| Wizzup | freemangordon: ok to test now? | 00:06 |
|---|---|---|
| freemangordon | better not, I have to make a new version first | 00:07 |
| freemangordon | but, for some reason Scan stopped giving me results :( | 00:07 |
| Wizzup | ok | 00:07 |
| Wizzup | btw | 00:07 |
| Wizzup | http://wizzup.org/dirlist/droidfam/ | 00:07 |
| Wizzup | not sure if I should make better ones or if these are ok | 00:07 |
| Wizzup | the thought is just to capture all the droid4 and bionic phones in one place | 00:07 |
| freemangordon | omg | 00:08 |
| Wizzup | :) | 00:09 |
| Wizzup | I'll make one in July showing the "family", with n900, pinephone and others, all on | 00:09 |
| uvos | thats... a lot of mapphones | 00:09 |
| Wizzup | I actually have one or two more but w/e | 00:11 |
| uvos | my gf thinks im a crazy persion for having 3 :D | 00:12 |
| Wizzup | well if you need to put it in perspective, there's photos :p | 00:12 |
| uvos | haha i allready fwd it :P | 00:12 |
| parazyd | :D | 00:13 |
| freemangordon | Wizzup: new version is building ATM | 00:27 |
| Wizzup | cool | 00:27 |
| freemangordon | what I don;t get it why nokia code does not terminate connecting to hidden wlan if it cannot be found | 00:28 |
| freemangordon | do you see some value? | 00:28 |
| Wizzup | it might take a while to scan for it, but I see no point in not stopping | 00:29 |
| freemangordon | see https://github.com/maemo-leste/connui-internet/blob/master/src/settings/easy-wlan.c#L761 | 00:30 |
| freemangordon | it doesn;t make sense to me | 00:30 |
| freemangordon | Wizzup: you may upgrade and test | 00:36 |
| freemangordon | Wizzup: hmm, seems I should not rely on BSSAdded only, as once added it is never removed | 00:39 |
| freemangordon | so, if while scanning I don;t receive BSSAdded for the given hidden wlan, I shall iterate over all the BSSs on ScanStop. Is that correct? | 00:40 |
| Wizzup | BSSes should expire over time | 00:40 |
| freemangordon | but not immediately | 00:41 |
| Wizzup | Yes, iterate over them on scandone/scanstop | 00:41 |
| freemangordon | yeah, ScanDone | 00:41 |
| Wizzup | right, they can persist across scans | 00:41 |
| freemangordon | ok, so it doesn;t make sense to listen for BSSAdded | 00:41 |
| Wizzup | for icd2 it does, since you can show added networks in real time | 00:42 |
| freemangordon | as it will be fired only once (the first time I issue scan) | 00:42 |
| freemangordon | yeah (icd2) | 00:42 |
| freemangordon | but, you still can try to connect with the current code, it will succeed the first time you try | 00:43 |
| Wizzup | ok | 00:43 |
| Wizzup | freemangordon: best if I reboot, I guess | 00:45 |
| Wizzup | (rather than re-launch connui-conndlgs) | 00:45 |
| freemangordon | hmm, yeah | 00:45 |
| freemangordon | both should work | 00:45 |
| freemangordon | restart connui-conndlgs first | 00:45 |
| Wizzup | I just rebooted the device, more clean way | 00:46 |
| freemangordon | ok | 00:46 |
| freemangordon | Wizzup: is there a way to force wpa_supplicant to forget about that specific ssid? | 00:47 |
| Wizzup | about the result of the scan you mean? | 00:49 |
| freemangordon | yes | 00:49 |
| Wizzup | I see: | 00:49 |
| Wizzup | FlushBSS ( u : age ) –> nothing | 00:49 |
| Wizzup | Flush BSS entries from the cache. | 00:49 |
| freemangordon | because I don;t want to wait for scan to complete and then to iterate over all of the networks just to get the results for 'my' network | 00:49 |
| Wizzup | freemangordon: yeah worked (connecting) | 00:51 |
| Wizzup | it got stuck forever when I wrote the wrong ssid though | 00:51 |
| Wizzup | but I think you remarked on that earlier | 00:51 |
| freemangordon | mhm | 00:51 |
| freemangordon | I don;t understand why nokia did it like that | 00:52 |
| Wizzup | hm, now opening osso controlpanel causes it to close/crash | 00:52 |
| Wizzup | (could be something on my side) | 00:52 |
