| Wizzup | sicelo: on upower, shall I just build the one from trixie (if that is new enough) for daedalus? | 02:34 |
|---|---|---|
| sicelo | Trixie version is fine. although if there are no other concerns, I think git master is preferable | 05:57 |
| freemangordon | Wizzup: we use cgroups v2, right? | 07:49 |
| freemangordon | arno11: instead of doing sudo, perhaps you can check what cgrulesengd can do for us (in cgroups v2 you can assign nice value to a group) | 07:53 |
| freemangordon | so we can distribute /etc/cgrules.conf with rules for Xorg, mce, h-d, etc | 07:54 |
| freemangordon | and we can use ohm rules from fremantle | 07:54 |
| freemangordon | s/use/port | 07:55 |
| freemangordon | hmm, at least VM is using cgroup v1 | 07:58 |
| freemangordon | arno11: see /usr/share/policy/etc/current/syspart.conf | 08:05 |
| freemangordon | I am really tempted to consider moving to systemd, we get all those things there with unit files | 08:15 |
| freemangordon | anyway... | 08:15 |
| freemangordon | arno11: seems we can use inotify to watch new processes creation | 08:16 |
| freemangordon | see https://paste.debian.net/1364525/ | 08:25 |
| freemangordon | 'processed': https://paste.debian.net/1364526/ | 08:28 |
| freemangordon | everything else is done through cgroups | 08:28 |
| freemangordon | for Xorg, we should be able to set nice value from the startup script | 08:34 |
| freemangordon | hmm, we do it, but autologin starts new pam session, which starts Xorg with nice 0 | 08:50 |
| freemangordon | so, we shall create some daemon/script/whatever that monitors processes creation and sets the appropriate nice value | 08:51 |
| Wizzup | no, let's not make a new daemon | 11:15 |
| Wizzup | https://github.com/OpenRC/openrc/blob/master/user-guide.md | 11:19 |
| Wizzup | https://wiki.gentoo.org/wiki/OpenRC/CGroups#Activating_cgroup_feature_support | 11:20 |
| Wizzup | let's do it this way | 11:20 |
| Wizzup | for xsession scripts, we can use cgrun if we must hack it in there | 11:20 |
| Wizzup | and ifwe do have cgrulesengd, why do we need another daemon? | 11:21 |
| mkf | hello. | 13:08 |
| mkf | i've got otg working today. :) | 13:09 |
| sicelo | nice progress! what image is your work based on btw? | 13:09 |
| mkf | sunxi image in the website, but it's quite change right now. | 13:10 |
| mkf | *changed | 13:10 |
| mkf | also updated to chimaera | 13:10 |
| mkf | have wireless worked for you? i'm not sure if i'm doing something wrong or i have fired the chip... :( | 13:10 |
| mkf | freemangordon, Wizzup: ^ | 13:11 |
| freemangordon | mkf: wifi works on my sunxi, but I had to patch the driver | 13:59 |
| freemangordon | my tablet has some exotic realtek chip | 13:59 |
| freemangordon | Wizzup: cgrulesengd can't control nice value | 14:00 |
| freemangordon | or rather - we use cgroups v1, which can't control niceness | 14:00 |
| freemangordon | if we can enable cgroup v2, then we don;t need another daemon | 14:03 |
| freemangordon | hmm... | 14:03 |
| freemangordon | oh, wait, seems v2 is mounted under /sys/fs/cgroup/unified | 14:09 |
| freemangordon | so yeah, we can go with v3 | 14:09 |
| freemangordon | *v2 | 14:09 |
| Wizzup | freemangordon: then we rename in the xsession | 14:10 |
| Wizzup | renice* | 14:10 |
| freemangordon | no, we can't | 14:10 |
| freemangordon | xsession is run as user | 14:11 |
| Wizzup | why not? sudo renice $(pidof Xorg) | 14:11 |
| freemangordon | no, wait | 14:11 |
| Wizzup | https://unix.stackexchange.com/questions/102945/how-can-i-allow-a-user-to-prioritize-a-process-to-negative-niceness | 14:11 |
| Wizzup | looks like pam can let users set negative niceness | 14:11 |
| freemangordon | but we want to ue cgroups for that | 14:11 |
| freemangordon | *use | 14:11 |
| Wizzup | ok, if cgroups v2 can do it | 14:11 |
| freemangordon | yes, they can | 14:11 |
| freemangordon | they have the so called 'sched' controller | 14:12 |
| freemangordon | that seems to support niceness | 14:12 |
