sicelo | maybe leste should look at using CONFIG_LEDS_TRIGGER_PATTERN then use mce or something else to provide the led patterns | 12:36 |
---|---|---|
Wizzup | does this support phones like the n900 with dedicated hw? | 12:37 |
Wizzup | or do you mean just moving some of it to the kernel for non-specialised hw? | 12:37 |
sicelo | that looks like it supports nearly all leds | 12:38 |
sicelo | i guess others may feel doing it in hw leads to device-specific hacks | 12:40 |
Wizzup | supporting nearly all leds doesn't mean it supports the n900 specific led driver, right? | 12:41 |
Wizzup | the one that does all the patterns without hw wakeups | 12:41 |
sicelo | yes it doesn't. | 12:41 |
sicelo | but i am not sure ... you want to keep n900's hw setup? | 12:42 |
Wizzup | we support it in mce | 12:42 |
Wizzup | and it definitely saves power, no? | 12:43 |
sicelo | oh ok. i am not averse to the n900 interface. i think it's cool actually | 12:44 |
sicelo | just thought this could help those devices which don't have such functionality, e.g. pp | 12:44 |
Wizzup | right | 12:45 |
Wizzup | so will it have less wakeups or something, or? | 12:45 |
sicelo | no idea tbh :-) | 12:45 |
Wizzup | I'm trying to see how it helps, other than perhaps less context switches for mce | 12:45 |
Wizzup | since we already have a generic led sys interface | 12:45 |
sicelo | you tell sysfs the pattern you want and kernel handles it by itself | 12:47 |
sicelo | i know d4's cpcap can d patterns in hw too, but seems the function isn't currently exposed in the driver | 12:51 |
uvos__ | so cpcap can do patterns in hw (well in fw android dose it in fw) | 13:06 |
uvos__ | the problem with the pattern trigger is it creats a lot of wakeups when used without any hw support, mces implementation wakesup alot les at the expense of being able to do alot less | 13:08 |
uvos__ | so this dosent really help us, if the n900's hw accel works via pattern trigger thats cool | 13:08 |
uvos__ | but theres no real incentive to replaceing mces code that uses the hw here | 13:09 |
Wizzup | uvos: imho we could have closed that ticket, but fine to keep it open if you want :p | 14:26 |
uvos__ | well as you mentioned bluetooth dosent work | 14:44 |
uvos__ | also that bug contains plenty of usefull information for a proper solution | 14:44 |
uvos__ | maybe we can rename it | 14:44 |
Wizzup | fine by me | 14:47 |
sicelo | ok. so how are we planning on using the leds for notifications in leste, on devices other than N900? | 14:54 |
sicelo | e.g. droid 4 | 14:56 |
Wizzup | don't we have this working already? or? | 14:57 |
sicelo | no idea. never seen led doing anything useful on my d4, except steady green when charging, and steady white when powering off | 14:58 |
sicelo | anyway, it's not very important to me atm ... i just came across the kernel PATTERNS while looking at feedbackd source | 15:07 |
sicelo | Wizzup: you said to ping you regarding our HSI/ISI issue. hope you'll have some spare cycles maybe weekend | 15:07 |
uvos__ | the leds work fine on d4 | 15:15 |
uvos__ | works with modest | 15:15 |
uvos__ | otherwise we just have nothing that activates any patterns | 15:15 |
uvos__ | same is true for all devices | 15:16 |
BlagovestPetrov[ | Wizzup: https://bpa.st/QGWQ I have ignorred the warning with the flags but it warns about possible buffer overflows | 15:17 |
BlagovestPetrov[ | libicd-wireguard | 15:18 |
BlagovestPetrov[ | https://github.com/maemo-leste/libicd-wireguard/pull/1 | 15:32 |
Wizzup | BlagovestPetrov[: let me see if I can fix the code | 15:50 |
Wizzup | btw the changelog entries lacks some spaces I think | 15:50 |