| Wizzup | well when I open the internet connections dialog | 00:52 |
| Wizzup | hm, worked when I did it again | 00:52 |
| Wizzup | ok, looks like it was nicely added to list of connections | 00:53 |
| freemangordon | yep, was just about to ask :) | 00:53 |
| Wizzup | freemangordon: the crash might be a local thing (cellular) | 00:53 |
| freemangordon | because here it never finished connecting (no dhcp sever setup) | 00:53 |
| Wizzup | heh | 00:54 |
| freemangordon | ok, so we have a fix now that works only once :) | 00:54 |
| Wizzup | what do you mean? | 00:54 |
| freemangordon | I may consider sending flush | 00:54 |
| freemangordon | BSSAdded is never called more than once here | 00:54 |
| Wizzup | right, ok | 00:55 |
| freemangordon | Wizzup: building a version which flushes cache before scanning for a hidden AP | 07:47 |
| freemangordon | however, I think there might be some issue in icd wpa plugin - it seems to try to connect forever to a hidden AP which is non-existent | 07:48 |
| freemangordon | I tried the same on fremantle and it gives error dialog in 10 seconds or so | 07:49 |
| freemangordon | that could be part of the issue that connui dialog has to be closed if first connection attempt fails | 07:49 |
| freemangordon | you can easily trigger that by entering wrong ssid for a hidden AP - wifi icon will start blinking forever | 07:50 |
| freemangordon | also, it is a good idea to mask psk and others like it when logging from the plugin | 08:00 |
| Wizzup | freemangordon: ok, it's possible I lack some timeout | 09:42 |
| diejuse | apt install xfce4-appfinder | 10:09 |
| diejuse | it works and looks perfect on MLeste | 10:10 |
| freemangordon | how does it look? | 10:11 |
| freemangordon | I mean - screenshot please :) | 10:11 |
| diejuse1 | https://drive.google.com/file/d/1SLt1kIB3_ALAiWxQhcN2sH4Bma2vqU3A/view?usp=drivesdk | 10:36 |
| diejuse | freemangordon: There is the image. | 10:37 |
| freemangordon | diejuse1: yeah, looks nice | 11:10 |
| freemangordon | Wizzup: what about https://pastebin.com/mxfVBsAy ? | 13:25 |
| freemangordon | this seems to fix "connecting forever" bug | 13:25 |
| Wizzup | freemangordon: sorry, in plane now, will check later tonight, pls ping me if I do not look at it tonight | 14:11 |
| freemangordon | Wizzup: no hurry, also, the patch seems to break normal connections | 14:46 |
| freemangordon | lets have a look/discussion when you have more time | 14:46 |
| mighty17[m] | Would you guys like adding support for another device (omap4430) ofc I'll do the porting and stuff, but I need directions | 16:32 |
| Wizzup | mighty17[m]: what device? | 18:00 |
| mighty17[m] | <Wizzup "mighty17: what device?"> Samsung Galaxy Tab 2 | 18:33 |
| mighty17[m] | Iirc I asked before but that was in freenode channel | 18:33 |
| Wizzup | mighty17[m]: sure, if you have display, wifi and usb or something working, we can start looking at building an image for it | 18:38 |
| Wizzup | I assume you don't have powervr working yet | 18:38 |
| mighty17[m] | <Wizzup "I assume you don't have powervr "> in pmOS i do! | 18:39 |
| mighty17[m] | https://wiki.postmarketos.org/wiki/Samsung_Galaxy_Tab_2_7.0%22_(samsung-espresso3g) | 18:39 |
| Wizzup | ok, cool | 18:40 |
| Wizzup | which ddk is that? | 18:40 |
| Wizzup | mighty17[m]: I am thinking what makes the most sense perhaps is for you to download a droid4 image that we have, and then modify the kernel/modules and a few other things to fix your device | 18:41 |
| Wizzup | we have this for device specific config: https://github.com/maemo-leste/leste-config | 18:41 |
| mighty17[m] | <Wizzup "which ddk is that?"> 1.17 | 18:42 |
| Wizzup | ok, for leste you'll want an older one | 18:42 |
| Wizzup | I'd suggest to use the one we use for the d4 | 18:42 |
| Wizzup | (the latest one does not support X) | 18:42 |
| mighty17[m] | 1.9 as it supports X | 18:42 |
| Wizzup | yes | 18:42 |
