belcher_ has joined #maemo-leste
belcher has quit [Ping timeout: 268 seconds]
belcher_ is now known as belcher
antranigv has quit [Quit: ZNC 1.8.2 - https://znc.in]
antranigv has joined #maemo-leste
Pali has quit [Ping timeout: 252 seconds]
l_bratch has quit [Ping timeout: 260 seconds]
l_bratch has joined #maemo-leste
kvw_5_ has joined #maemo-leste
kvw_5 has quit [Ping timeout: 240 seconds]
xmn has quit [Quit: ZZZzzz…]
xmn has joined #maemo-leste
ecs has quit [Remote host closed the connection]
ecs has joined #maemo-leste
ecs has quit [Remote host closed the connection]
Entitlement has quit [Ping timeout: 246 seconds]
ecs has joined #maemo-leste
Entitlement has joined #maemo-leste
pere has quit [Ping timeout: 240 seconds]
jonsger has joined #maemo-leste
pere has joined #maemo-leste
Pali has joined #maemo-leste
jonsger has quit [Ping timeout: 240 seconds]
inky has quit [Ping timeout: 240 seconds]
inky has joined #maemo-leste
Entitlement has quit [Ping timeout: 260 seconds]
Entitlement has joined #maemo-leste
<kek> parazyd closed an issue: https://github.com/maemo-leste/bugtracker/issues/500 (Vim is not usable inside screen / tmux sessions)
<kek> parazyd closed an issue: https://github.com/maemo-leste/bugtracker/issues/361 (Enter doesn't work in terminal if the terminal is started quickly after boot)
inky has quit [Read error: Connection reset by peer]
inky has joined #maemo-leste
uvos has joined #maemo-leste
pere has quit [Ping timeout: 260 seconds]
Oksana has quit [Ping timeout: 246 seconds]
sunshavi has quit [Remote host closed the connection]
inky has quit [Ping timeout: 246 seconds]
<sicelo> One thing spinal was great at is documentation - just been poking around leste's kernel for n900, and his work is quite impressive
sunshavi has joined #maemo-leste
<sicelo> Anyway, if I was to make an MR for the N900 kernel, it'd have to be against master? Or beowulf/devel ? parazyd
* sicelo goes to check how Beowulf is pronounced
sunshavi has quit [Remote host closed the connection]
inky has joined #maemo-leste
pere has joined #maemo-leste
sunshavi has joined #maemo-leste
R0b0t1 has quit [Ping timeout: 240 seconds]
R0b0t1 has joined #maemo-leste
<parazyd> sicelo: Against beowulf-devel would be the best
<sicelo> Thanks
<parazyd> I didn't rework it into a repo like droid4-linux yet
<parazyd> Which IMHO is nicer, with the orphan branches
<sicelo> np:-)
<sicelo> How did it go with N9? A few weeks
<sicelo> s/A few weeks//
inky has quit [Read error: Connection reset by peer]
inky has joined #maemo-leste
* enyc meows
jonsger has joined #maemo-leste
<parazyd> Didn't get to it :/
inky has quit [Ping timeout: 240 seconds]
<siceloa> ok :)
inky has joined #maemo-leste
inky has quit [Ping timeout: 246 seconds]
pere has quit [Ping timeout: 268 seconds]
inky has joined #maemo-leste
kgoetz has quit [Ping timeout: 260 seconds]
pere has joined #maemo-leste
kgoetz has joined #maemo-leste
<enyc> Hrrm, osso-xterm not have some characters in font that others do...
<enyc> xterm better hrrm
<enyc> parazyd: anyhow n900 image seems 'ok' once added -devel and custom kbd ....but....
<enyc> wifi fell off and once reconnected ipv6 autoconfigged addy no longer work
<enyc> was ok before fell over
<enyc> somehow also had clock wrong by 1hr as if ntp/timezone issue
<enyc> hrrm reboot and now ipv6 configged but not work again...
<enyc> I suspect kernel bugs... be nice if 5.4 or 5.10 LTS kernel but I appreciate thats easier said than done...
<enyc> fremantle+power k2.6.28 -> autoconfig wifi ipv6 was ok....
<parazyd> We don't have proper ip6 support yet
<parazyd> But this is unrelated to kbd, right?
<enyc> parazyd: shoultdnt need proper ip6 for basic autoconfig to be functional
<enyc> correct kbd unrelated just very useful
<parazyd> I meant in icd
<parazyd> It probably configures well in the lower layer
<parazyd> As in, I've seen it work on my droid4 WiFi on my local network
<parazyd> But this is probably just RA
<enyc> actually something not right with dns lookup
<parazyd> With IPv6? Quite possible
<parazyd> Did yo upgrade do get dnsmasq running?
<enyc> no e.g. elinks not work
<parazyd> In latest -devel I put dnsmasq to be the DNS system provider
<enyc> dnsmasq proc is running
<parazyd> Is /etc/resolv.xo
<parazyd> .conf pointing to localhost?
<enyc> resolv.conv says 127.0.0.1 i.e. legacy localhost
<parazyd> Then you have the nameservers in /var/run/resolv.conf*
<parazyd> Probably just IPv4 ns there
<parazyd> It's a "bug" we can fix
<parazyd> Thanks for letting me know
<enyc> curiously dns lookups sorted themselves out
<parazyd> :o
<enyc> parazyd: even ipv4 ns can lookup AAAA
<uvos> this reminds me that i need to get off my ass one of these days and make my custom/complex routing/filtering setup work in v6....
