belcher_ has joined #maemo-leste
belcher has quit [Ping timeout: 260 seconds]
belcher_ is now known as belcher
xes_ has joined #maemo-leste
kvw_5_ has joined #maemo-leste
kvw_5 has quit [Ping timeout: 252 seconds]
cockroach has quit [Quit: leaving]
t_rex has quit [Quit: WeeChat 3.0]
pagurus has joined #maemo-leste
<kek> parazyd created a repository: https://github.com/maemo-leste-upstream-forks/libsdl2
MartijnBraam has quit [Ping timeout: 245 seconds]
MartijnBraam has joined #maemo-leste
<parazyd> uvos: Did you compile your libsdl2 with the mesa and libglvnd from our repos?
pagurus has quit [Remote host closed the connection]
pagurus has joined #maemo-leste
Pali has joined #maemo-leste
uvos has joined #maemo-leste
parazyd has quit [Quit: parazyd]
parazyd has joined #maemo-leste
parazyd has quit [Changing host]
parazyd has joined #maemo-leste
Pali has quit [Ping timeout: 245 seconds]
<uvos> parazyd: yes i compiled it on d4
<uvos> but i have mesa with pvr patches there
<uvos> so no
<uvos> why?
<uvos> is the above in the repo?
<Entitlement> parazyd - [ libsdl2-binaries » architecture=armhf,label=armhf #1 Console [Jenkins] ]
<parazyd> In file included from include/SDL_opengles.h:33,
<parazyd> from /build/libsdl2-2.0.9+dfsg1/src/video/SDL_video.c:41:
<parazyd> /usr/include/GLES/gl.h:63:25: error: conflicting types for 'GLsizeiptr'
<uvos> hmm
<parazyd> We don't patch our mesa
<uvos> i also compiled the sdl mecurial verson
<parazyd> I guess I could try the chimaera/bullseye version
<uvos> try disabeling OPENGL support
<Entitlement> uvos - [ SDL :: View topic - Error building SDL2 [unix] ]
<parazyd> Do we really want that?
<uvos> parazyd: its probubly beifical for us on n900/mapphones, pinephone not so mutch
<uvos> as this forces sdl applciations to do gles if they can
<uvos> they often use gl by default
<uvos> we could compile without ogl on armv7 that correlates well with gles only socs
<uvos> dosent help arm64 failing tho
<Wizzup> I guess problems might occur when certain programs only do GL
<uvos> true
<parazyd> I'll first try a newer version of sdl2
<parazyd> Because our mesa is equivalent to what is in Debian unstable
<parazyd> At least header-wise
<uvos> annoying how mutch crap we have to fork
<parazyd> It'll solve itself through time
<Entitlement> uvos - [ Conflicting type declarations in GL & GLES3 · Issue #162 · KhronosGroup/OpenGL-R... ]
<parazyd> Ours seems like GLES1 though
<uvos> dident debain drop gles1 headers?
<uvos> maybe this is why
<parazyd> They're gone in beowulf and were reintroduced later
<parazyd> Perhaps, yeah
<parazyd> So I think I should try with newer sdl2
<parazyd> As opposed to adding our own patches
cr4y1_ has joined #maemo-leste
cr4y1_ has quit [Ping timeout: 260 seconds]
<Entitlement> parazyd - [ [RFC PATCH] watchdog: Adding softwatchdog - Peter Enderborg ]
<Entitlement> Wizzup - [ Funding from NGI EU - Maemo Leste ]
<buZz> Wizzup: congratz :)
<dreamer> Wizzup: woooot! \o/
<buZz> it read to me like this 'next generation' project is just reinventing Pidgin :P
<buZz> but sweet @ funding
<Wizzup> it's also about maemo integration, making a maemo ui, and the tor/wireguard parts, and the sharing parts
<Wizzup> + in depth telepathy knowledge will help with phonecalls and such too
<tmlind> cool :)
<buZz> :) very nice
<parazyd> :)
<buZz> Wizzup: oh, what i missed on that page you linked, was the name of the fund ;) you only posted the abbreviation
<parazyd> buZz: https://dapsi.ngi.eu/
<Entitlement> parazyd - [ DAPSI – Data Portability & Services Incubator ]
<Wizzup> parazyd: maybe we should fix it
<parazyd> Data Portability & Services
<parazyd> Wizzup: It's a hyperlink
<Wizzup> well we do link to it
<Wizzup> yeah
<buZz> parazyd: yeah i know, i just ment , its just a couple words, would have saved me a click ;)
<buZz> same @ NGI :P
<buZz> all abbreviation make sense once you know what they mean
<parazyd> buZz: The point is to click and get informed :) And potentially apply yourself!
