<libv>
i was talking about linux-sunxi (our 3.4 kernel)
<rm>
my point is, if that's the expected pace of mainline development, then if H3 is "only for mainline" then we maybe could use it by 2020
<mripard>
rellla: I think it would be a better investment to start working on mainline directly, instead of trying to catch up with 2 years of development that never happened
<libv>
rm: mainline, or vendor kernel
<jelle>
I also like to get my orange pi pc mainline supported, but I'm new to this kind of hacking ;)
<mripard>
rellla: but you know, I'm biased
<libv>
there is nobody there to clean up the vendor kernel and to merge the stuff with linux-sunxi
<libv>
... 3.4 kernel
<mnemoc>
:(
<rellla>
rm: with audio merged, we have only 2(+1) outstanding things, before i think we have ~full support
<mripard>
rm: hey, if there's anything missing for you, we always accept patches you know :)
<rm>
sure, I am not blaming anyone
<rellla>
thanks to mripard and nove, that gave a statement to my email in ml-
<rm>
just, "too bad"
<mripard>
rm: just so that we end up having anything you want before 2020
<libv>
rm: even i have to own up to the fact that mainline has the drive and that nobody is motivated to work on sunxi-3.4
<mripard>
rellla: and honestly, I think your work on vdpau is much more valuable than the work you could do on porting the H3 on 3.4
<mripard>
rellla: but maybe that's just me :)
<rellla>
mripard: biased :) i think, mainline is the right way
<rellla>
mripard: i paused my eudyptula challenge at level 5 and read up LDD3 atm. very much new things for me :p
<rellla>
mripard: but work on libvdpau-sunxi seems to be a dead end. we have to rewrite that if at sometime display and cedrus kernel drivers get real.
<rellla>
it would be much more fun, if you know, that it is not for the recycle bin \o/
cnxsoft1 has joined #linux-sunxi
<mripard>
rellla: would things change if I can bring you a display driver ? :)
cnxsoft has quit [Ping timeout: 244 seconds]
cnxsoft1 is now known as cnxsoft
<mripard>
rellla: and it's definitely an effort I wouldn't like to be wasted.
<libv>
in-circuit uses sunxi-tools, some mali package and some kernel package
<libv>
buildroot-2014.08/package/sunxi-mali/
codekipper has joined #linux-sunxi
<NiteHawk>
the sompi is A20 - i guess that mali means they're also on a 3.4.x kernel?
<rellla>
mripard: we can only base that on mainline, if we have display AND cedrus in there (or cedrus at least a external module).
<mripard>
cedrus can be an external module, or we can try to merge it in staging
<codekipper>
they've backported loads of stuff from 3.10 into their kernel to support big little.
<codekipper>
I also had a go at A31 ontop of our linux-sunxi kernel but as much of the hardware was a bitch to work with then again I fell out of love with it
<libv>
and they use the cubieboard2 uboot target, but bump down the dram clock to 384 in a patch to uboot
IgorPecovnik has quit [Ping timeout: 264 seconds]
<libv>
(in-circuit)
<libv>
they use our sunxi kernel, and use commit 5bf12905ff1, which is an ssvb patch for a newer mali version
FlorianH has quit [Remote host closed the connection]
FlorianH has joined #linux-sunxi
<rellla>
mripard: it would not be that huge thing imho, to adapt the existing libvdpau to a newer display driver and/or cedrus. in the meantime i'll stay fixing vdpau API things and doing ugly workarounds, that are needed because of the sunxi special 3.4 display implementation ...
philippe_fouquet has joined #linux-sunxi
<rellla>
but maybe it would be better to do a complete rewrite, if sunxi-era-2.0 starts
afaerber has joined #linux-sunxi
hansg has joined #linux-sunxi
hansg has quit [Remote host closed the connection]
<mripard>
rellla: maybe, I was just suggesting easier temporary solutions :)
MaDMaLKaV has joined #linux-sunxi
f15h has joined #linux-sunxi
adj_ has joined #linux-sunxi
<libv>
is the wiki able to send out emails atm?
<rellla>
libv: i think no.
<rellla>
libv: sent myself a mail ~5 min ago, but nothing received. no error message in the wiki.
vivesing has joined #linux-sunxi
<libv>
same here
<libv>
and my musclememory seems to have forgotten the password to my account again (which i need to switch to root)
<libv>
mnemoc, Turl: ^
<NiteHawk>
keepass to the rescue... (that's what i use)
<libv>
yeah, i should set up a password manager indeed
<vivesing>
hello all i am trying to install openwrt on banana pi r1 but facing some problems...
<vivesing>
the error i get is
<libv>
especially since my crap customer loves to change passwords, and the last one has a part which sounds like part of my sunxi pass
<vivesing>
cannot setup simplefb no node found
<vivesing>
when booting the bananapie and then screen goes off
f15h has quit [Remote host closed the connection]
<vivesing>
any idea ?
Net147 has quit [Ping timeout: 246 seconds]
<NiteHawk>
vivesing: the error message seems to indicate that you're trying to use simplefb, but it isn't found. since that's provided by u-boot, make sure you have a suitable (recent enough) version and that it's configured accordingly (to actually provide a simplefb implementation)
<traeak>
2 people on this thread actually played the celviano, both liked it very much...interesting
<traeak>
sry just keeping an eye on this casio digital
<wens>
also wondering if anyone actually confirmed sunxi's uart is indeed 8250_dw, since the dw quirk (LCR bit 20) is not documented in Allwinner manuals
<mripard>
wens: we had some issues at the very beginning of the kernel port
<mripard>
when we used the plain 8250
enrico__ has quit [Quit: Bye]
<mripard>
and it was actually because of the LCR bit
<mripard>
it's why we changed it
vishnup has quit [Remote host closed the connection]
<traeak>
oops wrong irc
<traeak>
heh
cptG_ has quit [Quit: Konversation terminated!]
domidumont has joined #linux-sunxi
massi_ has quit [Remote host closed the connection]
FR^2 has quit [Quit: Connection reset by peer]
afaerber has quit [Quit: Verlassend]
ricardocrudo has joined #linux-sunxi
afaerber has joined #linux-sunxi
khuey_ is now known as khuey|away
domidumont has quit [Ping timeout: 246 seconds]
arossdotme-planb has quit [Ping timeout: 256 seconds]