| freemangordon | at least AI said so :) | 14:13 |
| mkf | freemangordon: mine has that too. :) | 14:13 |
| mkf | what patches you made? | 14:14 |
| Wizzup | freemangordon: I am not sure if I will believe AI, but ok | 14:14 |
| freemangordon | "echo 10 > /sys/fs/cgroup/my_cgroup/sched.niceness" | 14:15 |
| freemangordon | mkf: what exactly? | 14:15 |
| Wizzup | freemangordon: ok, well, let me fix this n900 fstab issue first, but it sounds like cgroups is the answer here | 14:16 |
| freemangordon | mhm | 14:16 |
| Wizzup | freemangordon: btw, any chance to look at git.maemo.org? if we're happy with it I might do the migration some time soon | 14:16 |
| Wizzup | or at least proceed further with the migration | 14:16 |
| Wizzup | I'll check out some sign in via github kind of things, too | 14:16 |
| freemangordon | sorry, but I lost the context why do we need to migrate out of github | 14:17 |
| freemangordon | iow, what is wrong with github? | 14:17 |
| Wizzup | it's closed source, it's microsoft, it's US based, the CI/CD is not our own, and I don't want to say build packages on github infra | 14:17 |
| Wizzup | and I don't trust them to not throw us off when anyone complains | 14:17 |
| freemangordon | ok :) | 14:17 |
| freemangordon | oh, so we can do builds on commit with our infra? | 14:18 |
| Wizzup | the reason we went for it initially was that we thought the network effect would attract people | 14:18 |
| Wizzup | re: builds on commit - I plan to make that work, yes | 14:18 |
| freemangordon | ok, cool | 14:18 |
| freemangordon | ok, will have a look soon | 14:19 |
| mkf | freemangordon: rtl8723bs or 8702as or something like that... | 14:19 |
| freemangordon | no, it is important the *exact* chip | 14:19 |
| freemangordon | because if it is BS, it is the same as mine and I would be able to help | 14:20 |
| mkf | how do i know what's the chip exactly? | 14:20 |
| freemangordon | check in fex | 14:20 |
| mkf | ok | 14:21 |
| freemangordon | mkf: all the changes I have made https://paste.debian.net/1364571/ | 14:26 |
| freemangordon | at the end is drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c | 14:26 |
| freemangordon | BTW, your issues with TS might be related with wrong panel config | 14:26 |
| freemangordon | see unk_qc760bg1_1024_600_72_mode | 14:27 |
| freemangordon | bbl | 14:31 |
| mkf | mine is 800 x 480 tho | 14:34 |
| freemangordon | how do you know? from fex? | 14:35 |
| mkf | fex. | 14:35 |
| freemangordon | ok | 14:36 |
| mkf | however the image isn't the original one shiped with device, and wifi doesn't work in android, so i'm not sure if fex is to be trusted about wifi | 14:37 |
| mkf | sunxi wiki says i should have a realtek, and i do see a realtek chip in my board | 14:37 |
| freemangordon | ok | 14:37 |
| mkf | but fex (of incompatible?) android image says i have an rda. which i don't think is true. | 14:37 |
| freemangordon | did you enable the driver? | 14:37 |
| freemangordon | it is in staging | 14:38 |
| mkf | i have, but didn't applied the patches yet | 14:38 |
| mkf | let me recompile | 14:38 |
| freemangordon | wait | 14:38 |
| freemangordon | did it probe properly? | 14:38 |
| freemangordon | *does it* | 14:38 |
| mkf | how do i know? | 14:38 |
| freemangordon | lsmod? | 14:38 |
| mkf | it loads it fine (added it in /etc/modules) | 14:39 |
| freemangordon | you should not load it by hand | 14:39 |
| mkf | oh. ok. | 14:39 |
| freemangordon | lemme boot my device to be sure what happens | 14:39 |
| mkf | if i don't load it, who will load it then? | 14:39 |
| freemangordon | umm... who loads the other modules? | 14:40 |
| freemangordon | udev I would say | 14:40 |
| mkf | idk, i load all modules by hand. :) | 14:40 |
| freemangordon | oh, how's that? | 14:40 |
| mkf | gotcha. | 14:40 |
| mkf | idk, silead didnt worked properly and i thought modules must be manually loaded. | 14:41 |
| freemangordon | nope | 14:41 |