<buZz> ;) right but thats just adding unneeded rabbitholes :P
<parazyd> 🐰
<tmlind> uvos: so that no console with usb charger on boot, i think the console input overruns confuse openrc somehow
<freemangordon> nice :)
<freemangordon> (funding)
ravelo has joined #maemo-leste
uvos has quit [Ping timeout: 252 seconds]
cr4y1_ has joined #maemo-leste
<enyc> hrrm
<enyc> does anybody elses' leste (on n900 preferably??) manage to ntpdate anywhere?
<enyc> consistent "ntpdig: no eligible servers" for me
<enyc> In fremantle+cssu I have some panel app for "update time now" or so
<enyc> yes, "sync-time-now-widget"
<Wizzup> Entitlement: I have that problem on my n900 too
<enyc> Wizzup: hrrm does rdate work perchance?
<Wizzup> haven't tried
Pali has joined #maemo-leste
<enyc> hrm rm thats' interestingy ,my older cssu is not connecting to AP any more, with newer WPA2+WPA3 enabled access point OpenWRT ...
<enyc> when leste does connect
uvos has joined #maemo-leste
ollieparanoid[m] has quit [Ping timeout: 245 seconds]
ollieparanoid[m] has joined #maemo-leste
<horseshoecrab> enyc: ditto
<enyc> horseshoecrab: to which?
<horseshoecrab> the ntpdate thing
<horseshoecrab> but i was being an idiot
<horseshoecrab> i wasnt passing any args, and now i pass a server it seems to work.
<enyc> horseshoecrab: what server(s) work for you?
<horseshoecrab> i just used the vlan interface of my switch
<enyc> horseshoecrab: hrrm aah like .. by ipv4 address explicitly, no hostname ?
<horseshoecrab> pool.ntp.works
<horseshoecrab> *pool.ntp.org
<enyc> not for me ;-(
<horseshoecrab> well it doesnt return "no eligable servers"
<enyc> horseshoecrab: do you have n900? do you have beowulf-devel enabled ?
<horseshoecrab> i get:
<horseshoecrab> sudo ntpdate pool.ntp.org
<horseshoecrab> [sudo] password for crab:
<horseshoecrab> {"time":"2021-04-24T17:40:44.385643-0100","offset":-18.526743,"precision":0.076484,"host":"pool.ntp.org",ip:"85.199.214.99","stratum":1,"leap":"no-leap","adjusted":false}
<horseshoecrab> i do have n900
<horseshoecrab> im not sure about beowulf devel though
<horseshoecrab> i just got beowulf in sources.list
<horseshoecrab> does that mean im not on the bleeding edge ?! /o\
<horseshoecrab> seriously, whats beowulf-devel? im new to maemo-leste and devuan. i never thought id be using a systemd free distro but i like it.
<horseshoecrab> *again
<horseshoecrab> (not counting the several ancient frankendebians that i never quite managed to properly dist upgrade...)
<parazyd> enyc: There's no difference to the ntp setup between main and devel
<parazyd> enyc: Have you perhaps tried to use chrony?
<parazyd> enyc: Also note there is ntpdate and ntpdate-debian
<enyc> parazyd: no, i'm aware that there are alternatives but this strikes me as odd =)
<parazyd> enyc: Can you try forcing ipv4 ?
<parazyd> ntpdate -4
<enyc> parazyd: hrrm its' now failing in a different way ;p
<enyc> with gai related trackback in the code or so
<enyc> -4 now doesn't help but ntpdate and an ipv4 address, is what works
<Entitlement> parazyd - [ ntp tweaks · Issue #379 · maemo-leste/bugtracker · GitHub ]
<parazyd> Wizzup and I wanted to try out chrony, but didn't really get around to it
<enyc> I've had chronyd working fine
<enyc> on other systems
<parazyd> What's mostly important is that it doesn't do wakeups and is kind to power usage
<enyc> parazyd: tbh i'd be happy with a time-sync-widget like had on cssu ....
