**** BEGIN LOGGING AT Fri Jun 01 02:59:58 2012 Jun 01 06:21:05 in the process of trying to install shr on a new palm pre+ Jun 01 06:41:21 cenobyte_: do you know support for pre was dropped from FSO lately? :( Jun 01 06:41:35 cenobyte_: are you a capable interested developer? Jun 01 06:42:31 ya, i read that Jun 01 06:42:49 i am not very capable but interested Jun 01 06:43:09 i don't even know Python well enough to be dangerous Jun 01 06:43:26 i'm using the automatic installer for shr as we speak Jun 01 07:08:34 freesmartphone.org: 03morphis 07cornucopia * r3ac63c1a8653 10/fsogsmd/ (src/lib/at/atresultparser.vala tests/testatresultparser.vala): fsogsmd: lib: at: extend our AT result parser to handle multiple lines with different prefixes Jun 01 07:08:34 freesmartphone.org: 03morphis 07cornucopia * ra2afe52ae85c 10/fsogsmd/ (6 files in 3 dirs): fsogsmd: lib: at: rename AtResultParser class to AtResultIter Jun 01 07:09:50 ok so i have shr on my pre+ but have no wifi and i do not see a firmware on the main site Jun 01 07:15:44 cenobyte_: what do you see in dmesg? Jun 01 07:18:49 a whole lot, but mainly usbcore: modem power off Jun 01 07:18:55 and on Jun 01 07:19:59 the wifi settings say wifi power is off and no way to slide to on Jun 01 07:20:26 cenobyte_: i mean about the wifi firwmare Jun 01 07:21:27 well at first boot it said i may need non-free firmware Jun 01 07:21:40 so i went to the site but did not find it for pre+ Jun 01 07:24:18 is there a pdf reader for shr? Jun 01 08:01:22 freesmartphone.org: 03morphis 07cornucopia * re8c642854da9 10/fsogsmd/ (src/lib/at/atresultiter.vala tests/testatresultiter.vala): fsogsmd: lib: at: fix some bugs in the new AtResultIter class Jun 01 08:01:22 freesmartphone.org: 03morphis 07cornucopia * r65287b899c50 10/fsogsmd/ (3 files in 3 dirs): fsogsmd: lib: at: implement parsing of +CCLC query responses Jun 01 08:33:48 JaMa|Off: ping Jun 01 09:21:05 JaMa|Off: if you have some time can you take a look at this: http://git.shr-project.org/git/?p=meta-smartphone.git;a=commit;h=adf624ccc0b9ad77936cabe4ffa592a05b58a023 Jun 01 09:29:51 how would I define a CAPability like e.g. CAP_CHOWN for a 'normal' user? From a sysadmin POV Jun 01 09:30:36 morphis: ipng Jun 01 09:30:55 I guess a RTFM with *pointer* to the right Mpage will suffice Jun 01 09:31:16 hi mickey_office Jun 01 09:31:26 any clue about CAPs? Jun 01 09:32:08 mickey_office: pong Jun 01 09:33:29 DocScruSTM: good morning, i'm afraid not Jun 01 09:33:42 morphis: hi. what do you have in mind with the AtResultParser? going away from regexes? Jun 01 09:34:23 mickey_office: don't really know; I just didn't have fun writing complex regular expressions like for +CCFC Jun 01 09:34:41 i can do them, if you want Jun 01 09:35:03 you like the use of regular expressions more than having a different parser? Jun 01 09:35:13 i view using regexes as a big advantage over things like ofono Jun 01 09:35:37 i thought a lot about all that before starting with the vala-based fsogsmd Jun 01 09:35:53 ok Jun 01 09:36:02 why they are a big advantage? Jun 01 09:36:02 mickey_office: ++ Jun 01 09:36:10 and regexes offer a great way to deal with the slight differences of the 07.07 implementation different modems use Jun 01 09:36:21 yep Jun 01 09:36:24 portability Jun 01 09:36:28 correct Jun 01 09:36:42 means in detail? Jun 01 09:36:46 often, you "just" have to formulate the regex a bit more allowing to add compatibility to another vendor Jun 01 09:37:03 like one vendor adds more space than the other Jun 01 09:37:07 correct Jun 01 09:37:16 or optional arguments Jun 01 09:37:33 which we use in some places Jun 01 09:37:39 yes I saw Jun 01 09:40:59 hm Jun 01 09:41:42 with regular expressions you have less code lines to maintain but one (maybe complex) one Jun 01 09:42:17 mickey_office: in general I was just playing around with parsing AT commands, just to come closer to a better understand of