| freemangordon | everything here is auto-probed | 14:41 |
| freemangordon | keep in mind that even with that patch wifi has issues connecting every now and then | 14:41 |
| mkf | oh well. | 14:42 |
| freemangordon | [ 13.172909] r8723bs: module is from the staging directory, the quality is unknown, you have been warned. | 14:43 |
| mkf | it's not laoded automatically. | 14:43 |
| mkf | *laoded | 14:43 |
| mkf | silead did, so i think it's not probed. | 14:43 |
| mkf | or my dts is faulty. hm. | 14:43 |
| freemangordon | yeah, could be | 14:44 |
| freemangordon | what dts do you use? sun8i-a33-q8-tablet.dts ? | 14:44 |
| mkf | a23-q8-tablet | 14:46 |
| freemangordon | oh, yours is not a33 | 14:46 |
| mkf | http://cloud9p.org/paste/cf08 | 14:47 |
| mkf | yeah but these two are close enough. | 14:47 |
| freemangordon | rtl8723bs: sdio_wifi@1 { | 14:47 |
| freemangordon | so yeah, same chip | 14:47 |
| mkf | (i have modified this file) | 14:47 |
| mkf | https://elixir.bootlin.com/u-boot/v2025.04-rc4/source/arch/arm/dts/sun8i-a23-q8-tablet.dts | 14:49 |
| mkf | original file | 14:49 |
| freemangordon | there is nothing about wifi there | 14:50 |
| mkf | yeah. :( | 14:51 |
| freemangordon | hmm, wait | 14:51 |
| freemangordon | &mmc1 | 14:51 |
| freemangordon | it is missing "status = "okay"; ,no? | 14:52 |
| mkf | sun8i-q8-common.dtsi does define &mmc1 | 14:52 |
| mkf | and it has status ok | 14:52 |
| freemangordon | ok | 14:52 |
| mkf | tho it doesn't mention what wifi model it has | 14:52 |
| freemangordon | (I have kernel trees here, no need to paste upnaptched files) | 14:52 |
| freemangordon | *unpatched | 14:52 |
| mkf | sorry, i'm not very fluent in device tree tounge. :) | 14:53 |
| mkf | okay. thanks. | 14:53 |
| * freemangordon checks | 14:53 | |
| mkf | opened the tablet, apperently it has a 87?3As | 14:53 |
| freemangordon | so not BS? | 14:54 |
| mkf | no. | 14:54 |
| freemangordon | well,... | 14:54 |
| mkf | but these two share the same driver | 14:54 |
| freemangordon | that explains it | 14:54 |
| freemangordon | do they? | 14:54 |
| freemangordon | ok, but are you sure it is on emmc bus? | 14:54 |
| mkf | https://linux-sunxi.org/Wifi#SDIO-based | 14:54 |
| mkf | at least here says they share the same driver | 14:55 |
| freemangordon | yeah | 14:55 |
| mkf | i'm not sure if they are on emmc bus tho. | 14:55 |
| freemangordon | "SDIO" | 14:56 |
| mkf | ah. | 14:56 |
| mkf | wait emmc is too sdio? huh. | 14:56 |
| mkf | i thought they are distant standards | 14:56 |
| mkf | okay that explains a bit why a wifi should be on emmc | 14:57 |
| freemangordon | mkf: do you have anything "sdio" in dmesg? | 15:00 |
| freemangordon | also, check /dev/bus/sdio | 15:01 |
| gnarface | i have the same SDIO unit on a 2GB pine64+ board, and the debian kernel didn't have the right dtb patch, i had to manually add it | 15:01 |
| gnarface | device was showing up unpowered | 15:02 |
| gnarface | same driver too, lemme know if you can get more than 20 megabits out of it in host mode | 15:02 |
| gnarface | it'll work but i can't make it as fast as it's supposed to be (150MB/s!?) | 15:03 |
| freemangordon | gnarface: see https://paste.debian.net/1364571/, I think there is a bug in the driver | 15:04 |
| freemangordon | (patch at the end) | 15:04 |
| freemangordon | without that it does not even connect to my router here :) | 15:05 |
| freemangordon | or does it once in 20 reboots ;) | 15:05 |
| gnarface | hmm | 15:06 |
| gnarface | interesting... | 15:06 |
| mkf | gnarface, thank you. i think i never seen a n card go beyond 10Mb (18 Mb is theorical limit of 150MB cards) | 15:06 |
| gnarface | freemangordon: you mean just this hunk for drivers/staging/rtl8723bs/os_dep/ioctl_cfg80211.c? you think that'll fix the speed issue? | 15:08 |
| gnarface | or do i need more of this? | 15:08 |
| freemangordon | no idea | 15:08 |
| gnarface | hmmm | 15:08 |