BlagovestPetrov[ | right, empty line | 15:53 |
BlagovestPetrov[ | fixed | 15:54 |
BlagovestPetrov[ | I don't see calendar-ui-widgets in GitHub | 15:54 |
Wizzup | freemangordon: I'm almost done trying to port eds-backend-telepathy, but will need you to review it | 16:01 |
BlagovestPetrov[ | tinymail is also failing because of multiple declarations. I'm trying to fix it.. | 16:02 |
Wizzup | BlagovestPetrov[: no, I cannot build calendar-ui-widgets because it all depends on eds-backend-telepathy | 16:03 |
Wizzup | anything that uses abook cannot be built | 16:03 |
Wizzup | which is most of the remaining packages | 16:03 |
BlagovestPetrov[ | but I don't even see it as a repository | 16:04 |
BlagovestPetrov[ | it may be part of another package | 16:04 |
BlagovestPetrov[ | it's not even in beowulf | 16:05 |
freemangordon | BlagovestPetrov[: hmm? https://github.com/maemo-leste/calendar-ui-widgets | 16:08 |
BlagovestPetrov[ | weird | 16:09 |
freemangordon | see what apt-get source calendar-ui-widgets in beowulf will tell you :p | 16:09 |
BlagovestPetrov[ | I tried only with apt-cache :) | 16:10 |
* BlagovestPetrov[ uploaded an image: (130KiB) < https://libera.ems.host/_matrix/media/v3/download/matrix.petrovs.info/qjQOqLoNJCCemZjoZLsOJYYd/Screenshot_20221202_170936.png > | 16:10 | |
freemangordon | Wizzup: ok, will try | 16:10 |
freemangordon | well, M$ | 16:10 |
BlagovestPetrov[ | :D | 16:11 |
uvos__ | the search is toaly uesless | 16:12 |
freemangordon | no idea why it would not be visible/searchable | 16:12 |
uvos__ | it dosent even find "sdl" in upstream forks | 16:12 |
freemangordon | uvos__: usually it works | 16:12 |
uvos__ | altho we have libsdl1.2 and libSDL2 as repos | 16:12 |
uvos__ | freemangordon: no this has never worked | 16:12 |
uvos__ | it randomly dosent work for certain querys | 16:13 |
uvos__ | and for those querys it never has or will | 16:13 |
freemangordon | ok | 16:13 |
uvos__ | this has also been reported to gh often (at least you can find posts all over) | 16:14 |
uvos__ | with no improvement | 16:14 |
uvos__ | so yeah M$ | 16:14 |
BlagovestPetrov[ | windows update on one of the data nodes.. | 16:16 |
uvos__ | https://github.com/maemo-leste-upstream-forks?q=sdl&type=all&language=&sort= | 16:16 |
uvos__ | there for an example | 16:16 |
uvos__ | the same problem also exits when searching though code on gh | 16:16 |
uvos__ | so this causes lots of false conculstions | 16:16 |
uvos__ | like "this function is never used" etc | 16:17 |
uvos__ | its just horrible | 16:17 |
freemangordon | BlagovestPetrov[: Wizzup: WTF is actually this https://github.com/maemo-leste/libicd-wireguard/blob/master/src/libicd_wireguard_config.c#L161 ? | 16:19 |
freemangordon | gcc is absolutely right | 16:19 |
freemangordon | why is strncat used in first place ? | 16:20 |
freemangordon | omg https://github.com/maemo-leste/libicd-wireguard/blob/master/src/libicd_wireguard_config.c#L119 | 16:20 |
uvos__ | freemangordons head explodeing | 16:21 |
* freemangordon gets stack overflow | 16:21 | |
uvos__ | yeah this function hat alot to gain from using g_str_* functions :P | 16:22 |
freemangordon | BlagovestPetrov[: do you have an idea how to fix that properly? and if you have, do you want to? | 16:22 |
freemangordon | especially g_str_concat ;) | 16:22 |
BlagovestPetrov[ | I'm not sure | 16:22 |
freemangordon | ok, I will fix that | 16:23 |
Wizzup | freemangordon: just a sec, I made it compile, will push momentarily | 16:26 |
Wizzup | freemangordon: looks like parazyd wrote that https://github.com/maemo-leste/libicd-wireguard/commit/2f00166d4947763d384b0d039b7b73ec7ea4e452 | 16:27 |
