| Wizzup | tmlind: bencoh: I'll hold off with the sysctl change until we can confirm that it helps, iirc both me and tmlind still saw the work timer run | 12:06 |
|---|---|---|
| tmlind | Wizzup: ok, to test, the wakeups in powertop for an idle system should be lower at around 8 - 10 per second. and battery life should be longer naturally :) | 14:19 |
| Wizzup | tmlind: ok, is this with wifi on or off? | 14:44 |
| tmlind | Wizzup: yeah looking at it on ssh over wlan, screen blanked | 14:46 |
| Wizzup | ok | 14:47 |
| Wizzup | tmlind: powertop --sample=30 ? | 14:49 |
| tmlind | Wizzup: sure, i just used the default value | 14:54 |
| Wizzup | ok | 14:54 |
| Wizzup | I'm seeing 4.9 wakeups/seconds right now, with the value set to 5000ms (which I think was the default?) | 14:54 |
| Wizzup | maybe I need to reconnect the wifi | 14:54 |
| Wizzup | will monitor for a bit longer, jumped up some now | 14:55 |
| tmlind | Wizzup: maybe also check the p= value you're seeing with while true; do /etc/init.d/droid4-powermanagement status; sleep 60; done | 15:05 |
| tmlind | my sim card on m-l has expired, but i'm seeing numbers over 200mW.. with timer usage down, this should be way lower | 15:07 |
| * tmlind bbl | 15:08 | |
| uvos__ | tmlind: lsof | grep /dev/input/event1 ? | 15:16 |
| uvos__ | ls -l /dev/input/by-path/platform-mapphone_touchscreen-event | 15:17 |
| uvos__ | (while the display is off) | 15:18 |
| Wizzup | tmlind: suprisingly p=600 or so for me, but the wakeups are low | 15:19 |
| uvos__ | lsof | grep /dev/input/event1 is wrong here | 15:21 |
| uvos__ | elogind and xorg have the input device open again | 15:21 |
| uvos__ | Filtered Touchscreen id=7 [slave pointer (2)] | 15:21 |
| uvos__ | This device is disabled | 15:21 |
| uvos__ | its disabled in xinput | 15:21 |
| uvos__ | but xorg dosent release | 15:22 |
| Wizzup | uvos__: hmmm | 15:24 |
| Wizzup | looks like mce elogind, mce and xorg have it open | 15:24 |
| Wizzup | I thought we fixed this | 15:24 |
| uvos__ | mce dose not here | 15:24 |
| Wizzup | ok, so I guess we need to fix this first, before we look at other issues | 15:25 |
| Wizzup | # lsof | grep /dev/input/event1 | 15:25 |
| Wizzup | elogind-d 3110 root 18u CHR 13,65 0t0 444 /dev/input/event1 | 15:25 |
| Wizzup | mce 3667 root 17r CHR 13,65 0t0 444 /dev/input/event1 | 15:25 |
| Wizzup | mce 3667 3695 gmain root 17r CHR 13,65 0t0 444 /dev/input/event1 | 15:25 |
| uvos__ | note that you have to check if event1 is the ts | 15:25 |
| Wizzup | mce 3667 3696 gdbus root 17r CHR 13,65 0t0 444 /dev/input/event1 | 15:25 |
| uvos__ | it depends on probe order | 15:25 |
| Wizzup | Xorg 3857 user 25u CHR 13,65 0t0 444 /dev/input/event1 | 15:25 |
| Wizzup | Xorg 3857 3891 InputThre user 25u CHR 13,65 0t0 444 /dev/input/event1 | 15:25 |
| Wizzup | oh | 15:25 |
| Wizzup | my bad, sry | 15:25 |
| uvos__ | ls -l /dev/input/by-path/platform-mapphone_touchscreen-event | 15:25 |
| Wizzup | yeah, looks like X and elogind have it open | 15:25 |
| Wizzup | but this problem last time was because mce wasn't able to talk to X I believe | 15:25 |
| uvos__ | i can now | 15:26 |
| uvos__ | xinput shows you mce is disableding it | 15:26 |
| uvos__ | its a different problem | 15:26 |
| Wizzup | how do you check this with xinput? | 15:26 |
| uvos__ | xinput list --long | 15:26 |
| Wizzup | ⎜ ↳ Filtered Touchscreen id=7 [slave pointer (2)] This device is disabled | 15:26 |