| mighty17[m] | i use the linux_openpvrsgx kernel anyways, it requires some patches for charger and touch (and dts ofc) | 18:43 |
| Wizzup | great | 18:44 |
| Wizzup | so maybe start with a droid 4 or bionic image, add your own kernel+modules, and check leste-config for other files you might want to tweak | 18:45 |
| Wizzup | and then we can look at adding a leste-config package for it, and then auto build images for it | 18:45 |
| mighty17[m] | i was checking droid4 and the entire https://github.com/maemo-leste/leste-config/tree/master/leste-config-droid4/etc/mce/mce.ini.d seems unneeded for tab 2 (henceforth called as espresso/3g) | 18:47 |
| Wizzup | does the tablet have a rgb status led? | 18:48 |
| Wizzup | and does it have an ambient light sensor? | 18:48 |
| Wizzup | and vibrator, etc? | 18:49 |
| Wizzup | sometimes you need to provide a sysfs path for those kind of things | 18:49 |
| Wizzup | of course for an initial port this does not matter | 18:49 |
| mighty17[m] | no led, light sensor is still not working (i cant get gpadc working) | 18:49 |
| Wizzup | but you''ll probably want some kind of nce.ini.d | 18:49 |
| Wizzup | not even a charging led? ok | 18:49 |
| mighty17[m] | <Wizzup "of course for an initial port th"> yeah for now i will do what u said, use d4 img and my kernel | 18:49 |
| mighty17[m] | <Wizzup "not even a charging led? ok"> yup not even that | 18:49 |
| Wizzup | ok | 18:50 |
| Wizzup | well, you can omit it and see how it goes | 18:50 |
| mighty17[m] | a lot of things need to be omitted xD | 18:51 |
| Wizzup | exciting nonetheless | 18:52 |
| mighty17[m] | ``` | 18:53 |
| mighty17[m] | /usr/share/X11/xorg.conf.d/40-ts.conf.leste | 18:53 |
| mighty17[m] | /usr/share/X11/xorg.conf.d/99-omap.conf.leste | 18:53 |
| mighty17[m] | what are these two | 18:53 |
| Wizzup | is this in leste-config-mapphone ? | 18:55 |
| Wizzup | https://github.com/maemo-leste/leste-config/blob/master/leste-config-mapphone/usr/share/X11/xorg.conf.d/99-omap.conf.leste | 18:55 |
| mighty17[m] | yes yes found it | 18:55 |
| Wizzup | ok | 18:55 |
| Wizzup | not sure about 40-ts.conf | 18:55 |
| Wizzup | I don't see it anymore in leste-config | 18:55 |
| mighty17[m] | https://github.com/MightyM17/leste-config/commit/424739f9cdbdd22097a88c3d8cb6276d00bff158 | 18:56 |
| mighty17[m] | <Wizzup "not sure about 40-ts.conf"> its in pinephone folder somehow | 18:57 |
| mighty17[m] | pocophone* | 18:57 |
| Wizzup | well, various devices could have had such a ts.conf | 19:04 |
| Wizzup | mighty17[m]: I suggest (later, when submitting a PR) that you create a new directory for your device rather than modifying the droid4 one | 19:04 |
| mighty17[m] | <Wizzup "mighty17: I suggest (later, when"> yes ofc, i wanted to copy it now and do changes but t'was too much work for initial port | 19:11 |
| Wizzup | :) | 19:12 |
| mighty17[m] | <Wizzup ":)"> now what, how do i build it, (suggestion add dummy device package) | 19:53 |
| uvos | mighty17[m]: note that leste-config-droid4 is NOT a compleate or sufficant device specification for leste | 20:26 |
| uvos | mighty17[m]: only the combination of leste-config-mapphone and leste-config-droid4 has everything needed for the d4 | 20:26 |
| uvos | your device is not a mapphone so it dose not have the cpcap devices assumed by *-mapphone packages | 20:27 |
| mighty17[m] | correct | 20:27 |
| mighty17[m] | i think i removed then | 20:27 |
| mighty17[m] | them* | 20:27 |
| uvos | to create a new device you must | 20:28 |
| uvos | 1 provide a seris of kernel patches that apply to https://github.com/maemo-leste/droid4-linux/tree/droid4-pending-v5.11 (sorry about the bad name of this repo) | 20:29 |
| uvos | best base your patches on upstream 5.11 then they will apply to this branch | 20:29 |
| uvos | these patches must be of resonable quality that they can be upstreamed to the mainline kernel. | 20:30 |
| uvos | 2. you must add your device to https://github.com/maemo-leste/image-builder | 20:30 |