the whole thing Jun 01 09:42:27 s/understand/understanding/ Jun 01 09:42:28 morphis meant: mickey_office: in general I was just playing around with parsing AT commands, just to come closer to a better understanding of the whole thing Jun 01 09:42:59 mickey_office: btw. what do you think about dbus properties? Jun 01 09:44:05 morphis: i don't like them too much, that's why I didn't use them since now. Plus, many bindings did not have full support for them Jun 01 09:44:17 i don't mind introducing them, though Jun 01 09:44:21 at least Vala supports them fully now Jun 01 09:44:30 yes Jun 01 09:44:49 I was thinking about using them (already adjusted specs and vala-dbus-binding-tool to support them) Jun 01 09:44:59 means not redefining the whole API Jun 01 09:46:11 and yes vala supports them but you have to take care about the PropertyChanges event on your own which will blow up your code Jun 01 09:46:38 right. the thing is, I don't see a real benefit. I view API that heavily use properties as having no semantics Jun 01 09:46:46 since everything goes via stringified properties Jun 01 09:46:51 so there's hardly any API contract Jun 01 09:47:02 but that's just my view Jun 01 09:47:05 bbiab, meeting Jun 01 09:49:58 hmmm, setcap(8) Jun 01 09:50:39 which AIUI is the hyper-SUID Jun 01 09:52:06 anyway man 7 capabilities is quite a nice read Jun 01 10:10:22 freesmartphone.org: 03morphis 07vala-dbus-binding-tool * r89e160e8d9fc 10/src/vala-dbus-binding-tool.vala: Complex types needs an owned get accessor Jun 01 10:17:33 mickey_office: I am currently reworking the call forwarding API; it would be nice if you can take short look at it (it's mostly the same as before but grouped together with a new interface org.freesmartphone.GSM.CallForwarding Jun 01 10:19:03 freesmartphone.org: 03morphis 07specs * rc3e1a135a92b 10/ (4 files in 3 dirs): Add new interface org.freesmartphone.GSM.CallForwarding Jun 01 10:19:03 freesmartphone.org: 03morphis 07specs * r484bf6b4f941 10/ (configure.ac html/index.html): Bump version to 2012.06.01.1 Jun 01 10:19:22 freesmartphone.org: 03morphis 07libfso-glib * rb90ea1fafcc2 10/configure.ac: Keep version in sync with fso-specs Jun 01 10:19:53 mickey_office: some method to retrieve the current configuration for the forwarding rule is still missing Jun 01 10:31:38 freesmartphone.org: 03morphis 07specs * r1354b51efbc6 10/html/ (index.html index.html.in): Add org.freesmartphone.GSM.CallForwarding interface to index.html page Jun 01 10:34:17 freesmartphone.org: 03morphis 07specs * r688f96bf6706 10/org.freesmartphone.GSM/org.freesmartphone.GSM.CallForwarding.xml.in: Use doc:summary instead of doc:description for arguments Jun 01 10:34:55 freesmartphone.org: 03morphis 07specs * ra9a50e4af055 10/html/org.freesmartphone.GSM.CallForwarding.html: Add missing regenerated html file Jun 01 10:44:16 morphis: looks good on quick glance Jun 01 10:44:18 morphis: ok, i'll take a look. have to reboot now, some services have gone zombies... Jun 01 10:44:25 * JaMa|Off off for rest of today.. Jun 01 10:45:01 JaMa|Off: ok Jun 01 10:45:09 JaMa|Off: we need to do the same for fsodeviced Jun 01 11:20:11 061 Jun 01 11:44:34 is it on purpose that +CLVL=255 was not ported from ogsmd to fsogmsd ? Jun 01 11:44:35 on ti calypso Jun 01 12:29:23 hello all Jun 01 12:31:05 I m working on ARM linux kernel..... I m facing problem regarding multiple screen resolution..... it is hardcoded in the kernel ..... and its takes the value from this struct .... http://pastebin.mozilla.org/1649123 Jun 01 12:31:21 how can i pass this values dynamically... Jun 01 12:47:31 snkt: you will need to either fix the code or get the person who wrote it to do that Jun 01 14:44:43 nvm I found the dbus api to set modem volume Jun 01 14:46:44 still might be good to trigger it for ti calypso automagically Jun 01 15:28:54 hi, how to use platform_battery on recent SHR for om-gta02 Jun 01 17:10:14 