<parazyd> Is it open source?
<enyc> then look into time sync fun once updated to deb11 based
<parazyd> We can put it in extras
<enyc> parazyd: its very simple ... not 100% sure how to tell spec, https://talk.maemo.org/showthread.php?t=72991
<Entitlement> enyc - [ [Announce] sync-time-now-widget for N900 - maemo.org - Talk ]
<parazyd> ah it just runs rdate
<enyc> yes very very simple, synchronous (i.e. can make ui hang)
<enyc> maybe add an & =)
<parazyd> It's written in C
<parazyd> Let me clean it up and make a package
<enyc> yay! =0
<horseshoecrab> i ust tested with ipv6 and ntpdate works for me with a dns name that only returns AAAA
<enyc> horseshoecrab: default package? apt installed a specific ntpdate ?
<horseshoecrab> just the default
<horseshoecrab> i used the maemo-leste current version from a little while back
<horseshoecrab> and just apt updates since then
<horseshoecrab> ive updated it through the dns / resolv.conf fix.
<horseshoecrab> hmmm.
<horseshoecrab> actually now you say that, i might have installed ntp
<horseshoecrab> nah don't have /etc/ntp.conf
<horseshoecrab> so im fairly sure this is stock maemo-leste
<horseshoecrab> i dont have ntpq which frustrates and annoys me a little bit...
<horseshoecrab> because im not sure what ntp server it uses
<horseshoecrab> but i suspect its the switches vlan interface via dhcp if resolv.conf / dhcpcd or whatever is doing what its told
<horseshoecrab> i wish i knew why i have to set a static v6 for this dude
<horseshoecrab> and it doesnt get anything via slaac.
<horseshoecrab> i spent so long puzzled over that before i just decided to configure a static one day and since then its been rock solid.
<horseshoecrab> but the strange thing is: i dont think this is a maemo-leste specific issue.
<horseshoecrab> i had that behavior when i was running vanilla deb from a debootstrap with one of palis kernels i believe.
<horseshoecrab> whats the n900s place in the ecosystem? is it still the reference hardware or now getting squeezed into the shadows?
<kek> parazyd closed a pull request: https://github.com/maemo-leste/status-area-applet-battery/pull/5 ([do not merge yet] Port to newer upower api, dh porting, deprecate old glib api)
cockroach has joined #maemo-leste
<parazyd> enyc: The widget is in extras. I didn't bother making it more "secure". Basically it'll run any `rdate` in $PATH, but it seems to work I think.
* parazyd now upgrades jenkins
<enyc> horseshoecrab: for static v6 the ideal is that a ipv6 libicd module whatnot, dhcpv6 stateful client, and so-on ...
<enyc> 21:50 < parazyd> Pali: I think we need a module like libicd-network-ipv4 but for ipv6
<enyc> horseshoecrab: for me the SLAAC happens on leste but unreliable
<enyc> parazyd: hrrm, new time sync widget seems to jam.... "hildon-status-menu is not respenging. Close application?"
<enyc> parazyd: what rdate command is called exactly, I want to test manually...
<Entitlement> parazyd - [ sync-time-now-widget/sync-time-now-menu-widget.c at master · maemo-leste-extras/... ]
<enyc> parazyd: and the hildon menu runs as *root* ??
<parazyd> dsme spawns it, as user IIRC
<enyc> parazyd: in any case... 'sync clock now' not working for me ... in cli I have to as root 'rdate -4 pool.ntp.org' to get far, it seems to be failing with native ipv6 pool entries by dns or NAT64 to ipv4 ntp pool entries
<enyc> without root/sudo Operation not permitted
<parazyd> Yeah I guess it doesn't work
<parazyd> Perhaps a sudoers entry for rdate in this case could be added
<enyc> just checking on fremantle
<enyc> maybe there panel runs as root for some reason
<enyc> uerr hildon-status-menu ar whatever it is
<parazyd> Yeah /usr/bin/hildon-status-menu
<enyc> parazyd: hrrm fremantle "user" can use 'rdate' which in that case is provided by busybox
<parazyd> Then it's suid probably
<enyc> not that i can see....