| uvos | 3. you must provide the package list needed by your device to https://github.com/maemo-leste/hildon-meta/blob/master/debian/control | 20:31 |
| uvos | 4. you must provide a pr on leste-config that creates a package in leste-config (an be minimal or even empty) | 20:31 |
| uvos | if you do that you device will be supported as a "experiamental" device | 20:32 |
| mighty17[m] | <uvos "best base your patches on upstre"> is 11 necessary, i have a driver which was added in 5.12 | 20:33 |
| uvos | mighty17[m]: yes | 20:33 |
| mighty17[m] | ok i will add it as a patch | 20:33 |
| mighty17[m] | also my cpu is clocked to 1GHz compared to 1.2 on d4 | 20:33 |
| uvos | mighty17[m]: so? your dts should specify this | 20:34 |
| mighty17[m] | <uvos "if you do that you device will b"> seems like quite a lot | 20:34 |
| uvos | or use omap4 defaults whitch are 1.0 | 20:34 |
| mighty17[m] | <uvos "mighty17: so? your dts should sp"> what? doesnt the omap4.dtsi specify that | 20:34 |
| mighty17[m] | ok nvm | 20:34 |
| uvos | mighty17[m]: yes it dose we override that for mapphones | 20:35 |
| uvos | mighty17[m]: but as your device is a omap normal speed one its fine as is | 20:35 |
| uvos | if you want your device to be marked "supported" you must achive functionality somewhate close to the droid bionic see the table here: https://leste.maemo.org/Motorola_Droid_Bionic | 20:35 |
| uvos | and you must test the device regularly and at least one other active persion in the comunity must also have the same device | 20:36 |
| mighty17[m] | thats quite a lot | 20:37 |
| mighty17[m] | mostly i wont be working on 3g/modem | 20:37 |
| mighty17[m] | https://wiki.postmarketos.org/wiki/Samsung_Galaxy_Tab_2_7.0%22_(samsung-espresso3g) | 20:37 |
| uvos | on a tablet that fine | 20:37 |
| uvos | your device is a tablet right? | 20:37 |
| uvos | no modem is perfecly fine | 20:38 |
| uvos | functionaliy is also not so important as regular testing really | 20:38 |
| mighty17[m] | <uvos "your device is a tablet right?"> yes | 20:38 |
| mighty17[m] | <uvos "functionaliy is also not so impo"> i think i know a person who can do testing | 20:38 |
| uvos | anyhow lets get you into experiamental state first :P | 20:39 |
| mighty17[m] | indeed xD | 20:39 |
| uvos | i have a driver which was added in 5.12 | 20:40 |
| uvos | what driver was this btw? | 20:40 |
| mighty17[m] | yas530 | 20:40 |
| mighty17[m] | magnetometer | 20:40 |
| uvos | oh | 20:40 |
| uvos | dont patch that in | 20:40 |
| uvos | just drop it in 5.11 | 20:40 |
| mighty17[m] | like cp the driver lol | 20:41 |
| uvos | dont do this, just leave the magnetometer broken in leste untill we move to 5.12+ | 20:41 |
| mighty17[m] | <uvos "1 provide a seris of kernel patc"> https://github.com/maemo-leste/droid4-linux/tree/droid4-pending-pvr-omapdrm-v5.11 ? | 20:41 |
| uvos | no | 20:41 |
| uvos | just patches on top mainline 5.11 or droid4-pending-v5.11 | 20:42 |
| uvos | with droid4-pending-v5.11 prefered | 20:42 |
| uvos | if you have patches to the pvr driver please provied thos (and only those) on top of droid4-pending-pvr-omapdrm-v5.11 | 20:43 |
| uvos | i ment with mainline v5.11 prefered sorry | 20:43 |
| mighty17[m] | <uvos "if you have patches to the pvr d"> Oh I thought that was the kernel (like openpvrsgx) | 20:44 |
| uvos | no | 20:44 |
| uvos | same with openpvrsgx | 20:44 |
| uvos | its the kernel | 20:44 |
| uvos | tree | 20:44 |
| uvos | but in reality we just merge that in | 20:44 |
| mighty17[m] | Ah okay | 20:45 |
| uvos | and delop on top of mainline | 20:45 |
| mighty17[m] | Basically fork of openpvrsgx | 20:45 |
| uvos | right we develop on the droid4-pending-* branch and merge in droid4-pending-pvr-omapdrm-v5.11 just before compileing | 20:46 |
| uvos | so our commits are never at risk of dependancies to the pvr driver/patches | 20:47 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!