IRC log for #maemo-ssu on 20161013

03:53.50*** join/#maemo-ssu povbot (~supybot@office.pov.lt)
03:53.50*** mode/#maemo-ssu [+v povbot] by ChanServ
04:31.31*** join/#maemo-ssu DocScrutinizer05 (~saturn@openmoko/engineers/joerg)
05:10.32*** join/#maemo-ssu freemangordon (~ivo@46.249.74.23)
06:00.50*** join/#maemo-ssu freemangordon (~ivo@46.249.74.23)
06:07.16*** join/#maemo-ssu arcean (~arcean@62.159.77.164)
06:14.53*** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali)
10:06.58*** join/#maemo-ssu arcean (~arcean@62.159.77.164)
10:06.59*** join/#maemo-ssu freemangordon (~ivo@46.249.74.23)
10:07.54*** join/#maemo-ssu NIN101 (~core@mail2.quitesimple.org)
10:07.54*** join/#maemo-ssu ruskie (ruskie@sourcemage/mage/ruskie)
10:07.54*** join/#maemo-ssu sailus (~sailus@salottisipuli.retiisi.org.uk)
10:07.54*** join/#maemo-ssu sixwheeledbeast^ (~zem@2a01:4f8:c17:101::2)
10:07.54*** join/#maemo-ssu merlin1991 (~merlin@Maemo/community/cssu/merlin1991)
10:09.39*** join/#maemo-ssu povbot (~supybot@office.pov.lt)
10:09.39*** join/#maemo-ssu DrCode (~DrCode@5.28.134.3)
10:09.39*** join/#maemo-ssu gregoa (~gregoa@colleen.colgarra.priv.at)
10:09.39*** mode/#maemo-ssu [+v povbot] by tepper.freenode.net
10:10.04*** join/#maemo-ssu useretai- (useretail_@gateway/shell/elitebnc/x-klnlszjxodgqhjkc)
10:10.04*** join/#maemo-ssu jurov (~jurov@ec2-52-28-28-204.eu-central-1.compute.amazonaws.com)
10:10.04*** join/#maemo-ssu _laco (~laco@static.183.80.201.138.clients.your-server.de)
10:26.55*** join/#maemo-ssu NIN101 (~core@mail2.quitesimple.org)
10:26.55*** join/#maemo-ssu ruskie (ruskie@sourcemage/mage/ruskie)
10:26.55*** join/#maemo-ssu sailus (~sailus@salottisipuli.retiisi.org.uk)
10:26.55*** join/#maemo-ssu sixwheeledbeast^ (~zem@2a01:4f8:c17:101::2)
10:26.55*** join/#maemo-ssu merlin1991 (~merlin@Maemo/community/cssu/merlin1991)
10:27.14*** join/#maemo-ssu useretai- (useretail_@gateway/shell/elitebnc/x-klnlszjxodgqhjkc)
10:27.15*** join/#maemo-ssu jurov (~jurov@ec2-52-28-28-204.eu-central-1.compute.amazonaws.com)
10:27.15*** join/#maemo-ssu _laco (~laco@static.183.80.201.138.clients.your-server.de)
10:27.21*** join/#maemo-ssu freemangordon (~ivo@46.249.74.23)
10:27.49*** join/#maemo-ssu povbot (~supybot@office.pov.lt)
10:27.49*** join/#maemo-ssu DrCode (~DrCode@5.28.134.3)
10:27.49*** join/#maemo-ssu gregoa (~gregoa@colleen.colgarra.priv.at)
10:27.49*** mode/#maemo-ssu [+v povbot] by tepper.freenode.net
10:28.08*** join/#maemo-ssu brolin_empey (~brolin@brolin.be)
10:28.08*** join/#maemo-ssu APic (apic@apic.name)
10:28.08*** join/#maemo-ssu Guest64392 (~Milhouse@host81-154-9-248.range81-154.btcentralplus.com)
10:28.08*** join/#maemo-ssu sparetire (~sparetire@unaffiliated/sparetire)
10:28.34*** join/#maemo-ssu XDS2010 (sid1218@gateway/web/irccloud.com/x-ielfcccnuchjcctl)
10:28.34*** join/#maemo-ssu jon_y (~enforcer@2001:e68:4074:46ff:e01f:8e6d:cdca:d0e5)
10:28.34*** join/#maemo-ssu amiconn (~amiconn@rockbox/developer/amiconn)
10:28.34*** join/#maemo-ssu kerio (kerio@bad.memes.vs.dank.ninja)
10:28.34*** join/#maemo-ssu atk (Arch-TK@fsf/member/Arch-TK)
10:29.01*** join/#maemo-ssu APic (apic@apic.name)
10:29.59*** join/#maemo-ssu ChanServ (ChanServ@services.)
