| n00mann[m] | Hello | 00:31 |
|---|---|---|
| n00mann[m] | I am ordering an n900 | 00:32 |
| n00mann[m] | What i am in for lads | 00:32 |
| Guest86 | Hi | 00:36 |
| Guest86 | Linux phone is a cool toy | 00:37 |
| Guest86 | The nokia n950 looks interesting, i hope that i get one someday | 00:38 |
| Guest86 | A local guy had it but not sure if he will pass it to me | 00:38 |
| n00mann[m] | Surely the support wouldn’t be the best | 00:38 |
| kona | n00mann[m]: idk, my n900 runs leste just fine, but it is a little light on memory so even with zramfs you have to be patient. | 02:54 |
| n00mann[m] | Can it run mpv say a 360p h264 videos oke? | 03:18 |
| n00mann[m] | I imagine links browser should work okey but does qtbrowser work as well? | 03:20 |
| n00mann[m] | * I imagine the links browser | 03:21 |
| freemangordon | Wizzup: xorg segfaulted here, but I suspect it is not the driver | 09:57 |
| freemangordon | going to try to port sp-rich-core | 09:58 |
| Wizzup | ok | 09:58 |
| Wizzup | n00mann[m]: droid4 is probably a better pick unless you're in the US | 09:59 |
| Wizzup | but the n900 should work okayish, we don't have calls working properly yet | 09:59 |
| freemangordon | hmm, maybe someone on the channel might volunteer to do it | 10:12 |
| freemangordon | anybody wants to port sp-rich-core (and friends) to leste? | 10:12 |
| freemangordon | http://repository.maemo.org/pool/fremantle/free/s/sp-rich-core/ | 10:12 |
| freemangordon | it also needs sp-endurance | 10:12 |
| freemangordon | Wizzup: you may want to enable coredumps too | 10:18 |
| freemangordon | uvos: vkb seems to have issues sending utf8 characters to non-gtk apps | 10:20 |
| freemangordon | IIRC it is you that enabled that | 10:21 |
| freemangordon | ok, trying to send cyrillic text to empathy-chat and then playing with the menu makes xorg freeze | 10:23 |
| freemangordon | aaaand... we have a coredump :) | 10:24 |
| freemangordon | https://pastebin.com/1cUChkBK | 10:28 |
| freemangordon | ok, seems we are missing this https://gitlab.freedesktop.org/xorg/xserver/-/commit/35e5a76cc1d02801fadd49d12e60664b02e4bebc | 10:40 |
| freemangordon | Wizzup: I think we shall upgrade xorg | 10:50 |
| freemangordon | there are too many fixes since our version IIUC | 10:50 |
| Wizzup | hmm | 10:54 |
| freemangordon | also how to satisfy xorg systemd build dependency? | 10:55 |
| freemangordon | I installed libelogind-dev to no use | 10:55 |
| Wizzup | I was going to suggest that | 10:55 |
| Wizzup | are you trying to build upstream xorg now? | 10:55 |
| Wizzup | we forked it also for some other patches | 10:56 |
| freemangordon | no, this is our xorg | 10:56 |
| freemangordon | and I cannot build it | 10:56 |
| freemangordon | without installing systemd-dev | 10:56 |
| Wizzup | https://github.com/maemo-leste-upstream-forks/xorg-server/blob/maemo/ascii/debian/control#L100 | 10:56 |
| Wizzup | sudo apt-get build-dep . (from debian/ ?) | 10:56 |
| freemangordon | lemme try | 10:57 |
| Wizzup | hmm that's the wrong branch | 10:58 |
| Wizzup | I think spinal made maemo/ascii pinned | 10:58 |
| freemangordon | The following NEW packages will be installed: | 10:58 |
| freemangordon | libpolkit-gobject-consolekit-1-0 libsystemd-dev libsystemd0 | 10:58 |
| freemangordon | anyway, will install | 10:58 |
| freemangordon | it seems jenkins pulls libsystemd during the build | 11:00 |
| freemangordon | I will test the ^^^ patch with our xorg, that's whay I want to build it | 11:00 |
| freemangordon | *why | 11:01 |
| freemangordon | Wizzup: seems the patch applies fine, will push to -devel | 11:02 |
| freemangordon | can;t wait 3 hours until xorg builds on d4 | 11:02 |
| Wizzup | sure | 11:03 |
| _uvos_ | freemangordon: it is impossible to send non english charecters to x11 if an english keymap is selected | 11:04 |
| _uvos_ | x11 keycodes are only 8 bit wide | 11:04 |
| freemangordon | ugh | 11:04 |
| _uvos_ | it will work if you select an cryllic keymap for x11 | 11:04 |
| _uvos_ | but then the hwkbd wont work | 11:04 |
| _uvos_ | it is impossible for both to work at the same time | 11:05 |
| _uvos_ | due to fundamental x11 limits | 11:05 |
| freemangordon | great | 11:05 |
| freemangordon | ok, thanks | 11:05 |
| _uvos_ | the solution i have planned | 11:05 |