| freemangordon | but at least it allows me to connect to my router 9 of 10 times | 15:08 |
| freemangordon | gnarface: btw, what do you use to test speed? | 15:09 |
| gnarface | seems promising... | 15:09 |
| gnarface | usually testing with netcat | 15:09 |
| gnarface | just time and netcat of a cached 50mb file from a ram disk to /dev/null | 15:09 |
| gnarface | 50MB i mean | 15:10 |
| freemangordon | yeah | 15:10 |
| mkf | gnarface: megabit or megabyte? | 15:10 |
| gnarface | heard some accounts of 60 megabits with wpasupplicant but they hadn't tested in host mode, so there also might be a speed discrepancy between host mode and client mode | 15:11 |
| gnarface | mkf: MB == megabyte, mb == megabit | 15:11 |
| freemangordon | lemme see if I'll be able to run speedtest.net :) | 15:11 |
| mkf | aren't cards rated for mb? | 15:11 |
| gnarface | yes, but the SDIO bus is rated in MB, i thought... | 15:12 |
| mkf | hm. | 15:13 |
| gnarface | basically it's supposed to be good for 150MB/s, current kernel/dtb settings show it should at least be working at 40MB/s, but the best first-hand accounts i've heard can't get more than 6MB/s out of it and in host mode i can't get more than 2MB/s out of it | 15:13 |
| freemangordon | ok, according to speedtest.net, my DL speed is 39.54, my UL is 46.78 | 15:13 |
| gnarface | MB or mb? | 15:13 |
| freemangordon | Mbps | 15:14 |
| gnarface | hmm, still basically double what i'm getting | 15:14 |
| gnarface | so maybe that patch is helping | 15:14 |
| freemangordon | those speeds are pretty normal, as my inet is 100Mb or something | 15:14 |
| freemangordon | and also, chromium uses some CPU, so that affects speets too I would say | 15:15 |
| freemangordon | *speeds | 15:15 |
| freemangordon | lemme re-test | 15:16 |
| freemangordon | 20.57/53.09 | 15:17 |
| freemangordon | but web page trying to show speed meter in real-time does not help much | 15:19 |
| arno11 | freemangordon: Wizzup: great @cgroup | 15:19 |
| mkf | freemangordon: are you running chromium on a33? | 15:20 |
| freemangordon | yes | 15:20 |
| mkf | does that work well? | 15:20 |
| mkf | (is it jib?) | 15:21 |
| freemangordon | no, *chromium* | 15:21 |
| mkf | oh. | 15:21 |
| freemangordon | works ok | 15:21 |
| mkf | i see. | 15:21 |
| freemangordon | with ublock origin, it is 'ok' | 15:21 |
| mkf | wait a sec | 15:22 |
| freemangordon | gnarface: so, with speedtest page scrolled so meter is hidden, I get 45.11/63.96 | 15:22 |
| freemangordon | mkf: hmm? | 15:23 |
| gnarface | freemangordon: that's in client mode though, it'd be interesting to see if it can maintain those speeds in host mode | 15:23 |
| freemangordon | no idea, I admit I have never used host mode :) | 15:24 |
| gnarface | that is more like what i've heard other people say they were getting out of it in client mode though | 15:24 |
| mkf | kernel nags about lack of regulatory db for 802.11 | 15:24 |
| mkf | could that might be a reason? | 15:24 |
| freemangordon | yeah, we knbow | 15:24 |
| freemangordon | no | 15:24 |
| gnarface | doesn't installing the crda package get rid of that error? | 15:25 |
| mkf | hm ok | 15:25 |
| freemangordon | gnarface: no, because we build our kernels | 15:25 |
| gnarface | oh | 15:25 |
| freemangordon | and there is something with the keys | 15:25 |
| mkf | i built my own, does it effects that? | 15:25 |
| freemangordon | kernel? | 15:25 |
| mkf | yeah. | 15:25 |
| freemangordon | sure | 15:25 |
| mkf | i have another wifi card, that's usb | 15:26 |
| mkf | and apperently it gets detected enough so ip a can see it | 15:26 |
| mkf | tho idk how to connect to wifi in maemo thru terminal (gui doesn't show a thing) | 15:26 |
| mkf | do we use iwl in maemo? | 15:26 |
| freemangordon | use wpa_supplicant | 15:27 |
| freemangordon | or rename to wlan0 | 15:27 |
| freemangordon | then maemo will see it | 15:27 |
| freemangordon | Wizzup: ok, won't work with cpu.weight.nice | 15:39 |
| freemangordon | this does not change niceness, but some scheduling priority within the group | 15:40 |