10:29.59*** mode/#maemo-ssu [+o ChanServ] by tepper.freenode.net
10:33.01*** join/#maemo-ssu sparetire (~sparetire@unaffiliated/sparetire)
10:44.16*** join/#maemo-ssu XDS2010 (sid1218@gateway/web/irccloud.com/x-jmuecsivcytsprmp)
12:11.01*** join/#maemo-ssu how900 (~Neo900@neo900/coreteam/communications/how)
12:12.00how900moin moin
12:12.22*** join/#maemo-ssu parazyd (~parazyd@unaffiliated/parazyd)
12:12.25how900freemangordon: did you see parazyd's baby? https://files.devuan.org/devuan_jessie_beta/testing/devuan_jessie_1.0.0-beta2.rc1_armhf_n900.img.xz
12:12.32parazydmeh
12:12.38parazydit's the classic
12:13.47*** mode/#maemo-ssu [+o DocScrutinizer05] by ChanServ
12:15.50DocScrutinizer05[Notice] -asimov.freenode.net- #maemo-cssu forward channel is #maemo-ssu
12:15.54*** mode/#maemo-ssu [-o DocScrutinizer05] by ChanServ
12:19.17how900:)
12:46.46*** join/#maemo-ssu NishanthMenon (~nmenon@unaffiliated/nishanthmenon)
12:58.43DocScrutinizer05how900: won't there be an announcement for it on devuan or even on tmo?
13:06.44DocScrutinizer05parazyd: ^^^
13:07.26DocScrutinizer05also would be great to give it a try and make a video of N900 running it
13:07.28DocScrutinizer05:-)
13:08.08parazydi can do it with how900 when he comes to our office :)
13:08.26parazydshould be soon(tm)
13:12.16how900DocScrutinizer05: sure, parazyd wanted to push the images before addressing the community.  He sent an email... https://lists.dyne.org/lurker/message/20161012.203353.ad85a13e.en.html
13:13.31how900parazyd: if you write a nice howto I can test it since I have an N900 sitting here without a SIM card ;)
13:15.10parazydhow900: https://git.devuan.org/sdk/arm-sdk/blob/master/doc/quirks.md#nokia-n900
13:15.27how900parazyd: I should have known :)
13:17.36how900ok there's room for improvement :)
13:18.09bencohparazyd: "To actually run the maemo kernel using a new kernel, revert::
13:18.13bencoh" what does that mean?
13:18.38parazydbencoh: booting maemo with 4.x
13:18.47parazydpali can explain
13:18.55bencohparazyd: s/kernel// then?
13:19.16parazydyeah well, this is a paste i got from wizzup
13:19.23bencohoh okay
13:19.59bencohthis doc looks pretty useful btw
13:20.32parazydyep
13:22.11bencohoh btw .... http://pavelmachek.livejournal.com/137806.html
13:22.21DocScrutinizer05>>* Charging (seems to work)<< hmmm...
13:22.42bencohDocScrutinizer05: I'd expect it to work with bq24150 kernel driver
13:22.56bencohit shouldn't need any userland
13:23.25DocScrutinizer05ack, if there's any bq24150 driver. Is it upstream?
13:23.30bencohit is
13:23.35DocScrutinizer05great :-)
13:23.47bencoh(it has been for quite some time actually)
13:23.59DocScrutinizer05never checked :-)
13:24.27bencohI wished pavel was a tiny bit more ... verbose in his livejournal posts
13:24.47DocScrutinizer05so finally a driver based on my drafts made it into mainline \o/  ;-D
13:25.10parazydDocScrutinizer05: it works, but you have to script the led flashing yourself
13:25.17bencohlooks like he's doing some nice work but hardly gives any details :/
13:25.41DocScrutinizer05parazyd: that's pretty clear. Nobody would want the LED "hardwired"
13:25.52parazyd:)
13:25.55bencohparazyd: well, you wouldn't want to kludge bq24150 with lp55xx driving functions :)
13:26.45DocScrutinizer05parazyd: though, there are some weeeeiiird things about LEDs, where you can tell LED driver "flash when MMC access" and simliar obscure stuff
13:27.11DocScrutinizer05bencoh: ^^^
13:27.33DocScrutinizer05I never thought that was a really sustainable approach though
13:28.11DocScrutinizer05particularly it doesn't scale, nor is it really tangible for any config
13:28.23bencohDocScrutinizer05: in hw you mean?
