libera/#maemo-leste/ Saturday, 2025-04-26

freemangordonWizzup: strip? it is build with -O207:28
freemangordonand no '-g'07:29
freemangordonok, on a33 with chimaera, the same test program uses 58280 RES08:16
freemangordon76704 RES on n900 and 81632 RES on d4 with daedalus08:21
freemangordonWizzup: ok, what I see as a major strace difference chimaera vs daedalus, is that chimaera uses lots of lstat64 calls when looping through theme images, while daedauls uses readlink08:39
freemangordonsee https://paste.debian.net/1371648/08:48
arno11freemangordon: those readlink lines are also the main diff between strace running with calculator (stock) vs calculator (qt5ct tweak) FYI09:23
arno11*on daedalus09:24
freemangordonarno11: by saying "diff" do you mean they are absent or that lstat is used instead of readlink?09:25
arno11i mean they are absent09:26
arno11hmm there are not absent in fact, just a lot more without tweak09:32
arno11*hundreds more09:32
freemangordonok, have to run now, but later on will put some traces in gtk/maemo style plugins, to see who does that09:43
arno11ok09:46
arno11Wizzup: able to get ringtone working: the problem was only @idle, screen off. renicing dbus-daemon @-5 (like fremantle) solves the issue13:07
arno11but still the sphone crash if you don't click on 'mute ringing' before answering13:09
arno11so it seems to be 2 different issues13:09
Wizzupseems like a race condition if renice matters13:15
arno11hm seems -5 is not enough @100% of time, -8 is fine13:19
arno11yeah13:19
WizzupI'm worried that just renicing something solves problems13:20
WizzupWe really ought to fix any underlying problems13:20
Wizzupsurely dbus-daemon isn't soo slow that it cna't deal with a few things in 5-10 seconds13:20
arno11ok13:21
arno11at least, renice could be useful for sicelo to avoid missed calls :P13:24
uvosarno11: im working on fixing the sphone crash15:59
uvosnot that this will help you15:59
uvossince after the fix sphone will still exit, it just wont assert, since its in an situation it cant recover from15:59
arno11ok that's fine16:00
arno11btw, any idea why if i click on 'mute ringtone' before answering, it avoids the crash ?16:01
uvossphone crashes because libpulse reports that it lost its connection to the pa server16:02
uvosmuteing ringtone causes the hifi stream to be idle when the ucm profile gets switched16:02
uvospresumably this avoids pa breaking somehow16:03
uvospresumably what is happening is that pulseaudio crashes when a voice call is entered while the hifi stream is active16:03
uvospulseaudio crashing then takes down sphone with it16:03
arno11ah ok makes sense16:07
arno11could it be possible to add a function to mute ringtone when clicking on 'answer' to avoid this kind of situation ?16:09
uvosno16:09
uvosswitching proflile while the steam is active is fully valid16:10
uvostheres a bug somewhere that needs to fixed16:10
Wizzupyes, as I also stated, pa fails on my d4 regularly in asserts16:11
WizzupI would not be surprised if this is the same on n90016:11
Wizzupand this gets logged to /var/log16:11
arno11i didn't see any other troubles atm but i didn't try mpv (which is usually a bit buggy on n900)16:12
arno11but yeah maybe something a bit different in PA 16, no ?16:12
Wizzuplooks like the ai crawler have found our git.maemo.org and are crawling the wholdr droid4-linux repo16:13
Wizzupso useful16:13
arno11oh cool16:15
arno11wow lot of failures in /var/log @PA16:17
arno11*indeed16:17
arno11i wonder if our default.pa is correct btw16:19
Wizzupwhat kind of failures16:22
arno11[pulseaudio] module.c: Module "module-switch-on-port-available" should be loaded once at most. Refusing to load.16:23
arno11*when sphone crashes @ringtone16:23
Wizzupthat does not seem like a failure16:25
arno11sorry was not talking about failures i see, just PA msg @crash16:25
arno11i.e @failure:16:26
Wizzupso you're saying there is no crash message logged?16:26
Wizzupif my PA crashes I see 'assertion failed... aborting'16:26
Wizzupor similar16:26
Wizzupser16:26
arno11ok, no 'assertion failed' for me, but let me double check16:27
arno11indeed, no similar failure on my device16:27
arno11at least in syslog, maybe visible elsewhere ?16:28
arno11ah i remember..yeah i see similar failure in sphone logs16:31
arno11'assertion...failed at../src/pulse/thread-mainloop.c:166, function pa_threaded_mainloop_stop(), Aborting16:32
WizzupI think it's user.log but not sure on top of my head16:35
siceloWizzup: hoping your battery woes are over?16:46
Wizzuplooks like it so far16:57
sicelolovely :-)16:57
Wizzupquite insane how many /16's I had to block to get rid of the overly aggressive crawlers17:35
siceloyes those are a real nightmare17:44
freemangordonWizzup: seems realpath() implementation in glibc has changed20:04
freemangordonhttps://public-inbox.org/libc-alpha/c2a203a5-5817-b0f4-3f56-1ffca9f971da@cs.ucla.edu/T/20:04
freemangordonthis is called several times by sapwood engine for each image20:05
freemangordonhttps://git.maemo.org/leste/sapwood/src/commit/e8f249b8a26075467a14d0d93966c6f590611181/engine/theme-pixbuf.c#L18820:10
freemangordonhowever, to me it seems we have an issue with fs on n90020:11
freemangordonthat code should not take that much time, even for 1000 files20:12
Wizzupyeah, I doubt that this is the problem20:14