morphis, hi Jun 01 17:10:23 morphis, why did you remove the palmpre configs? Jun 01 17:15:06 GNUtoo-desktop: I removed the whole palmpre support from FSO Jun 01 17:15:41 morphis, why? Jun 01 17:15:53 as I already said on the ML Jun 01 17:15:59 not to maintain it I guess Jun 01 17:16:04 yes Jun 01 17:16:19 yes I tought you wouldn't remove it but just put it as unsuported Jun 01 17:16:20 no palmpre support anymore in FSO means no configs for it at all Jun 01 17:16:34 maybe you need for AUTOREV this one here: http://git.shr-project.org/git/?p=meta-smartphone.git;a=commit;h=adf624ccc0b9ad77936cabe4ffa592a05b58a023 Jun 01 17:16:42 so if a maintainer was to come and wanted to support palmpre for instance for debian it would be easier for him Jun 01 17:17:05 easier to do what? Jun 01 17:17:10 but I guess you don't even want to work on keeping the stuff for palmpre building Jun 01 17:17:37 the start for bringing palmpre to work again would have been easier Jun 01 17:17:54 if the support would be still there? Jun 01 17:18:09 it's not gone at all, 0.11 and 0.10 still have it Jun 01 17:18:22 so if someone wants to work on it, it can be easily added back again Jun 01 17:18:39 but you're right in makes life for me as FSO maintainer easier Jun 01 17:18:42 less dependencies Jun 01 17:18:51 less things I have to check for the release Jun 01 17:19:06 ok Jun 01 17:19:21 morphis: hey! I wonder if you do any real-life testing on FSO lately. I know there was a nasty issue when i'm in a conversation and somebody else calls in. fsogsmd behaved all right (i guess) but the audio scenarious were applied improperly. Jun 01 17:19:28 morphis, what do we do for htc devices, because sometimes people come up and want their machine to work with fso Jun 01 17:19:38 And i couldn't figure out the correct rule for that, and default rules.yaml was broken (in this regard). Jun 01 17:19:58 but until no one is going to work on palmpre support it will stay removed Jun 01 17:20:28 PaulFertser: real life, not really Jun 01 17:20:40 PaulFertser: just executing different test cases not written down yet somewhere Jun 01 17:21:18 PaulFertser: there is a bug report from JaMa|Off about a crashing fsodeviced when scenarios relies to not available ALSA controls Jun 01 17:21:30 morphis: ok, i see. If you by any chance will have some time to think about this use-case (being on phone and receiving another call) please consider this issue. Jun 01 17:21:31 PaulFertser: with latest SHR? Jun 01 17:22:06 morphis: i've read it of course, but it's unrelated. Nothing crashes for me. It's just that another incoming call most probably makes gsm voice scenario switched off due to fsoeventsd rule. Jun 01 17:22:16 morphis: nope, not latest. But i'm afraid it's still the same. Jun 01 17:22:34 PaulFertser: but that sounds like something is wrong with the daemon/ruleset applying the audio scenarios Jun 01 17:22:49 like justing testing if a call is coming but doesn't check if a call is already active Jun 01 17:23:15 morphis: yep, and i think that default ruleset is still broken. And there was some subtle issue that made it unclear how to write a proper ruleset. Jun 01 17:23:18 PaulFertser: but thats a SHR bug I would say Jun 01 17:23:49 which subtle issue? Jun 01 17:23:52 morphis: if it's not a limitation of fsoeventsd, then, well, kind of. Because fsoeventsd ships with a sample config, and i'm using that one, not the SHR. Jun 01 17:24:18 PaulFertser: there is still no real fsoeventsd, it's still oeventsd Jun 01 17:24:28 morphis: probably i couldn't figure out how to make it "ring" if there's an incoming call only when there's no onogoing call present. Jun 01 17:24:33 "active" Jun 01 17:24:35 hm Jun 01 17:24:55 PaulFertser: we're talking about the scenarios for the GTA02, right? Jun 01 17:25:39 So, i'm talking, the audio is all right, i have that call "ACTIVE", then someone else calls me, i get another call