<parazyd> Then I dunno
<enyc> oh yes sorry it is
<enyc> confusing
<enyc> there is an 'sbin' rdate pointing to busybox
<enyc> but is al/so /usr/bin/rdate which *is* setuid
<enyc> users' path uses that separate /usr/bin/rdate
<parazyd> haha
<parazyd> Yeah then I guess we need a sudoers entry for this
<parazyd> Mind doing it?
<enyc> old openrdate package not in debian package database
<parazyd> It's just rdate
<enyc> fremantle cas "openrdate" nontheless
<enyc> parazyd: I could try, but not up on git or sudo atm
<enyc> also I note pool.ntp.org. is not (supposed) to be directly used, supposed to register a subzone and so on but I guess leste not big yet ;p
<enyc> in any case reviewing the underpinning command used would be snsebile i think
Guest47005 is now known as t_rex
<parazyd> ok let me fix it
<enyc> sudo rdate -n -s 2.debian.pool.ntp.org may be more sensible.... testing with nat64 ...
<enyc> ok nat64 works with rdate on newer debian not leste. There is some problem with leste and ipv6 routing.
<enyc> that is not an rdate problem
uvos has quit [Ping timeout: 265 seconds]
<enyc> in short, sudo rdate -n -s 0.debian.pool.ntp.org points to a zone which does not yet declare AAAA pointers, except where downstream NAT64 fabricates them.
<enyc> whereas, sudo rdate -n -s 2.debian.pool.ntp.org will allow ipv6 too and I might suggest use of but leste needs newer n900 kernel or whatever to start unbreaking the auto-ipv6 (unreliable at present...)
<parazyd> Added the sudoers entry, you can upgrade
<parazyd> As for IPv6, nothing I can do atm
<enyc> parazyd: sure, my biggest suspicion is kernel ... In any case curious what will be needed to move n9xx-kernel to a debian LTS kernel e.g. 5.10 series =)
<enyc> iirc there was some problem with some driver not upstreamed or so
<Wizzup> I think icd2 needs support for it
<Wizzup> don't think it's driver related
<enyc> Wizzup: no, this is additional problem
<Pali> maemo extras has ipv6 icd plugin
<enyc> routing unreliable etc. even with autoconfi address
<enyc> as per fremantle, where it was reliable even with automatic address and no usermode support
<Entitlement> Wizzup - [ maemo.org - package overview for Libicd-network-ipv6 ]
<enyc> some sort of intermittent ipv6-packet/routing blackhole occurs on leste even if kernel autoconfig address does come in
<Wizzup> it's in non-free
<Wizzup> there is this, but I think that's kernel related mostly? http://maemo.org/packages/view/ipv6-support/
<Entitlement> Wizzup - [ maemo.org - package overview for IPv6 support ]
<Wizzup> or does it rely on the closed source ipv6?
<Entitlement> Wizzup - [ Reverse engineering libicd-network-ipv6 · Issue #139 · maemo-leste/bugtracker · ... ]
<Entitlement> Pali - [ IPv6 support for Nokia N900 (version 0.2) - maemo.org - Talk ]
<Wizzup> I think that's linked from that issue
<Pali> I will try to explain state of ipv6 support...