<enyc> as in fremantle
<enyc> ipv6 routing not reliable hrrm
<enyc> falls over on leste
<parazyd> That depends on how the /var/run resolv.confs are configured on fremantle
<enyc> network unreachable etc
<parazyd> I haven't investigated this enough yet
<parazyd> That sounds like a missing route
<parazyd> (Ip6 route)
<siceloa> ipv6 wasn't working properly on fremantle. you had to set it up manually (in fact was not even in the stock kernel iirc)
<enyc> yes gone ip ip -6 ro
<enyc> siceloa: it was reliable for me with cssju+power kernel, no manual config
<parazyd> In theory RA should work, but I dunno how this correlates to icd2
<enyc> parazyd: RA drops ipv6 ro after a while on leste
<enyc> hey ho :)
<parazyd> How could I reproduce such an environment if I don't have outbound ip6 at my house?
<parazyd> So I can test and fix this
<parazyd> Wizzup: Do we need something like libicd-ipv6 actually?
<enyc> parazyd: ideally, set up tunnelbroker.net ipv6 tunnel
<enyc> parazyd: and separate radvd
<parazyd> On my router?
<enyc> yes, *or* another machine which can do 6in4 tunnel endpoint via router dmz facility can sometimes work
<parazyd> Because I already have ip6 in my lan and it works
<parazyd> ok
<enyc> parazyd: what sort of router? does ISP provide global IPv4 address?
<enyc> I have a set of OpenWRT with a very helpful isp ;p
<enyc> /48 ipv6 and /28 legacy route =)
<parazyd> No, I have a dynamic address, but I have ipv6 connectivity between computers on my lan on openwrt
<enyc> parazyd: hangon, dynamic *global* ipv4 address, or cgnat etc?
<parazyd> global, as in my external/public IP
<enyc> ok so have global(public) ipv4
<parazyd> I don't have outbound ipv6
<enyc> and openwrt router
<enyc> in which case, tunnelbroker.net 6in4 tunnel FTW, you can have 'static' ipv6 and set up whatever to auto-point the ipv6-tunnel at current ipv4-global-address
<parazyd> ok
<enyc> I have older ipv6.he.net/tunnelbroker account with 'sage' stats
<enyc> they can blocke ipv6-IRC and ipv6-port-25-SMTP and can get this unblocked but not usually a real issue
<Wizzup> parazyd: yes we do @ ipv6, Pali said he could help
inky has quit [Ping timeout: 240 seconds]
ajr has joined #maemo-leste
ajr is now known as Guest47005
inky has joined #maemo-leste
<parazyd> ok, neat :)
<Pali> of course, I can help with IPv6, just let me know what is the issue
<Entitlement> enyc - [ [OpenWrt Wiki] IPv4/IPv6 Transition Technologies ]
<parazyd> Pali: I think we need a module like libicd-network-ipv4 but for ipv6
<Pali> in repository.maemo.org was one
<enyc> parazyd: not tried this myself but in theory you can have openwrt manage tunnelbroker ipv6 tunnel that gives you almost as good as native ipv6 on your network . Once working for other devices etc you can then start comparing fremantle leste etc
<parazyd> We don't really have proper ipv6 support in the ecosystem atm
<enyc> parazyd: indeed, though I would stress leste simple does not work as well as fremantle+cssu+power-kernel used to
<enyc> true ipv6 daemon... should support noticing radvd 'managed' network flag and handling DHCPv6-stateful-assignment client etc...
<enyc> again, OpenWRT supports stateless+stateful assignment downstream of openwrt ...
<enyc> not all networks privide dhcpv6-stateful-assignment, rely only on stateless ... also Android is exception in not supporting stateful assignment DHCPv6-client
* parazyd cancels Android
<enyc> I've not really surveyed which "desktop" linux distros do/don't tend to include DHCPv6-stateful-client ... I should check what the devuan doing in that regard, etc.
<Pali> dhcpv6 stateful client is supported basically by every OS / linux distribution, the only exception is Android
<Pali> I remember that 7+ years ago I did tests; the best support had Windows Vista+; linux dhcpv6 clients needed "hacks" / "workarounds
<Pali> after 1/2 later I did test again and NetworkManager was fixed and also new version of dhcpv6 clients were fixed
<Pali> so at least for 4-5 years every setup with NetworkManager should work out of the box and also manuall setup with existing dhcpv6 clients
<Pali> just Android stays broken and unusable
<Pali> every "solid" IPv6 network use statful DHCPv6 and not SLAAC; just to ensure that addresses are assigned in deterministic way
<Pali> lot of times this is requirement for firewall, to allow input traffic for certain ports/addresses
inky has quit [Ping timeout: 260 seconds]
inky has joined #maemo-leste
Oksana has joined #maemo-leste
brolin_empey_ has joined #maemo-leste
kek has quit [Read error: Connection reset by peer]
brolin_empey has quit [Remote host closed the connection]
kek has joined #maemo-leste
jonsger has quit [Ping timeout: 252 seconds]
R0b0t1 has quit [Ping timeout: 246 seconds]
<siceloa> Pali: since the u-boot changes were accepted upstream, how about releasing updated u-boot for Fremantle :)
<Pali> siceloa: IIRC it is not heavily tested like latest version in Extras repository
<Pali> and IIRC the only improvment is the USB tty support with ymodem and kermit
<Pali> and also onenand support
<siceloa> i think one of the things that would be nice (for Leste and others) is ability to use raw dtb, instead of having to append it in kernel. supposedly the newer one would work well for that
<Pali> well, this needs to be tested if it is working
<Pali> lot of u-boot commands and features are disabled at compile time due to big size of u-boot.bin binary
<Pali> but there are pending LTO patches for u-boot which decrease size of binary, so once they are merged, u-boot build for n900 can enable additional u-boot features
uvos has quit [Ping timeout: 240 seconds]
cr4y1 has joined #maemo-leste
cr4y1 has quit [Ping timeout: 260 seconds]