which is "INCOMING" (now i have two calls) and then audio stops (i bet due to oeventsd switching the alsa state). Jun 01 17:25:51 morphis: yes, the sample config for the gta02. Jun 01 17:25:58 ok Jun 01 17:27:23 My plan enables this "call waiting" functionality by default, so every time i'm having a conversation and somebody else calls me, i have to "Release" both the calls and call one of the numbers back to continue. :) Jun 01 17:27:44 PaulFertser: this one: http://git.freesmartphone.org/?p=framework.git;a=blob;f=etc/freesmartphone/oevents/om-gta02/rules.yaml;h=8f7e2a7a9fc849d816f2ad03b3f32d610bb5fc40;hb=HEAD Jun 01 17:27:46 Funny i seem to be the only one bothered by this, but otoh nobody told it works all right for him either. Jun 01 17:28:13 :) Jun 01 17:28:20 I use the gta02 as main phone too under SHR tough Jun 01 17:28:25 but I don't get many calls Jun 01 17:28:29 nor call often Jun 01 17:28:40 PaulFertser: it would be really nice if you can write this all together in a FSO bug report so it don't get lost Jun 01 17:28:46 so often if it doesn't work and I forget to test, I don't even notice that it doesn't work Jun 01 17:28:54 but I do notice if music is broken Jun 01 17:29:46 PaulFertser: or this one here: http://git.freesmartphone.org/?p=framework.git;a=blob;f=etc/freesmartphone/oevents/rules.yaml;h=695f5ff7a7c323a36fe13bbec3e7c1356d7e5122;hb=HEAD Jun 01 17:29:52 morphis: sure thing, i will. Jun 01 17:30:06 morphis: let me paste my config Jun 01 17:30:44 PaulFertser: the second rules.yaml does SetScenario(gsmhandset) when Or(CallListContains("outgoing"), CallListContains("active")) Jun 01 17:31:00 lindi-: wishlist item if i may :) make your awesome paste service accept connections over IPv6 :) Jun 01 17:32:22 freesmartphone.org: 03morphis 07cornucopia * r276b9340aeaa 10/RELEASE-TODO: Update RELEASE-TODO Jun 01 17:32:34 morphis: http://paste.debian.net/172398 Jun 01 17:34:02 morphis: i suspect that CallStatus() is ambigous: it's not clear what call it actually represents (and CallList can contain many). Jun 01 17:34:07 freesmartphone.org: 03morphis 07morphis/service-rework * rf458aa7863b5 10cornucopia/fsogsmd/src/plugins/dbus_service/ (5 files): Jun 01 17:34:07 freesmartphone.org: fsogsmd: dbus_service: start reworking internal bits of dbus service management Jun 01 17:34:07 freesmartphone.org: Signed-off-by: Simon Busch Jun 01 17:34:07 freesmartphone.org: 03morphis 07morphis/service-rework * r28cfe54c51f0 10cornucopia/fsogsmd/src/plugins/dbus_service/ (15 files): Jun 01 17:34:07 freesmartphone.org: fsogsmd: dbus_service: reimplement all missing services Jun 01 17:34:07 freesmartphone.org: Signed-off-by: Simon Busch Jun 01 17:34:08 freesmartphone.org: 03morphis 07morphis/service-rework * r79b9fe6c6f14 10cornucopia/fsogsmd/src/plugins/dbus_service/ (5 files): Jun 01 17:34:08 freesmartphone.org: fsogsmd: dbus_service: reintegrate resource registration and handling Jun 01 17:34:09 freesmartphone.org: Signed-off-by: Simon Busch Jun 01 17:34:09 freesmartphone.org: 03morphis 07morphis/service-rework * r6be731228364 10cornucopia/fsogsmd/src/plugins/dbus_service/device_powersupply_service.vala: Jun 01 17:34:10 freesmartphone.org: fsogsmd: dbus_service: add possible thrown exception type FreeSmartphone.Error for two methods Jun 01 17:34:10 freesmartphone.org: Signed-off-by: Simon Busch Jun 01 17:34:18 freesmartphone.org: 03morphis 07morphis/service-rework * r8578a4b12b50 10cornucopia/fsogsmd/src/plugins/dbus_service/ (plugin.vala service.vala servicemanager.vala): Jun 01 17:34:18 freesmartphone.org: fsogsmd: dbus_service: assign modem to services when modem construction is done Jun 01 17:34:18 freesmartphone.org: Signed-off-by: Simon Busch Jun 01 17:34:18 freesmartphone.org: 03morphis 07morphis/service-rework * r81f1f58981d8 10cornucopia/fsogsmd/src/plugins/dbus_service/modemfactory.vala: fsogsmd: dbus_service: add support for phonesim modem