| _uvos_ | is for the vkb to switch the keymap when it opens | 11:05 |
| _uvos_ | and swich back on close | 11:05 |
| _uvos_ | the hwkbd will be broken then while the vkb is open | 11:06 |
| _uvos_ | but its the best you can do | 11:06 |
| freemangordon | hwkbd is inactive anyways when vkb is opened | 11:06 |
| _uvos_ | the problem with this is we need some him->x11 keymap table | 11:06 |
| _uvos_ | not true | 11:06 |
| freemangordon | how's that? | 11:06 |
| Wizzup | inactive in the sense that nobody would be expected to use it | 11:06 |
| Wizzup | sure you can press keys on it | 11:07 |
| freemangordon | if you press any key on hwkbd, vkb will close | 11:07 |
| _uvos_ | sill that keypress will be wrong | 11:07 |
| freemangordon | we can ignore the first char | 11:07 |
| _uvos_ | also i skipped the check when the user request vkb open via the open key | 11:07 |
| _uvos_ | because it might be usefull sometimes | 11:08 |
| freemangordon | check for? | 11:09 |
| _uvos_ | vkb usual dosent open if silde is opened | 11:10 |
| _uvos_ | this check is skipped if the user explicitly requests vkb | 11:10 |
| _uvos_ | as opposed to it opening on field click | 11:10 |
| freemangordon | ok, but vkb closes on HWKBD keypress | 11:11 |
| _uvos_ | sure | 11:11 |
| freemangordon | so the only thing we shall do in that case is to ignore the first hwkbd char, while vkb has still not restored the mapping | 11:12 |
| _uvos_ | i gues | 11:12 |
| _uvos_ | also we need a table | 11:12 |
| _uvos_ | the problem here is | 11:12 |
| _uvos_ | that there is no 1 to 1 corrispondance of him input languages and x11 keymaps | 11:12 |
| _uvos_ | we really need to make a special x11 keymap for eatch him lang | 11:13 |
| Wizzup | freemangordon: btw, not urgent, but was your ioload problem solved with the new kernel? | 11:13 |
| _uvos_ | lots of work | 11:13 |
| freemangordon | _uvos_: can;t we just build it in run-time? | 11:13 |
| freemangordon | Wizzup: no :( | 11:13 |
| _uvos_ | maybe idk | 11:13 |
| _uvos_ | not sure about xcb internals to that degree | 11:14 |
| Wizzup | freemangordon: ok, well, I suppose we could push it through to stable soon in any case | 11:14 |
| _uvos_ | *xkb | 11:14 |
| freemangordon | Wizzup: https://github.com/maemo-leste/pine64-kernel/issues/7#issuecomment-1256370594 | 11:14 |
| freemangordon | Wizzup: wait, what? not only that, but TS does not work at all | 11:15 |
| freemangordon | I don;t see how we can push that to stable | 11:15 |
| freemangordon | and trust me, this is no "my ioload" problem, there is something terribly wrong in mmc driver my gut feeling tells me | 11:16 |
| rafael2k | nope | 11:16 |
| freemangordon | Wizzup: why are xorg tags that long? is it ok | 11:16 |
| rafael2k | unless all Mobian users also have the same... which I doubt | 11:17 |
| rafael2k | :/ | 11:17 |
| freemangordon | well, I see nothing specioal about my device | 11:17 |
| freemangordon | *special | 11:17 |
| rafael2k | which is your PP version? | 11:17 |
| freemangordon | braveheart | 11:17 |
| freemangordon | no idea how to provide more specific info | 11:17 |
| rafael2k | different than mine | 11:17 |
| freemangordon | mhm | 11:17 |
| freemangordon | the fact that I can reliably can reproduce the issue by just changing the kernel, should be enough | 11:18 |
| rafael2k | so better you try other "upstream" and point which one works fine, we can choose that branch | 11:18 |
| rafael2k | I can not help | 11:18 |
| freemangordon | you can :). by providing some hints how to build that | 11:19 |
| freemangordon | opr, how to choose the correct tree | 11:19 |
| rafael2k | debuild -uc -us | 11:19 |
| freemangordon | *or | 11:19 |
| rafael2k | ? | 11:19 |
| freemangordon | ah, it already have debian packaging? | 11:19 |
| freemangordon | *has | 11:19 |
| rafael2k | yeap | 11:19 |
| freemangordon | ok | 11:19 |
| Wizzup | freemangordon: hm @ ok ts doesn't work, I'll test it as well then | 11:19 |
| Wizzup | I will brb | 11:20 |
| freemangordon | rafael2k: ok, will try | 11:20 |
| freemangordon | but, lets push xorg first to -devel | 11:20 |
| rafael2k | I have a feeling distros are not supporting pretty well the braveheart anymore | 11:20 |
| rafael2k | which is sad indeded | 11:20 |
| rafael2k | anyway, lets try to find a a good kernel, or may be bisect the issue in order 5.15... | 11:21 |