<Pali> * obviously you need kernel with ipv6 support, stock nokia did not have, kernel-power was needed
<Pali> * for address assignment there are multiple independent options (and can be combined): static/manual, RA/SLAAC, DHCPv6-statefull (assigns one address, but can also more are possibles), DHCPv6-PD (assigns prefix of addresses)
<Pali> * for gateway/router assignment there is only one option: RA
<Pali> * for DNS server assignment there are multiple independent options and can be combined: static/manual, RA, DHCPv6-stateless, DHCPv6-statefull (part of address asignment)
<Pali> RA/SLAAC part is implemented in kernel, just needs to be enabled via sysctl and kernel automatically assigns SLACC address to interface
<Pali> for DNS server from RA, kernel exports it via netlink interface, so userspace can read it and overwrite /etc/resolv.conf
<Pali> all DHCPv6 modes are implemented in userspace and there are more implementations
<Pali> RA packet also inform about existance of DHCPv6 and also which modes of DHCPv6 are supported... so userspace needs to listen for RA packets, parse them and start DHCPv6 as required
<Pali> in maemo fremantle as DHCPv6 client was used "wide" and for detecting of DHCPv6 mode was used custom nokia c application (open source)
<Pali> integration of starting this application and dhcpv6 client into icd was done by that close source plugin
<Pali> DHCPv6-PD mode was not supported
<Pali> in maemo there is ipv6-support which depends on all those required software, plus in its scripts it enable (disabled) ipv6 support in random nokia/fremantle applications
<Pali> I have written software daemon which listen for RA packets and when "state" changes it can start script with info in env what was changed
<Pali> it can be used e.g. to automaticaly start dhcpv6 client (in required mode), or updating /etc/resolv.conf
<Pali> daemon puts all required data to env variables prior spawning user script
<Pali> I can provide you this my daemon (under GPL) if you think it can be useful
<Pali> but if you are going to strictly mimic libicd-network-ipv6 behavior with nokia open source c application (osso-rtsol) and helper scripts around in libicd-network-ipv6-script then I think you do not need it
<parazyd> Is it viable to RE libicd-network-ipv6 ?
<Pali> I do not know details of icd daemon
<Pali> for sure you need some plugin which integrates it... but if writing it from zero is better and easier then RE this binary I'm not sure
<parazyd> Ah Wizzup already opened a ticket https://github.com/maemo-leste/bugtracker/issues/139
<Entitlement> parazyd - [ Reverse engineering libicd-network-ipv6 · Issue #139 · maemo-leste/bugtracker · ... ]
<Wizzup> maybe we can add the above info to the ticket
<parazyd> It's 18K
<parazyd> The binary
<Wizzup> we can RE it, but we have other stuff to do first (but someone else can do it)
jonsger has joined #maemo-leste
ravelo has quit [Quit: Connection closed for inactivity]
cr4y1_ has quit [Ping timeout: 240 seconds]
jonsger has quit [*.net *.split]
cockroach has quit [*.net *.split]
pere has quit [*.net *.split]
sicelo has quit [*.net *.split]
_blasty` has quit [*.net *.split]
Armen has quit [*.net *.split]
ikmaak has quit [*.net *.split]
freemangordon has quit [*.net *.split]
DPA has quit [*.net *.split]
brolin_empey has quit [*.net *.split]
vectis_ has quit [*.net *.split]
meridion has quit [*.net *.split]
rafael2k has quit [*.net *.split]
thefinn93 has quit [*.net *.split]
tvall has quit [*.net *.split]
fLegmatik has quit [*.net *.split]
__20h__ has quit [*.net *.split]
ollieparanoid[m] has quit [*.net *.split]
MartijnBraam has quit [*.net *.split]
kgoetz has quit [*.net *.split]
[TheBug] has quit [*.net *.split]
nohit has quit [*.net *.split]
NIN101 has quit [*.net *.split]
l_bratch has quit [*.net *.split]
jameshjacks0njr has quit [*.net *.split]
MinceR has quit [*.net *.split]
enyc has quit [*.net *.split]
Kabouik has quit [*.net *.split]
danielinux has quit [*.net *.split]
linmob has quit [*.net *.split]
xes_ has quit [*.net *.split]
devil has quit [*.net *.split]