13:28.39parazydbut you know, it's there :D
13:28.40DocScrutinizer05nono, that's a LED kernel driver property
13:28.43bencohah, that
13:29.07bencohwell if it's a generic kernel thing then that's "fine" with me :)
13:29.13DocScrutinizer05somewhere deep in /sys/*
13:29.30bencoh(whether this generic property is really relevant is another story)
13:29.40DocScrutinizer05indeed
13:29.57DocScrutinizer05I used it with N900 for MMC access and it sort of "worked"
13:30.18bencohand if there is such a property for charging then implementing it might be worth it
13:30.19DocScrutinizer05dunno how generic it is
13:31.12DocScrutinizer05also no idea how 'specified' the list of "commands" is that this feature understands
13:31.28parazydthere is no charging prop
13:31.34DocScrutinizer05I know for sure that list has no "batcharge" value
13:32.49DocScrutinizer05for sure xyou could augment the list with such "batcharge" feature, but if you can get that upstream is another question
13:33.04DocScrutinizer05I hate the whole concept
13:34.13DocScrutinizer05such stuff is typically best handled in middleware
13:34.23DocScrutinizer05like MCE ;-)
13:36.22keriois batcharge a special batman move
13:36.47DocScrutinizer05no, it's a backpack for bats
13:37.16kerioDocScrutinizer05: i take it that you've never seen /sys/class/leds/*/trigger on a raspberry pi?
13:37.27kerioit spanned like 6 lines of a 80-column terminal
13:37.28DocScrutinizer05exactly
13:37.37keriowhen listing all the possibilities
13:37.42DocScrutinizer05duh!
13:37.44keriogranted, the raspbian kernel is not mainline, but
13:38.00DocScrutinizer05ok, so it's platform specific and not written in stone
13:38.01kerioif anything should go there, battery charging should be it
13:38.15kerio...but then again
13:38.16keriowhich led?
13:38.22kerioorange is not a color
13:38.23kerio:D
13:38.32DocScrutinizer05hehehe
13:38.46DocScrutinizer05this approach sucks in so many aspects
13:38.58DocScrutinizer05as braindamaged as it gets
13:39.50DocScrutinizer05I can see you want that for really (copper) hardwired signalling that you nevertheless can enable / disable
13:40.07bencohI suspect those properties should be defined in devicetree nowadays, but not sure
13:40.17bencoh(with some defaults as well)
13:40.23DocScrutinizer05bencoh: they are not hardware
13:40.28bencohI know
13:40.52DocScrutinizer05so what got DT to do with it?
13:41.57bencohDT often defines "platform-dependent" behavior because they don't have any better place of doing so
13:42.08bencohto do so*
13:42.31DocScrutinizer05the LED signalling of charging is not platform specific on N900, that's the whole point
13:43.27bencohapparently there is such a thing as default-trigger that should come from DT, for instance
13:43.38bencohand some drivers a battery-related trigger
13:43.42bencoh+implement
13:43.49DocScrutinizer05it's a general purpose LED that can signal anything it likes. Assigning charging to it is completely arbitrary
13:44.12bencohapparently some people expect it to be board-vendor-defined
13:44.23DocScrutinizer05that's braindead
13:44.46bencohdunno
13:44.48parazydbraindead's a good miniseries
13:44.52parazydwatch it
13:45.28bencohsee Documentation/devicetree/bindings/leds/common.tx btw
13:45.38DocScrutinizer05nobody in their sane mind want vendor define the LED behavior for them
13:46.34DocScrutinizer05isn't that exactly what we want to get rid of and thus we use linux instead of a proprietary OS?
13:48.25bencohthat's actually the other way around
13:49.02DocScrutinizer05anyway as long as DT just defines the list of properties/values for /sys/*/LED*/*/trigger, I don't mind
13:50.13DocScrutinizer05this would still not explain how to make the LED shine orange.