Wizzup@ realpath20:14
Wizzupmaybe we can revisit some tunables that we may have set in the past in sysctl - I don't think we do that any more, right20:14
freemangordonlike what?20:15
freemangordonalso, I may implement some filename caching in sapwood, but not sure what side effects it may have20:16
freemangordonI wonder if we have some kernel issue, or dunno20:19
Wizzupfreemangordon: don't know, I just remembered arno tried some things, but if this is a fs regression somehow it'd be kernel side20:19
Wizzupbut that should be the same on chimaera20:19
freemangordonhaving swap on emmc should not slow down the system20:20
Wizzupit could be we have slightly more ram pressure and suddenly vfs is not ok? no idea20:20
freemangordonor that, yeah20:20
freemangordonanyway, calling that code 1000 times should not take 3 seconds, even on n90020:20
freemangordon(1000 being some arbitrary value)20:21
freemangordonok, the real value is 40920:26
freemangordonWizzup: how to install the new upower?20:31
sicelojust apt update20:34
siceloin fact, dpkg -l upower ... what does yours return?20:34
freemangordonupower         1:0.99.11.1-1+m720:36
freemangordonso, this one has epoch20:37
freemangordonI don;t think we have upower in the repos20:37
arno11Wizzup: i doublechecked, no more custom sysctl stuff or similar in daedalus and chimaera-devel. still some weird stuff only in stock chimaera20:42
freemangordonhow to check on the device?20:42
arno11you mean for sysctl or upower ?20:43
freemangordonsysctl20:44
arno11check in /etc/sysctl.d20:44
arno11for n900 perf something20:44
arno11the file should be empty20:44
freemangordonthere is only a symlink20:47
arno11yeah so that's 'fine'20:48
freemangordonmhm20:48
arno11brb20:49
freemangordonWizzup: btw, noatime has huge impact on n90020:55
arno11on daedalus ? i use it and don't feel too much diff21:00
arno11but maybe depending of sdcard21:00
freemangordonmine is fast, on theory21:01
freemangordon*in theory21:01
arno11ok21:01
arno11what kind of improvement you see btw ?21:03
arno11(just curious, to test)21:03
freemangordonssh console session becomes way more responsive21:03
arno11ah ok, interesting (i don't use it most of the time)21:04
siceloWizzup: another leste-config MR ... to update the hwclock once ntpdate has updated system clock - https://git.maemo.org/leste/leste-config/pulls/6121:46
sicelobeen driving me mad for some time21:46
freemangordonwhat the "-rw-r--r-- 1 root root 102598896 Jan  3  2023 /usr/lib/arm-linux-gnueabihf/libLLVM-15.so.1" ?!?21:49
uvoswhats wierd about that?21:55
uvosyou suprised llvm is big? or that you have an old version installed?21:55
freemangordonbig21:58
sicelofreemangordon: upower - weird, it's definitely here, https://maedevu.maemo.org/leste/pool/main/u/upower/  ... but indeed i can't find with `apt-cache policy upower`22:00
uvosfreemangordon: llvm is kinda amazingly small given what it supports22:01
freemangordonuvos: ok, won't argue, it is just that 100MB so seems kind of oversized to me :)22:01
uvosi suspect we could make it smaller by limiting the targets we compile in, theres like 50 gpu isa backends in there that are known to be pretty large22:02
uvoswe obv dont need to compile shaders for random desktop gpus..22:02
freemangordonanyway, I don;t think it is related to the issue we see on n90022:02
uvossure only a small part of that should end up in ram on n90022:03
freemangordonthere is some general slowness, it feels like device runs on 125 MHz all the time22:03
freemangordonbut that's not the case, for sure22:03
uvosi presume the other obvious case is also not present22:04
freemangordonit seems that QT (or general) memory usage has increased22:04
uvos(too mutch memory pressure)22:04
freemangordonyeah, perhaps we hit that22:04
freemangordonas my test program startup time varies between 3s and 10s22:07
arno11hmm, on my device, i don't see too much diff in term of memory usage, neither general slowness22:09
uvosyou run with stuff turned off no?22:10
arno11only for qt22:10
uvosok22:10
uvoslike if your desperate for ram you could drop conversations/tp/vcm and re enable sphones messageing, at the cost of ritch messaging support ofc.22:11
uvosbut that should save quite the chunk22:11
arno11conversation + vcm etc don't use too much ram honestly22:12
arno11and new converstions_slim is nice22:12
freemangordonusing maemo-launcher strips 1 second off startup time of the test app22:12
freemangordonso it start in 2700ms22:13
arno11freemangordon: i'm a bit surprised your device seems slow. i mean my daedalus seems smoother and faster (apart qt) thand fremantle22:16
arno11*than22:16
freemangordonwell, maybe I access it mostly over ssh22:22
freemangordonand wifi latency seems big22:22
arno11ah ok22:27
arno11btw i doubt ram or swap usage is a problem. at least not more than chimaera (see tg-desktop working)22:36
arno11and at least with sd card swap22:37
Wizzupfreemangordon: I think we have upower withotu epoch now22:44
freemangordonWizzup: where? https://paste.debian.net/1371784/22:51
freemangordondo I miss a repo?22:52
siceloah, i see ... you don't have -devel enabled23:17
siceloi also didn't until just now .. the correct upower does show up now23:17
freemangordonoh, I didn't know there is daedalus-devel repo :)23:29

Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!