again Jun 01 17:34:19 freesmartphone.org: 03morphis 07morphis/service-rework * r76fa4a2ec2cc 10cornucopia/fsogsmd/src/plugins/dbus_service/gsm_pdp_service.vala: fsogsmd: dbus_service: implemented required roaming_allowed property Jun 01 17:34:19 freesmartphone.org: 03morphis 07morphis/service-rework * r30bdfbc403e6 10cornucopia/fsogsmd/src/ (4 files in 2 dirs): fsogsmd: dbus_service: implement service provider interface to make service accessible Jun 01 17:35:13 PaulFertser: yes Jun 01 17:35:18 freesmartphone.org: 03morphis 07morphis/service-rework * r3f6b3c74c1a9 10cornucopia/fsogsmd/src/plugins/dbus_service/modemfactory.vala: fsogsmd: dbus_service: remove configuration for qualcomm_palm modem type which is gone Jun 01 17:35:58 PaulFertser: I don't really like this oeventsd concept anymore, it's limited for special use cases Jun 01 17:36:45 morphis: there was that cool idea about using prolog-based framework Flora (?) + something to provide for some ultimate flexibility but alas... Jun 01 17:37:01 yes I know, it's still in git.shr-project.org Jun 01 17:37:53 morphis: well, this oeventsd concept sort-of works, so probably fixing this one extra special case would be nice since it adds some immediate pain to the users (or probably only me). Jun 01 17:38:06 yes for sure Jun 01 17:38:17 I just have this one in my mind for a long time Jun 01 17:39:01 morphis: btw, i'm really glad you stepped up in FSO maintenance, and i can hardly express how much i appreciate all your hard work. Thanks for the discussion, i'll file a ticket. Jun 01 17:42:12 morphis: btw, what's your take on the future of the telephony API within FSO? It really seems too low-level and GSM-centric. Jun 01 17:42:29 Also, there's no support for the bluetooth headsets now (since ophoned was broken). Jun 01 17:44:05 PaulFertser: good question Jun 01 17:44:24 in the first step I want to implement most missing features in the API and doing some cleanups Jun 01 17:44:38 for example call forwarding was not supported for a long time now Jun 01 17:44:53 but it will stay lowlevel Jun 01 17:45:10 means we need further integration in some higher level thing like telepathy Jun 01 17:45:35 I saw recently a video where they showed ofono telepathy integration Jun 01 17:45:51 doing voice calls and sending SMS message with empathy is really nice Jun 01 17:45:55 * GNUtoo-desktop loves bluetooth headset Jun 01 17:46:08 it's the only way to make gta02 work Jun 01 17:46:08 yeah and bluetooth headset is also another nice thing Jun 01 17:46:16 but I never really looked at bluetooth integration Jun 01 17:46:25 (I don't have bassfix) Jun 01 17:46:32 or tried running bluetooth on the gta02/gta04 Jun 01 17:46:42 I run bluetooth all the time Jun 01 17:46:48 so if anybody has ideas how we can get this done tell me Jun 01 17:46:53 so ask me Jun 01 17:47:06 also as pespin Jun 01 17:47:13 he wrote emtooth2 Jun 01 17:47:30 I also somewhat documented it in the hardware part of the fso wiki Jun 01 17:47:42 if you want I can document it more Jun 01 17:47:56 morphis: the code in ophoned to handle bt headset was good enough. Jun 01 17:48:15 GNUtoo-desktop: that would be really nice Jun 01 17:48:20 ok Jun 01 17:48:31 morphis, do you know what A2DP and SCO are? Jun 01 17:49:27 morphis, http://wiki.freesmartphone.org/index.php/Implementations/fsoaudiod/RoutingUsecases#Openmoko_GTA02 Jun 01 17:49:36 GNUtoo-desktop: bluetooth protocols to provide audio data Jun 01 17:49:38 it's aalready well documented Jun 01 17:49:41 yes Jun 01 17:49:50 but there A2DP and SCO are different Jun 01 17:49:59 A2DP is stereo. no mic Jun 01 17:50:08 SCO is mono, with mic Jun 01 17:50:22 ok Jun 01 17:50:30 A2DP on the freerunner is routed trough software(alsa) Jun 01 17:50:46 btw. on the wiki page is pulseaudio mentioned again as unwanted; why everybody hates pulseaudio? Jun 01 17:50:47 SCO is routed trough hardware(the sound card controls) Jun 