| rafael2k | *our 5.15 tree | 11:21 |
| rafael2k | freemangordon: we can also try to look the changes in the patchset between 5.15.21 and 5.14.48 and see wtf broke braveheart | 11:22 |
| freemangordon | I tried, to no use | 11:23 |
| freemangordon | but didn;t spent much time | 11:23 |
| rafael2k | with my "building on device" setup I'll certainly take 2 years to bisect the issue | 11:23 |
| freemangordon | I will focus on that, I just want to fix those several outstanding xorg issues first | 11:24 |
| rafael2k | I agree that is more important for now (and guarantee that Maemo 5.15.68 for PP >= 1.2 is working fine) | 11:24 |
| freemangordon | I trust you, it is that it does not work on my device, which is kind of de-motivating :) | 11:25 |
| rafael2k | I can work on 6.0 these coming week | 11:25 |
| rafael2k | may be just grab Megi's 6.0 branch and build for experimentation | 11:26 |
| freemangordon | I am almost sure the issue will be there | 11:26 |
| rafael2k | :/ | 11:26 |
| freemangordon | I prefer to try to bisect | 11:26 |
| rafael2k | I have the same feeling tbh | 11:26 |
| freemangordon | so I will start building release after release until I hit the issue | 11:28 |
| freemangordon | and then it will become more or less clear what it is about | 11:29 |
| freemangordon | what worries me though is that it will take ages on the device | 11:29 |
| freemangordon | omg, git-buildpackage pulled mysql | 11:30 |
| norayr | i don't know if it is connected but xorg crashes for me on droid4 | 11:33 |
| norayr | sometimes when i use it | 11:33 |
| norayr | doesn't crash when i don't use it. | 11:33 |
| norayr | so might be connected to typing | 11:33 |
| norayr | but not sure | 11:33 |
| norayr | i mean it never crashed if it just stays there with pidgin running | 11:34 |
| norayr | but when i use it it crashes | 11:34 |
| freemangordon | this is what I am trying to fix now | 11:34 |
| norayr | but maybe it also crashes when i use hwkbi | 11:34 |
| norayr | i don't know | 11:34 |
| norayr | yes yes that would be great if u fix it. | 11:35 |
| rafael2k | freemangordon: ok! lemme know once you find the offending version, I can help checking the diff of the patches | 11:40 |
| freemangordon | ok, seems cannot repro xorg crash after the upgrade | 11:59 |
| Wizzup | freemangordon: that is with the extra patch? | 12:02 |
| freemangordon | yes | 12:02 |
| freemangordon | bbl | 12:04 |
| sicelo | but the n900 should work okayish, we don't have calls working properly yet / < freemangordon> hmm, maybe someone on the channel might volunteer to do it | 22:49 |
| sicelo | do what? make n900 calls work? | 22:50 |
| freemangordon | sicelo: port sp-rich-core to leste | 23:15 |
| sicelo | ah :-) | 23:22 |
| * freemangordon wonders how hard it would be to port h-i-m gtk plugin to gtk3 | 23:26 | |
| Wizzup | not too hard I think | 23:40 |
| freemangordon | mhm | 23:40 |
| Wizzup | I looked at it briefly a while ago | 23:40 |
| Wizzup | it is mostly the same api | 23:40 |
| freemangordon | looking at it ATM | 23:40 |
| freemangordon | yes | 23:40 |
| freemangordon | and I think this would be better than hacking around with x11 | 23:40 |
| Wizzup | sure, but x11 fallback is not bad | 23:41 |
| Wizzup | we also need it for qt5 :) | 23:41 |
| Wizzup | atm | 23:41 |
| freemangordon | yes, but I think we can use him there as well | 23:41 |
| Wizzup | yes | 23:41 |
| Wizzup | I had that working as a hack | 23:42 |
| freemangordon | I am first to agree him is not the best, but OTOH - what is? | 23:42 |
| freemangordon | do we really have a replacement? | 23:42 |
| freemangordon | unless I promised to do something else first (which I don;t remember) I think the next I will start working on is gtk3 h-i-m plugin | 23:43 |
| freemangordon | should be low-hanging fruit and will allow us to use browsers with non-english characters | 23:44 |
| freemangordon | and will also ease the usage on PP | 23:44 |
| Wizzup | ok | 23:45 |
| Wizzup | agree | 23:45 |
| freemangordon | both FF and chromium are gtk3, right? | 23:45 |
| Wizzup | not sure | 23:51 |
| * Wizzup zzz | 23:51 | |
| freemangordon | good night | 23:51 |
| sicelo | freemangordon: if I'm not mistaken, you had planned to work on vrfb next | 23:59 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!