freemangordon | yeah | 16:29 |
uvos__ | on the bright side, its probubly very fast this way compeard to using g_str_* :P | 16:31 |
uvos__ | way less mallocs | 16:31 |
Wizzup | freemangordon: https://github.com/maemo-leste/eds-backend-telepathy/tree/porting-wip | 16:31 |
Wizzup | freemangordon: the main thing that needs a look I think is 'status' handling when it's actually not an error | 16:33 |
Wizzup | there is no 'new variant' of E_DATA_BOOK_STATUS_SUCCESS | 16:33 |
Wizzup | which might mean the respond functions shouldn't actually use e_client_error_create at all, and rather perhaps pass NULL ? | 16:33 |
Wizzup | I suppose the old e_data_book_create_error function returned NULL if status was E_DATA_BOOK_STATUS_SUCCESS | 16:34 |
Wizzup | looks like it | 16:35 |
Wizzup | e-book-backend-ldap.c does this with ldap_error_to_response | 16:35 |
freemangordon | uvos__: right... until you get buffer overflow :) | 16:41 |
freemangordon | Wizzup: just a minute to push WG fix | 16:41 |
Wizzup | freemangordon: I think what I said is above is correct and the status handling needs some additional ctx | 16:43 |
freemangordon | ugh https://github.com/maemo-leste/libicd-wireguard/blob/master/src/libicd_wireguard_config.c#L208 | 16:44 |
freemangordon | I think this whole code needs review | 16:44 |
Wizzup | ok, well, I need to go and arrange some dinner for myself, so I'll be back in an hour or more | 16:44 |
Wizzup | I can rework the status handling as above, but please look if you see other problems | 16:45 |
Wizzup | bbl | 16:45 |
freemangordon | ok | 16:45 |
freemangordon | BlagovestPetrov[: could you pull https://github.com/maemo-leste/libicd-wireguard master and try to compile it? If it is ok, please, pull in/release for beowulf-devel as well | 16:50 |
Wizzup | (he doesn't have privs to release or push branches atm) | 16:51 |
Wizzup | (I've been doing it) | 16:51 |
freemangordon | ah, ok | 16:54 |
freemangordon | Wizzup: E_CLIENT_ERROR_INVALID_ARG = 0 | 16:54 |
freemangordon | so setting status to 0 is not ok, IIUC | 16:54 |
BlagovestPetrov[ | it's ok in master | 16:54 |
BlagovestPetrov[ | tested.. | 16:54 |
freemangordon | great | 16:55 |
freemangordon | Wizzup: this https://github.com/maemo-leste/eds-backend-telepathy/commit/842e6af2b06ae8c55fdfeeffcd5f4bdb09d069c9#diff-e9311ef775aa11f0d335f34653a990095eba4bbf0ef0e7434e0acdd0648939ecR2576 | 16:55 |
freemangordon | so yeah, it seems e_data_book_respond_remove_contacts() (and others) shall be called with NULL error in case of success | 17:00 |
Wizzup | sicelo: btw yes I should have time tonight | 17:05 |
Wizzup | freemangordon: ok, will rewrite the logic then | 17:05 |
Guest224 | Is hildon-connectivity-mobile inside in any _stable_ SD-card images yet? | 17:14 |
uvos__ | no | 17:17 |
sicelo | Guest224: what device are you using | 17:18 |
sicelo | Wizzup: great. looking forward | 17:18 |
Guest224 | pinephone | 17:18 |
Guest224 | I also have N900..and might soon Droid4. | 17:18 |
sicelo | the full package :-) | 17:20 |
Guest224 | but now running PostmarketOS in pinephone, cause it works on the road, I hate PMOS, but it has 4G and when Leste has working connection in _stable_ then I change permanently to it. | 17:24 |
sicelo | i don't think you should *hate* pmOS ... it's fine to not like it though | 17:26 |
sicelo | why do you feel you need 'stable' specifically? | 17:26 |
nerdcore | i quite like pmOS, but it's been mashing my SMS messages together, and mislabeling them :( | 17:27 |
nerdcore | gonna try ML instead I think | 17:27 |