01 17:50:54 I think it's routed through HCI (host controller interface of the bluetooth chip) everywhere Jun 01 17:50:57 GNUtoo-desktop: good to know Jun 01 17:51:11 morphis, pulseaudio is unusable for gta02 Jun 01 17:51:16 we would have to drop gta02 Jun 01 17:51:19 SCO can be routed via HCI too but the chip should support it. Jun 01 17:51:28 and that's the worse thing to do Jun 01 17:51:35 it's the only somewhat working phone we have Jun 01 17:51:36 In gta02 the chip is configured to route SCO over a dedicated direct PCM link to the Wolfson mixer. Jun 01 17:51:37 GNUtoo-desktop: who said this? om back in 2007? Jun 01 17:51:52 morphis, no, openembedded people Jun 01 17:51:57 GNUtoo-desktop: I don't want to drop anything Jun 01 17:52:02 GNUtoo-desktop: really? Jun 01 17:52:08 morphis: talking about day-to-day usage, gta02 seems to be the best phone out there. Jun 01 17:52:09 because the armv4 don't have certain locking instructions Jun 01 17:52:13 so they have to be emulated Jun 01 17:52:16 ah ok Jun 01 17:52:25 and pulse make use of that Jun 01 17:52:36 s/locking/atomic/ Jun 01 17:53:18 In fact it's possible to reconfigure the EEPROM of the BT chip inside gta02 to route SCO over HCI, but it's non-trivial (non-standard) and doesn't make much sense, because it works fine for our purposes with routing over the dedicated PCM link too. Jun 01 17:54:36 SCO over PCM doesn't need any special alsa driver, only bluetoothd running. The app must set the proper rate and other parameters for the second interface of the wolfson mixer though, and it's done nicely in ophoned code. Jun 01 17:55:07 And then the app just issues a Play() (iirc) call to the bluetoothd, and SCO is activated. Jun 01 17:55:43 PaulFertser, it does need an alsa plugin Jun 01 17:55:45 sounds very easy Jun 01 17:55:54 opkg install libasound-module-bluez Jun 01 17:56:12 but maybe there is more than one waay Jun 01 17:56:21 my way is with the opkg install and the asoundrc Jun 01 17:56:33 ahhh sorry Jun 01 17:56:40 I was talking about A2DP Jun 01 17:56:47 you're talking abuot SCO Jun 01 17:56:49 sorry Jun 01 17:56:52 forget what I said Jun 01 18:00:52 GNUtoo-desktop: A2DP does need an alsa plugin. But SCO over PCM doesn't. Jun 01 18:01:05 ok Jun 01 18:01:05 morphis: http://trac.freesmartphone.org/ticket/685 i hope that's "good enough" for now. Jun 01 18:02:00 GNUtoo-desktop: yeah, A2DP seems to be working fine as is (except that it seems impossible to activate dmix for it :(((, but that's a bluez problem). The issue is that the BT headset for gsm calls is currently unsupported by FSO. Jun 01 18:02:29 GNUtoo-desktop: hello, did you have time to fix compat-wireless build ? I Jun 01 18:03:22 nschle85, not yet Jun 01 18:03:32 I'll look at it very soon Jun 01 18:04:09 PaulFertser: hello i did not manage to build compat wireless so i tried using bitbake to compile but there is some issue (no driver included in the packga :-)) Jun 01 18:05:05 nschle85: heh Jun 01 18:05:20 PaulFertser: it should, thanks a lot Jun 01 18:08:04 GNUtoo-desktop: may i also try wpa_supplicant 1.0 it was released some days ago, or do you expect problems for our supported phones ? Jun 01 18:08:35 I've no idea about possibles problem, sorry Jun 01 18:10:11 GNUtoo-desktop: do you know how i can copy the kernel sources to the phone to compile native kernel modules ? Jun 01 18:11:00 i am failing compiling compat-wireless on N900 i think complete source s are nneded only headers are not enough Jun 01 18:11:14 forget about that Jun 01 18:11:21 you nearly got it to compile unnder fso Jun 01 18:11:23 oops Jun 01 18:11:25 you nearly got it to compile unnder oe Jun 01 18:11:43 you just need to adapt the kernel I think Jun 01 18:11:51 and that won't change if you do it on target Jun 01 18:12:45 Oh, what was the exact issue? Jun 01 18:13:07 PaulFertser: relating to what ? Jun 01 18:13:30 that: Jun 01 18:13:38 nschle85: to your