 so setting that to True? Jul 29 19:46:00 is there ever a reason to not use status reports? Jul 29 19:46:14 hmm Jul 29 19:46:16 lindi-: status-report-request Jul 29 19:46:18 good question Jul 29 19:46:32 dos1: ah, and not status-report-indicator? Jul 29 19:46:33 since 99% of people don't use status reports, i guess there is one Jul 29 19:46:35 yup Jul 29 19:46:38 sorry, status-report-indicator is the other side Jul 29 19:46:46 status-report-request sounds good Jul 29 19:46:57 there's a certain inherent unreliability associated with calypso Jul 29 19:47:06 really? Jul 29 19:47:07 hmm, bad grammar Jul 29 19:47:15 i found sms sending pretty solid these days Jul 29 19:47:27 so I'm going to want every extra check if at all possible :) Jul 29 19:47:31 understandable Jul 29 19:48:08 unfortunately you have to do quite an extra amount of work for concatenated sms Jul 29 19:48:12 ogsmd will not assist you in that Jul 29 19:48:18 fsogsmd has that ;) Jul 29 19:48:33