13:52.50DocScrutinizer05a proper way to solve this stuff would be a LED "trigger" sysnode that understands a sort of handle to a (pair of) kernel events, and the charger emitting those events (it already does I think)
14:41.52*** join/#maemo-ssu qwqwqwqw (5bc18032@gateway/web/freenode/ip.91.193.128.50)
14:48.29DocScrutinizer05for led in LED:R LED:G; do echo 'kevent.bq24150.chargestart' > $led/event_on; echo 'kevent.bq24150.chargeend' > $led/event_off; done
14:53.51*** join/#maemo-ssu jon_y_ (~enforcer@2001:e68:4074:46ff:e01f:8e6d:cdca:d0e5)
15:02.09DocScrutinizer05but then, you actually rather (NOT) want something like echo 'kevent.bq24150.chargestart {engine1_load="9d80407f7f00e0047f0040007f007f007f00407f7f00e0047f007f000000;9d804000e080407fe0800000"; engine1_LEDs=000001100; start}; kevent.bq24150.chargestop [stop}' >lp5523/new_sysnode_for_this
15:03.06DocScrutinizer05so, better already do it in userland, by a daemon
15:20.31kerioDocScrutinizer05: yeah but only if you can inotify it or whatever
15:20.34keriopolling is :(
15:26.15*** join/#maemo-ssu Pali (~pali@Maemo/community/contributor/Pali)
15:33.24parazydkerio: http://z3bra.org/wendy/
15:33.49kerioty but i don't really linux anymor
15:33.50kerioe
15:33.56kerioi have seen the truth
15:51.10DocScrutinizer05kerio: knotify
15:51.33kerioanyway should i try that devuan
15:51.45kerioi don't even know what i could do
15:51.59keriolike
15:52.01DocScrutinizer05err sorry, kevents?
15:52.01kerioi have no wifi
15:52.33DocScrutinizer05https://www.freebsd.org/cgi/man.cgi?kqueue
15:54.40DocScrutinizer05https://wiki.linuxfoundation.org/networking/kevent
15:56.29DocScrutinizer05parazyd: inotify fails on sysfs
15:56.43DocScrutinizer05as well as other virtual filesystems
16:00.08DocScrutinizer05in maemo I think kerecv does that kevent stuff
16:26.47freemangordonhow900: no. which kernel it has?
16:27.21freemangordonparazyd: ^^^
16:27.37how900freemangordon: "The kernels are built with custom gcc
16:27.39how900toolchains that can be retrieved at https://pub.parazyd.cf/mirror/
16:27.47how900" see https://lists.dyne.org/lurker/message/20161012.203353.ad85a13e.en.html
16:28.02freemangordonok, but what is the kernel version?
16:28.15freemangordon4.?
16:29.55how900freemangordon: parazyd was mentioning 4.8 earlier.  I didn't test yet.
16:30.59freemangordon4.8 is ok, if it is the version includede. downloading atm, no idea when I will have time to test it
16:32.14freemangordonwonders where is his uSD card
16:32.20parazyd4.6-rc1 is on the imge
16:32.55freemangordonhmm, not good, lots of stuff missing
16:33.13parazydit was the default branch on pali's repo
16:33.17freemangordonesp in the terms of PM
16:33.26freemangordonoh, so you're not using upstream kernal?
16:33.31freemangordon*kernel
16:33.32parazydfeel free to make your own with arm-sdk: https://git.devuan.org/sdk/arm-sdk
16:33.37parazydfreemangordon: no
16:34.22freemangordonparazyd: BTW, I am not complaining, just saying ;)
16:36.32parazydyeah well, i try to do stable kernels for devuan
16:38.40freemangordonparazyd: the point is that in vanilla 4.8 there should be most of the stuff which is on Pali's tree as patches. However, I'll try to find some time in the upcoming weekend to play with that sdk
16:40.00freemangordonbut first I need to find my spare uSD card
17:04.49keriohas a 128gb uSD card
17:05.11keriofucking voodoo magic man
17:05.18keriotheres no way 128gb can be that small
17:06.29freemangordonkerio: what about 256GB?
17:06.39kerioHERETIC
17:50.12freemangordonparazyd: hmm, I see no kernel included in the image, is that expected?
18:10.33keriofreemangordon: uImage in /boot?
18:11.14kerioalso how do i use that image
18:11.21keriodo i shove it in a microsd partition and boot from it
18:53.00freemangordonkerio: iirc it should be on fat partition
18:53.30freemangordontry searching google on how to boot meego from usd
20:18.14bencohfreemangordon: btw http://pavelmachek.livejournal.com/137806.html?nojs=1 :(
20:18.40bencohdunno if you guys noticed
21:08.49*** join/#maemo-ssu xes (~xes@unaffiliated/xes)
21:28.33parazydfreemangordon: the kernel is in /root/ of the image
21:28.46parazydyou're supposed to do ubootmenu yourself
21:29.00parazydthrough maemo
21:29.12parazydif there's another way, let me know :)
21:48.45*** join/#maemo-ssu RedW (~redw@89-76-164-87.dynamic.chello.pl)
22:52.14*** join/#maemo-ssu RedW (~redw@89-76-164-87.dynamic.chello.pl)
23:23.20*** join/#maemo-ssu NishanthMenon (~nmenon@unaffiliated/nishanthmenon)

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.