problem compiling it with OE Jun 01 18:13:40 CONFIG_WL12XX_PLATFORM_DATA Jun 01 18:13:47 it's not an oe problem indeed Jun 01 18:13:52 that's what I'm telling him Jun 01 18:14:06 it's a compat-wireless<->kernel sources problem Jun 01 18:14:14 I'll try Jun 01 18:14:23 I did it for htcdream which has .32 already Jun 01 18:14:57 GNUtoo-desktop: but N900 kernel has already a WL1251 driver Jun 01 18:15:15 that's not a problem Jun 01 18:15:24 you want a new driver Jun 01 18:15:31 that's why you want compat wireless Jun 01 18:15:32 right? Jun 01 18:15:47 yes Jun 01 18:16:10 PaulFertser: actually it already has an ipv6 address since last week Jun 01 18:16:40 PaulFertser: it just needs to be restarted so that it rebinds to the socket. and tested. and documented, and firewall rules adjusted :) Jun 01 18:17:23 lindi-: wow, cool :) unfortunately, dyndns sucks too much to be able to add an AAAA record, or do not they anymore? Jun 01 18:17:28 but where comes CONFIG_WL12XX_PLATFORM_DATA into play ? Jun 01 18:17:55 Ah, platform data... Jun 01 18:19:15 PaulFertser: please explain, its years ago i compiled a linux kernel (it was for old XBOX i think :- ) Jun 01 18:19:50 nschle85, it's the platform data passed from the board file to the driver Jun 01 18:19:58 nschle85, the board file is in kernel source Jun 01 18:20:07 and the driver is in compat-wireless Jun 01 18:20:58 nschle85: what is the error you're getting? Jun 01 18:21:29 PaulFertser: no error, but the wl1251.ko is not included in the ipk Jun 01 18:21:29 none Jun 01 18:21:39 just that the module is not built Jun 01 18:21:41 let me fix it Jun 01 18:21:56 GNUtoo-desktop: thank you Jun 01 18:22:31 thanks me when it'll be done Jun 01 18:25:17 PaulFertser: GNUtoo-desktop: but which driver i have to build exactly ? wl1251 or wl12xx ? Jun 01 18:26:11 nschle85: if you're talking about driver-select script then probably wl1251 should be choosen, but often the safest way is to build all of the compat-wireless. Jun 01 18:27:17 PaulFertser: ok the i tried the right one Jun 01 18:28:19 PaulFertser: now ill try a newer wpa_supplicant to reproduce the problem Jun 01 18:29:05 1251_spi Jun 01 18:30:08 GNUtoo-desktop: 1251_spi as argument for driver_select ? Jun 01 18:30:12 no Jun 01 18:30:17 wl1251 is Jun 01 18:30:48 so it will result in wl1251.ko and wl1251_spi.ko ? Jun 01 18:34:34 and wl1251_sdio.ko Jun 01 18:38:36 ok kernel built manually, let me build compat wireless now Jun 01 18:45:02 nschle85, what is the compat wireless that you want? Jun 01 18:45:29 the version or what do you mean ? Jun 01 18:46:21 in the recipe i added the latest available Jun 01 18:48:09 ok Jun 01 18:48:14 the one for today is ok? Jun 01 18:49:23 i guess so :-) Jun 01 19:05:10 PaulFertser: when GNUtoo-desktop: finished fixing the build how do i provide the logfiles you need ? Jun 01 19:05:30 one second Jun 01 19:05:43 there is indeed pdata: Jun 01 19:05:56 arch/arm/mach-omap2/board-rx51-peripherals.c Jun 01 19:05:58 nschle85: pastebin and pass the link here :) Jun 01 19:06:10 all I have to do is to sync them and add the option Jun 01 19:06:38 PaulFertser: no, how do i enable debug messages ? Jun 01 19:07:08 PaulFertser: it was russian logic ? Jun 01 19:07:35 nschle85: have you already reproduced your problem with the latest compat-wireless? ;) Jun 01 19:08:14 PaulFertser: no but i wanted to ask now because we have 2 hours difference Jun 01 19:08:43 nschle85: ok Jun 01 19:10:35 nschle85: you can try to alter #define DEBUG_LEVEL (DEBUG_NONE) to DEBUG_ALL in drivers/net/wireless/wl1251/wl1251.h Jun 01 19:10:36 nschle85, also cryptographic stuff often require to have the correct hour/date setup Jun 01 19:10:51 yoh Jun 01 19:12:02 GNUtoo-desktop: i agree but changing wpa+wpa2 to wpa2 in router "solved" the problem so time settings are relevant here ? Jun 01 19:12:06 ./drivers/net/wireless/wl12xx/Kconfig:config WL12XX_PLATFORM_DATA Jun 01 19:12:18 