panzeroceania has quit [*.net *.split]
kvw_5_ has quit [*.net *.split]
mrgeanie has quit [*.net *.split]
avoidr has quit [*.net *.split]
RedW has quit [*.net *.split]
Jasper[m] has quit [*.net *.split]
cato`[m] has quit [*.net *.split]
Pali has quit [*.net *.split]
belcher has quit [*.net *.split]
tmlind has quit [*.net *.split]
Entitlement has quit [*.net *.split]
Wizzup has quit [*.net *.split]
mighty17 has quit [*.net *.split]
afontain_ has quit [*.net *.split]
lisq has quit [*.net *.split]
horseshoecrab has quit [*.net *.split]
buZz has quit [*.net *.split]
globbot has quit [*.net *.split]
bencoh has quit [*.net *.split]
avoidr has joined #maemo-leste
RedW has joined #maemo-leste
freemangordon has joined #maemo-leste
kvw_5_ has joined #maemo-leste
DPA has joined #maemo-leste
pere has joined #maemo-leste
vectis_ has joined #maemo-leste
brolin_empey has joined #maemo-leste
_blasty` has joined #maemo-leste
rafael2k has joined #maemo-leste
sicelo has joined #maemo-leste
jonsger has joined #maemo-leste
cockroach has joined #maemo-leste
ikmaak has joined #maemo-leste
Armen has joined #maemo-leste
thefinn93 has joined #maemo-leste
meridion has joined #maemo-leste
Jasper[m] has joined #maemo-leste
cato`[m] has joined #maemo-leste
buZz has joined #maemo-leste
globbot has joined #maemo-leste
bencoh has joined #maemo-leste
afontain_ has joined #maemo-leste
mighty17 has joined #maemo-leste
lisq has joined #maemo-leste
horseshoecrab has joined #maemo-leste
afontain_ has quit [Ping timeout: 240 seconds]
Entitlement has joined #maemo-leste
belcher has joined #maemo-leste
Pali has joined #maemo-leste
Wizzup has joined #maemo-leste
tmlind has joined #maemo-leste
__20h__ has joined #maemo-leste
fLegmatik has joined #maemo-leste
tvall has joined #maemo-leste
Jasper[m] has quit [Ping timeout: 248 seconds]
mrgeanie has joined #maemo-leste
xes_ has joined #maemo-leste
devil has joined #maemo-leste
panzeroceania has joined #maemo-leste
[TheBug] has joined #maemo-leste
kgoetz has joined #maemo-leste
Entitlement has quit [Ping timeout: 295 seconds]
mighty17 has quit [Ping timeout: 267 seconds]
kgoetz has quit [Ping timeout: 272 seconds]
venji10[m] has quit [Ping timeout: 258 seconds]
nohit has joined #maemo-leste
linmob has joined #maemo-leste
l_bratch has joined #maemo-leste
enyc has joined #maemo-leste
MinceR has joined #maemo-leste
NIN101 has joined #maemo-leste
Kabouik has joined #maemo-leste
danielinux has joined #maemo-leste
tvall has quit [Ping timeout: 276 seconds]
t_rex has quit [Ping timeout: 258 seconds]
panzeroceania has quit [Ping timeout: 287 seconds]
cato`[m] has quit [Ping timeout: 248 seconds]
fLegmatik has quit [Ping timeout: 276 seconds]
Entitlement has joined #maemo-leste
jameshjacks0njr has joined #maemo-leste
panzeroceania has joined #maemo-leste
uvos has joined #maemo-leste
<uvos> parazyd: did you manage wrt. sdl2?
<parazyd> No, got sidetracked. But I planned it for the morning.
<parazyd> Will ping you asap
uvos has quit [Remote host closed the connection]
uvos has joined #maemo-leste
<uvos> not to worry i wont make use of it today anymore anyhow :)
<parazyd> Issued a build now
<parazyd> Will see if it passes tomorrow :)
uvos has quit [Ping timeout: 248 seconds]
MartijnBraam has joined #maemo-leste
ollieparanoid[m] has joined #maemo-leste
fLegmatik has joined #maemo-leste
cato`[m] has joined #maemo-leste
Jasper[m] has joined #maemo-leste
mighty17 has joined #maemo-leste
kgoetz has joined #maemo-leste
venji10[m] has joined #maemo-leste
afontain_ has joined #maemo-leste
t_rex has joined #maemo-leste
tvall has joined #maemo-leste
jonsger has quit [Quit: jonsger]
Entitlement has quit [Ping timeout: 260 seconds]
<enyc> parazyd: my IPv6 concern (not sure been understood...) is that there is something else odd going on in Leste, not related to lac of ipv6-daemon
<enyc> Pali: ^^ i meant ;p
<enyc> something funny goes on with v6 packets getting blackholed somehow, as if kernel bug or so, even if v6 address + route SLAAC'ed okay
Entitlement has joined #maemo-leste
Barabacha has joined #maemo-leste