nerdcore | I see that there are ML instructions on building a custom image, but they do not explain how to replace the default user/pass... Is that easily done? | 17:27 |
Guest224 | sicelo: ok..hate might too strong word...but gui is horrible...when it is stable then it is "stable enough" for me, for beta testing :) | 17:29 |
sicelo | nerdcore: it's not pmOS mashing your messages ... most probably modem-manager. actually, to a large extent, pmOS doesn't make software | 17:29 |
sicelo | Guest224: you may want to consider devel, if you reall want to help with testing ;-) | 17:30 |
Wizzup | chimaera will have modem stuff by default | 17:31 |
sicelo | Guest224: pmOS doesn't have a gui ... which gui do you find horrible? phosh, plasma mobile, or sxmo? | 17:32 |
uvos__ | hell for a while it had hildon-desktop | 17:32 |
nerdcore | I like phosh because elsewhere I run GNOME, so it feels very comfortable :) | 17:32 |
nerdcore | sicelo: appreciate the input. It's a PinePhone (not Pro) device, so do you expect I may have similar issues in ML? | 17:33 |
nerdcore | I tried a couple different firmwares, to varying results. | 17:33 |
Guest224 | sicelo: thats true...I have tested phosh and plasma...there was some reason I didn't test sxmo (I dont remeber why..) | 17:33 |
nerdcore | i3/sxmo has a steep learning curve ;) | 17:33 |
sicelo | nerdcore: i don't know. ML uses ofono. maybe the message handling is different there ... not really sure. as for which distro to choose up to you. we're in ML chat so choose ML :-p | 17:34 |
nerdcore | yeah I've only tried 2-3 OS on the device so far. I know buZz from elsewhere so ML seems like a good choice ;) | 17:35 |
sicelo | great then | 17:35 |
nerdcore | its so easy to drop a new OS onto microSD, and leave the internal storage untouched. Great for testing new OSes | 17:35 |
nerdcore | will probably try the custom image builder too, for max authority heh | 17:36 |
sicelo | of course ML's DE is the lightest mobile DE, so I guess it really is the better choice. i read pinephone cpu isn't as beefy as people would like | 17:36 |
nerdcore | yeah, defoz chunky running phosh | 17:36 |
nerdcore | I mean, when I click the phone icon in the Messages application to call that person, it takes around 60s to launch the Phone application and then dial the number :( | 17:37 |
nerdcore | but it works ... eventually lol | 17:37 |
Wizzup | sicelo: no ncpu is ever as beefy as people like | 17:37 |
buZz | sicelo: i think pp nonpro has quite a slow gpu specifically | 17:37 |
buZz | the panfrost/lima stuff helps a lot though | 17:38 |
Wizzup | bottom line is that low footprint matters | 17:38 |
Guest224 | nerdcore: I tested ML shortly in september and it is fast in Pinephone...only reason why I moved to PmOS was that mobile-connectivity wasn't ready to dayly use. | 17:38 |
nerdcore | does ML use glibc? pmOS is using MUSL C and giving me some specific issues as well | 17:38 |
Wizzup | yespglibc | 17:38 |
Wizzup | it,s just debian/devun | 17:39 |
nerdcore | Guest224: "mobile connectivity" such as sharing your 4G with a laptop? | 17:39 |
buZz | Guest224: i've been daily'ing ML on d4 with data since ~april | 17:39 |
Wizzup | we're porting to bullseye atm | 17:39 |
nerdcore | excellent. I have some time in the coming week, so keen to try ML. I'll stick around. :) | 17:39 |
Guest224 | nerdcore: plain 4G connection. | 17:39 |
buZz | yay :) | 17:39 |
freemangordon | still, mobile connectivity is WIP | 17:39 |
nerdcore | hmm, I do use the mobile data. I turned my home wifi Off last year ;) | 17:40 |
buZz | the whole thing is WIP :P | 17:40 |