| Wizzup | we can leverage pam to allow user to set negative nice with renice | 15:40 |
| Wizzup | or maybe scheduling priority is enough | 15:40 |
| freemangordon | otoh, do we want to enable non-root to set niceness < 0? | 15:40 |
| freemangordon | I don't like the idea | 15:41 |
| freemangordon | I'd better invent some more secure way | 15:41 |
| freemangordon | or really, just create a 20-liner script (part of leste-config) that does inotifywait on /proc | 15:42 |
| Wizzup | it all feels like bad hacks imo | 15:42 |
| Wizzup | strange that we can't just put nice -20 in front of something. | 15:43 |
| freemangordon | if you allow user to do that, a user-started runaway process will hang the system | 15:43 |
| freemangordon | so I don't think that's a good idea | 15:43 |
| freemangordon | bbl | 15:44 |
| arno11 | Wizzup: freemangordon: maybe an alternative: i created a basic bash script (root) with the needed renice commands, linked to a xsession.post file (99my_script) and surprisingly works fine | 16:55 |
| Wizzup | I'm not surprise it works | 16:55 |
| Wizzup | we're just trying to figure out a way not to leverage passwordless sudo on user to work around this | 16:56 |
| Wizzup | as in, make it a bit more future proof, so to say | 16:56 |
| arno11 | yeah ofc | 16:57 |
| mkf | yipeee \o/ | 17:50 |
| mkf | wifi now probes | 17:51 |
| freemangordon | mkf: how? | 18:06 |
| mkf | i'm a dummy, it turns out i've compiled the incorrect driver. | 18:07 |
| freemangordon | heh | 18:07 |
| mkf | it loads this one now, and it's detected as wlan0 | 18:08 |
| mkf | alas maemo still shows no wifi networks | 18:08 |
| Wizzup | maybe you need to restart icd2 | 18:33 |
| mkf | uh. how can i stop hildon from starting? | 18:58 |
| mkf | or xorg even. | 19:04 |
| freemangordon | move dsme out of /etc/init.d/ | 19:20 |
| mkf | okay. | 19:20 |
| mkf | now got wifi in maemo working, altough i can't connect. trying to apply your patch..., which version of kernel this patch is for? | 19:21 |
| freemangordon | 6.12? | 19:21 |
| mkf | oh. | 19:23 |
| mkf | i suppose that driver got quite some love in versions between 5.8 and 6.12. | 19:24 |
| mkf | hm, linux 5.8 already have done what you did roughly. | 19:27 |
| mkf | nvm. | 19:31 |
| freemangordon | mkf: better use at least 6.12 | 20:14 |
| freemangordon | it works fine, at least here on a33 | 20:15 |
| arno11 | freemangordon: btw any idea about what's going on with Qt5 apps launch time ? | 20:36 |
| arno11 | gpu related or ? | 20:37 |
| arno11 | it is even a lot slower if i try to open more than one qt app. | 20:46 |
| arno11 | if i use rasterisation (force raster surface in qt5ct), it is almost as fast as what you described for d4. and i can open 4 or 5 qt apps @the same time with no troubles | 20:48 |
| arno11 | (i.e calculator, qalendar, hamsterfiler, conversations, tg-desktop) | 20:49 |
| arno11 | impossible OOTB with daedalus | 20:49 |
| mkf | freemangordon: will do after i got mali up :) | 21:33 |
| mkf | hm. strange, mali is apperently loaded but there is no /dev/dri/dri0 | 21:43 |
| Wizzup | mkf: why mali and not lima | 22:21 |
| mkf | yeah i meant lima | 22:21 |
| mkf | it seems /dev/dri/card0 is there but for some reason X can't open it (no such file or directory) | 22:22 |
| mkf | it even exists before X runs | 22:22 |
| sicelo | user permissions perhaps? | 22:23 |
| mkf | might be, but chown'd em to be sure and nothing changed | 22:24 |
| mkf | i've also ran X as root for test, and it can access /dev/dri/card0, but glxinfo still tells me i'm using llvmpipe | 22:24 |
| Wizzup | sicelo: re: upower, trixie is easier, but we can also do master | 23:44 |
| sicelo | sure, should be OK | 23:58 |
| sicelo | FYI, I currently have a build of upower that reports the following, https://paste.debian.net/1364641/ .... << freemangordon ;-) | 23:58 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!