nschle85, it was just a remark Jun 01 19:15:35 PaulFertser: change in compat-wireless sources ? and where is the log stored ? in dmesg ? Jun 01 19:15:48 nschle85: yes, yes Jun 01 19:16:12 nschle85: also, if you run a nice enough system (with klogd or some similar software) you'll have it logged in syslog as well. Jun 01 19:16:26 hmmm Jun 01 19:16:32 config WL12XX_PLATFORM_DATA Jun 01 19:16:37 depends on WL1271_SDIO != n || WL1251_SDIO != n Jun 01 19:16:38 nschle85: to see it in dmesg you might need to use "dmesg -n8" as those messages are all KERN_DEBUG Jun 01 19:17:09 nschle85, what do you prefer? working arround in the kenrel or in compat-wireless? Jun 01 19:17:10 PaulFertser: ok Jun 01 19:17:57 GNUtoo-desktop: i am a software architect but i do not know anything about the software Jun 01 19:18:36 GNUtoo-desktop: for me it would be simplier to have it in compat-wireless but i dont know what is right Jun 01 19:19:32 GNUtoo-desktop: so why working around ? what would be the solution ? Jun 01 19:20:50 just patch compat wireless to remove the check Jun 01 19:20:56 and look if it works Jun 01 19:21:46 remove that: Jun 01 19:21:48 ifdef CONFIG_WL12XX_PLATFORM_DATA Jun 01 19:21:52 endif #CONFIG_WL12XX_PLATFORM_DATA Jun 01 19:21:56 but not what's in between Jun 01 19:22:07 in config.mk Jun 01 19:24:35 GNUtoo-desktop: ill do Jun 01 19:37:58 PaulFertser: how do i create a patch for a diff between 2 directories which are not under any source control ? Jun 01 19:38:29 nschle85: diff -ru Jun 01 19:39:25 what about diff -Nurd ? Jun 01 19:39:44 is there a diff -Nerd btw? Jun 01 19:41:58 http://pastebin.com/cJ4AV2TS Jun 01 19:42:29 but which of them is a correct patch ? Jun 01 19:43:54 nschle85, yes Jun 01 19:44:34 nschle85, the diff -Nerd was a joke Jun 01 19:44:44 I don't see the difference between the 2 patches Jun 01 19:44:50 both seem correct Jun 01 19:45:01 GNUtoo-desktop: me too but would they work ? Jun 01 19:45:14 is the base directory correct ? Jun 01 19:45:28 try them Jun 01 19:45:36 I use quilt for making patches Jun 01 19:45:39 but you can do that else: Jun 01 19:45:49 bitbake -c configure Jun 01 19:46:03 and you do the patch manually Jun 01 19:46:09 and then you bitbake normally Jun 01 19:47:12 ahh i think quilt it the right tool Jun 01 20:03:15 GNUtoo-desktop: i used quilt but i cannot find the patch Jun 01 20:03:56 you forgott to do that before: Jun 01 20:03:56 mkdir patches Jun 01 20:03:57 or patch Jun 01 20:04:01 I don't remember Jun 01 20:04:08 maybe ls -la Jun 01 20:04:13 there should be an old directory Jun 01 20:04:17 liike .pc Jun 01 20:04:22 or soemthing like that Jun 01 20:05:26 GNUtoo-desktop: the directory patches already existed , and there is my patch :-) Jun 01 20:05:47 ok Jun 01 20:16:20 probably very old issue ...I found how to send +CLVL=255 , tried ti calypso dsp mode 0001 (none)/ 0105 (nr) / 0187 (aec+nr) with microphone muted on the gta02 side ... the received volume on the gta02 side is higher than it ever was (aec+nr ... none dsp mode was way lower I cannot tell why) Jun 01 20:16:36 but still no way I would ever be able to hear a thing in a train Jun 01 20:17:02 ie in a flat I have to keep the ear close to the earpiece Jun 01 20:20:30 prahal: i think antrik: was telling something about louder speaker during FSOSHRCON11 so you should contact him to tell about your current development Jun 01 20:20:31 I have also tweaked all mixer settings http://prahal.homelinux.net/~prahal/phone/shr/gsmhandset.new.5 is the best I achieve reception wise Jun 01 20:22:17 PaulFertser: dynsdns support AAAA nowadays (I don't know if a paying account is required though) Jun 01 20:24:37 I wondered if my speaker issue was not a gsm firmware issue (ie CLVL bringing things too low) Jun 01 20:25:00 prahal: probably only for paying customers, yes. But luckily HE's DNS service is free and awesome, dyndns can't compete with it :) Jun 01 20:25:05