freemangordon | connectivity as such works | 17:40 |
nerdcore | does /usr/bin/ssh work? | 17:40 |
freemangordon | but, UI for changing PIN for example does not | 17:40 |
freemangordon | or, manualy selecting operator | 17:40 |
freemangordon | etc | 17:40 |
nerdcore | because I can just run lynx on a VPS haha | 17:41 |
freemangordon | chromium works pretty well | 17:41 |
buZz | yeah, fast too | 17:41 |
freemangordon | but, you have to connect a keyboard | 17:41 |
buZz | and nice touch support | 17:41 |
freemangordon | otherwise it is very hard to enter the url ;) | 17:41 |
nerdcore | Is it possible to import a CSV or VCF file into the contacts application in ML? That would save me a lot of setup time | 17:42 |
buZz | maybe get a d4 to remote control the pp? | 17:42 |
buZz | hmm | 17:42 |
Guest224 | nercore: I can recomend Pinephone keyboard for Pinephone. | 17:42 |
freemangordon | nerdcore: vcf is ok | 17:42 |
* buZz suddenly sees a cyberdeck made of cellphones running leste | 17:42 | |
nerdcore | sometimes I use my PP via its serial UART in the 3.5mm headphone jack. I connect it to a rpi and run minicom | 17:42 |
freemangordon | actually it has UI to import a directory containing vcfs | 17:42 |
freemangordon | so you import all contacts at once | 17:43 |
nerdcore | excellent. Thanks freemangordon that will save me a lot of time. I had to type all my contacts by hand into pmOS :( | 17:43 |
buZz | the pp i borrowed had to go back :( | 17:43 |
nerdcore | i have a USB-C hub here with all sorts of connections like RJ-45 ethernet, and HDMI. Makes it possible to hook my PP up like a PC, but the battery dies in < 2h | 17:44 |
buZz | still doubting any new device, d4 is great but limited, pp same, pocophone f1 looks cool in support but no hw keyb | 17:44 |
buZz | specs of poco look good too | 17:44 |
nerdcore | i have a small BT keyboard here, but I find it equally as frustrating as OSK | 17:44 |
Guest224 | freemangordon: Is that vcfs importer tool on default SD-image, it is something that should? | 17:44 |
sicelo | buZz: yes the sdm845 kernel support is maturing fast | 17:44 |
freemangordon | it is part of the addressbook | 17:45 |
freemangordon | not a separate tool | 17:45 |
buZz | sicelo: do you think that device will get some futureproofing through it? | 17:45 |
sicelo | meaning? | 17:45 |
buZz | will-it-mainline | 17:45 |
buZz | hehe | 17:45 |
buZz | like how n900 can nowadays boot straigthup debian installers | 17:46 |
Guest224 | freemangordon: that is good! | 17:46 |
sicelo | mainline? I was talking about sdm845 mainline :-) | 17:46 |
buZz | hehe ok | 17:46 |
buZz | i have good hopes, but really needs some keyb, touchscreen input makes me feel demented, lol. | 17:46 |
Wizzup | nerdcore: you can also use syncevolution for contacts | 17:46 |
buZz | ML + selfhosted radicale = syncable agenda and contacts to my desktop and laptops | 17:47 |
buZz | 'cloudfree' | 17:47 |
Guest224 | Cloudfree sounds always good. | 17:48 |
Guest224 | ad text:"With Maemo Leste you can see the sun, because you don't have to stick in clouds." | 17:49 |
buZz | Guest224: we need more memes | 17:49 |
buZz | 'de-mist your phonelife' | 17:50 |
Guest224 | or just "cloudfree phonelife" | 17:51 |
buZz | 'get ready to uncloud, maemo leste is coming' | 17:51 |
buZz | haha yeah and then we can post pictures of ppl with stuff like 'my cellphone is a minecraft server' and some blurb | 17:52 |
Guest224 | "You can use Maemo Leste in offline, when you want." | 17:58 |
Wizzup | freemangordon: please re-check porting-wip | 17:59 |
Wizzup | freemangordon: merged it to master | 18:41 |