| uvos__ | jup | 15:27 |
| Wizzup | argh :( | 15:27 |
| arno11 | Wizzup: i've got calls with sound working partially on n900 ! i succeded to get specific logs to see what's going on. hope it helps | 15:41 |
| Wizzup | arno11: great, with what sw exactly? | 15:43 |
| arno11 | using cmt_alsa and two kernel modules wich were not loaded on boot | 15:47 |
| arno11 | *which | 15:49 |
| Wizzup | ok, please document it somewhere (here or mail) | 15:49 |
| Wizzup | did you get audio both ways? | 15:49 |
| arno11 | I'm currently documenting that | 15:50 |
| arno11 | and can send you the logs per email | 15:50 |
| arno11 | working great in one way and partially in the other | 15:51 |
| arno11 | microphone on n900 works but refuse to upload/write to modem | 15:52 |
| arno11 | there are specific errors in logs for that | 15:52 |
| arno11 | but i can hear my voice in hp | 15:52 |
| Wizzup | ok, please do document | 15:57 |
| arno11 | no probs | 15:57 |
| Wizzup | I think we want to make a UCM2 for the n900 soon | 15:57 |
| Wizzup | this would mean knowing some amixer sequences to switch from/to hp and such | 15:57 |
| Wizzup | (and speaker and earpiece) | 15:57 |
| arno11 | ok cool | 15:57 |
| arno11 | oh and something is wrong with earphone | 15:58 |
| arno11 | seems missing another kernel module | 15:58 |
| arno11 | or something wrong with alsa | 15:58 |
| arno11 | anyway i'll prepare an email this evening | 15:58 |
| Wizzup | great, ty | 15:59 |
| arno11 | ttyl | 15:59 |
| Wizzup | uvos__: I wonder if the XDG change caused this | 16:00 |
| Wizzup | I am pretty sure this was not a problem very recently | 16:00 |
| Wizzup | Xorg has XDG_RUNTIME_DIR=/run/user/1000 | 16:01 |
| Wizzup | I don't see that set for mce | 16:02 |
| Wizzup | freemangordon: run-standalone does not set XDG_RUNTIME_DIR now, not a problem if it is already set in the shell it is running from though | 16:06 |
| Wizzup | freemangordon: maybe we do need the profile.d | 16:06 |
| Wizzup | freemangordon: uvos__: I placed back the profile.d file and it does not make a difference | 16:39 |
| Wizzup | wait | 16:39 |
| Wizzup | I didn't make it executable | 16:39 |
| Wizzup | didn't matter | 16:53 |
| Wizzup | freemangordon: did we miss some fix somehow I wonder | 16:53 |
| sicelo | Wizzup: hope arno will use public email, e.g. ML, so we can also follow along. my n900 with working modem arrived last week, and i hope to find time to play with it soon | 17:06 |
| Guest224 | Interesting talking about Maemo Leste version numbering: https://talk.maemo.org/showthread.php?p=1575187 | 17:11 |
| sicelo | looks like crazy idea to me | 17:14 |
| sicelo | anyway, not something I'm interested in atm | 17:15 |
| Guest224 | Long Term Support version is so far away, but interesting idea. | 17:16 |
| sicelo | i mean versioning in line with kernel | 17:17 |
| sicelo | btw, leste is considered to be maemo 7, not 6 | 17:17 |
| sicelo | harmattan (on N9 & N950) was considered to be 6 | 17:18 |
| Guest224 | some day kernel version is 7 :D | 17:18 |
| sicelo | anyway, I'm out ... | 17:18 |
| Guest224 | and right now it is 6.1 | 17:18 |
| sicelo | Guest224: you're the OP? | 17:21 |
| Guest224 | OP? | 17:23 |
| sicelo | you're teroyk from the thread? | 17:24 |
| Wizzup | Original Poster | 17:25 |
| Guest224 | Maybe ;) | 17:26 |
| Guest224 | gotta go but that wasn't reason. | 17:28 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!