freemangordon | Wizzup: sorry, was having dinner | 18:45 |
Wizzup | np | 18:46 |
Wizzup | the only remaining question was about E_DATA_BOOK_STATUS_BOOK_REMOVED | 18:46 |
Wizzup | I quote: | 18:46 |
Wizzup | I wasn't sure how to map E_DATA_BOOK_STATUS_BOOK_REMOVED - I went for | 18:46 |
Wizzup | E_BOOK_CLIENT_ERROR_NO_SUCH_BOOK for now | 18:46 |
* freemangordon checks if there is more appropriate status | 18:47 | |
tmlind | uvos: got output on xt910 lcd with some copy paste android code :) | 18:48 |
tmlind | so the backlight controller is integrated into the panel | 18:49 |
tmlind | and the panel init sequence is a bit crazy set of mip dcs writes.. | 18:50 |
Wizzup | tmlind: sweet :) | 18:50 |
_uvos_ | tmlind: sweet :) | 18:51 |
_uvos_ | maybe this is true on d3 aswell | 18:51 |
Wizzup | _uvos_: echo | 18:51 |
_uvos_ | i wasent able to find who dose the backlight there | 18:51 |
Guest224 | nerdcore: btw..if you are frustrating battery life with USB-C devices then you should think Pinephone Keyboard it gives extra battery life (it has 6000mAh battery inside)..but you cannot charge when useing USC-C devices or at least this page saying it: https://talk.maemo.org/showpost.php?p=1574678&postcount=25 | 18:51 |
tmlind | heh now need to sort of understand what is going on.. it's the panel-mapphone.c dsi_mipi_430_cm_540_960_amoled_panel_enable | 18:51 |
_uvos_ | schould just look in android kernel | 18:51 |
Wizzup | Guest224: yeah I found out the hard way and broke mine | 18:52 |
nerdcore | ty | 18:53 |
Guest224 | Wizzup: :( ..it should have big warning manual. | 18:53 |
freemangordon | Wizzup: hmm, is mixing EClientError with EBookClientError ok? | 18:54 |
tmlind | _uvos_: well bunch of the mipi commands are now implemented in a generic way, see "MIPI DCS commands" in include/video/mipi_display.h | 18:54 |
tmlind | _uvos_: maybe check the panel id for d3? | 18:55 |
tmlind | xt910 has panel_id=0x90006, so you can grep for 0x00090006 in android panel-mapphone.c | 18:56 |
freemangordon | Wizzup: like, e_client_error_create() wants EClientError, but you do https://github.com/maemo-leste/eds-backend-telepathy/blob/master/src/e-book-backend-tp.c#L3141 | 18:57 |
Wizzup | freemangordon: good question, probably not... | 18:57 |
tmlind | i added motorola-mapphone-xt875-xt984-common.dtsi and moved the lcd regulator and backlight there | 18:57 |
freemangordon | Wizzup: which repo is EDS? | 19:00 |
freemangordon | upstrem-forks? | 19:00 |
Wizzup | yes | 19:00 |
Wizzup | evolution-data-server | 19:00 |
freemangordon | not visible :D | 19:01 |
freemangordon | umm | 19:01 |
freemangordon | 404 | 19:01 |
Wizzup | https://github.com/maemo-leste/evolution-data-server/ | 19:01 |
Wizzup | maybe not upstream-forks then | 19:01 |
freemangordon | well, this is not 'upstream-forks' ;) | 19:01 |
Wizzup | yup, I thought it was there | 19:02 |
freemangordon | np | 19:02 |
Wizzup | you can move it and I can update the jenkins job | 19:02 |
freemangordon | not important | 19:04 |
freemangordon | but, see what ldap does | 19:04 |
freemangordon | https://github.com/maemo-leste/evolution-data-server/blob/e298c82798843470ae92cbedac236268e574518e/src/addressbook/backends/ldap/e-book-backend-ldap.c#L1130 | 19:04 |
Wizzup | yeah, I saw that basically after I ported the code | 19:04 |
freemangordon | ok :) | 19:04 |
freemangordon | if you want, I can fix it, but tomorrow | 19:05 |
Wizzup | ok, I'll see if I feel like it today | 19:05 |
Wizzup | rtcom-eventlogger-ui also needs changes :( | 19:05 |
Wizzup | gcc detects that one case of the switch was not covered | 19:06 |
freemangordon | I really like this gcc10, not a single false positive so far | 19:07 |
Wizzup | yes, it's good in that sense | 19:08 |
Wizzup | annoying otherwise :P | 19:08 |
freemangordon | well, yeah, but at least something we get help from upstream :) | 19:09 |
uvos | tmlind: xt984? | 19:10 |
uvos | this is typo in irc only i hope :P | 19:10 |
Wizzup | freemangordon: yup | 19:11 |
uvos | tmlind: yeah i dident look at the android source at all for d3 i just checked if changeing the cpcap backlight led channel changes the backlight (this was not used on any device, i wonder why) on d3 or if any of the lm* channels do and dident find it | 19:13 |
uvos | ill get back to it at some point | 19:13 |
tmlind | uvos: ok, yeah sorry xt894 typo :) | 19:16 |
Wizzup | so notify_notification_attach_to_widget seems to be done from libnotify | 19:35 |
Wizzup | commit 27e05d0f9562a26163493d6cc1d5924b9a4ebf68 | 19:36 |
Wizzup | Author: William Jon McCann <jmccann@redhat.com> | 19:36 |
Wizzup | Date: Fri Oct 8 22:47:06 2010 -0400 | 19:36 |
Wizzup | Remove the ability to attach notifications to widgets or positions | 19:36 |
Wizzup | well, I'll continue this later | 19:37 |
Wizzup | It looks like python2 stuff will be hard to keep going, there's a lot more to import now, cython, python-notify, etc | 19:41 |
Wizzup | sicelo: ok let me get the trace, then I really have to go | 19:51 |
sicelo | awesome. thanks | 20:07 |
Wizzup | sicelo: hrm, it's not showing kernel console atm :( | 20:08 |
Wizzup | sicelo: when does it reset for you | 20:08 |
sicelo | just before h-d shows up | 20:10 |
sicelo | which i assume is when modem is brought online | 20:10 |
sicelo | anyway, i did remove ofono from default runlevel, and then h-d does come up. so it's definitely modem related. did yours boot into h-d right away? | 20:12 |
Wizzup | no, it was slow and serial was on tty0 | 20:14 |
sicelo | ah. | 20:14 |
Wizzup | I need to redo it with serial on ttyS* | 20:14 |
Wizzup | but...tomorrow morning :) | 20:14 |
Wizzup | I'mm keeping people waiting | 20:14 |
sicelo | doesn't the kernel already have ttyS2? which kernel did you try | 20:15 |
sicelo | at /proc/cmdline | 20:15 |
sicelo | root=/dev/mmcblk0p2 rootwait console=ttyO2,115200 | 20:15 |
sicelo | root@devuan-n900:~# uname -r | 20:15 |
sicelo | 6.1.0-rc6 | 20:15 |
sicelo | that's the one from repos | 20:15 |
sicelo | but sure, tomorrow | 20:15 |
Wizzup | I think my uboot overrides it | 20:16 |
sicelo | ok. tomorrow then | 20:17 |
Wizzup | yup. :) | 20:17 |
sicelo | you don't even need to recompile kernel for this .. uboot has `setenv bootargs ....` | 20:17 |
Wizzup | I forgot how long it took for the n900 to upgrade :P | 20:17 |
Wizzup | sicelo: yeah but I'm already outside now | 20:18 |
sicelo | uvos: please restore CONFIG_RFKILL in kernel. this broke wifi on N900 | 20:21 |
uvos | sicelo: not sure what you mean by "restore" | 21:19 |
uvos | neither our 6.1 nor 5.18 set this | 21:20 |
uvos | see | 21:20 |
sicelo | "restore" means current kernel does not have the config. it was there before. | 21:20 |
uvos | https://github.com/maemo-leste/droid4-linux/blob/maemo-6.1/arch/arm/configs/omap2plus_defconfig | 21:20 |
uvos | https://github.com/maemo-leste/droid4-linux/blob/maemo-5.18.y/arch/arm/configs/omap2plus_defconfig | 21:20 |
uvos | this dosent apear true | 21:20 |
uvos | if wifi broke this is something new | 21:20 |
sicelo | fine | 21:21 |
uvos | so dose setting CONFIG_RFKILL to something help in 6.1? | 21:21 |
uvos | Wizzup: freemangordon: https://github.com/maemo-leste/osso-xterm/pull/4 | 23:17 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!