<mnemoc>
it's absurd them all include info about how to install a toolchain for example
<mnemoc>
techn_: i totally agree tutorials should all point to the main repos, not people of board specific forks
<mnemoc>
but also be properly splitted by topic
<mnemoc>
making buildroot is not cubieboard-only
<mnemoc>
using livesuit is not buildroot only
<mnemoc>
main problem with livesuit support at the momemnt is that we don't have a u-boot branch that hooks correctly between boot0/boot1 and our kernel
<mnemoc>
as we require (optional in 3.0, but default) the bootloader to pass proper machine type and mem info
<mnemoc>
i was wondering if we should fork (and maintain) the allwinner-tools repo
<mnemoc>
there is a patch for fixing the kernel driver livesuit for linux needs lingering around
<lkcl>
techn_: oh. that's funny. everyone complained when i insisted that people be a bit more rigorous about the rhombus tech wiki.
<lkcl>
hm
<lkcl>
they didn't like that, so moved to linux-suni instead. now you're saying that linux-sunxi is disorganised. funny that.
Svet has joined #arm-netbook
<mnemoc>
lkcl: the reason was not organization, was scope. for RT the A10 is merely the SoC to use in the first EOMA card, linux-sunxi is 100% decided to these chips in particular
<mnemoc>
lkcl: and as you did with A10-specific info we are now suffering a flood of interest bound to the cubieboard
<mnemoc>
lkcl: having pages about the hackberry in RT's wiki is kind of.... weird
discopig has quit [Ping timeout: 276 seconds]
mysteryname has joined #arm-netbook
<mnemoc>
miniand can't have less in common with RT
<mnemoc>
lkcl: the split was intended to keep eoma's wiki/mailing list from been taken over by A10 fans
<mnemoc>
lkcl: any realistic timeframe for us to be able to fund the mini engineering eoma68 board, and open hardware a10/imx6/exynos5 card?
cat_x301 has joined #arm-netbook
gzamboni has quit [Ping timeout: 240 seconds]
mnemoc has quit [Excess Flood]
penguin42 has joined #arm-netbook
cat_n9 has quit [Ping timeout: 246 seconds]
mnemoc has joined #arm-netbook
cat_x301 has quit [Ping timeout: 265 seconds]
cat_n9 has joined #arm-netbook
cat_x301 has joined #arm-netbook
ZaEarl_ is now known as ZaEarl
WarheadsSE has joined #arm-netbook
vgrade has quit [Remote host closed the connection]
cat_n9 has quit [Read error: Connection reset by peer]
cat_x301 has quit [Ping timeout: 246 seconds]
popolon has joined #arm-netbook
IEF has quit [Read error: Operation timed out]
IEF has joined #arm-netbook
cat_n9 has joined #arm-netbook
ibot has quit [Ping timeout: 256 seconds]
cat_n9 has quit [Ping timeout: 246 seconds]
tuliom has joined #arm-netbook
rellla has joined #arm-netbook
rellla has quit [Client Quit]
cat_n9 has joined #arm-netbook
<mnemoc>
Turl: ping
<Maqs>
what's the difference between "stage/" and non-stage?
<Turl>
mnemoc: pong
<Turl>
hard to beat ~30s latency on a saturday morning :P
<mnemoc>
Maqs: stage has patches that will go to non-stage after tested
<Maqs>
ok, thanks
<mnemoc>
Turl: i want to make a "sunxi-bsp" repo, to help people build our stuff, also android's
<mnemoc>
also, did you manage to test the stage branches on android?
<Turl>
what would you find on this bsp repo?
<Turl>
mnemoc: I didn't test stage per se, but I've been running with mali_next since like forever and I have cpufreq in the mix now too
<Turl>
I'll kick a build with stage/3.0 later today though to test the display stuff
<mnemoc>
submodules to the other repos, and a couple of Makefiles and scripts to do the stuff
<Turl>
kick off*
<Turl>
mnemoc: you cannot really intend to do android as submodules :P
<mnemoc>
Turl: only it's head
<mnemoc>
Turl: but anyhow, that's where I need your help :)
<mnemoc>
`make zatab android` :p
<Turl>
there's no head
<Turl>
it's all managed by `repo'
<mnemoc>
`make mele_a1000 buildroot`
<mnemoc>
Turl: right, that's exactly the stuff I'll need to be able to setup via scripts
<Turl>
`make reposync; make lunch zatab; make bacon' would be pretty fail
<Turl>
considering it'd be wrapping make in make already :P
<mnemoc>
narf
<Turl>
it can be done, but it doesn't add any value imo
<mnemoc>
get rid of broken ape-proof-wannabe tutorials
<Turl>
hm?
<mnemoc>
or at least shrink them
<Turl>
ape-proof stuff would be good, but apes are too good at figuring out the best way to make stuff fail :<
<mnemoc>
there are tons of tutorials of pages and pages and pages of instructions, errors and patches on which most stuff is a boilerplate we can actually get rid of
<mnemoc>
meh. lunch. back in 30m
cat_n9 has quit [Ping timeout: 246 seconds]
tuliom has quit [Quit: Konversation terminated!]
merbanan has joined #arm-netbook
cat_n9 has joined #arm-netbook
mysteryname has quit [Ping timeout: 248 seconds]
Xlab has joined #arm-netbook
<Xlab>
Hi all. I'm trying to get /dev/fb1, but device doesn't exist. What is the current status of FB1 support in the kernel?
Xlab1 has joined #arm-netbook
Xlab has quit [Ping timeout: 245 seconds]
cat_n9 has quit [Ping timeout: 246 seconds]
Xlab1 has quit [Ping timeout: 276 seconds]
wingrime has joined #arm-netbook
QingPei has joined #arm-netbook
popolon has quit [Quit: Quitte]
cat_n9 has joined #arm-netbook
toxicpsion has joined #arm-netbook
toxicpsion has quit [Remote host closed the connection]
toxicpsion has joined #arm-netbook
<bsdfox_>
who runs dl.linux-sunxi.org/nightly?
alcides has joined #arm-netbook
<mnemoc>
cnxsoft
<mnemoc>
but it would be awesome if he could hand it over to someone more active
<bsdfox_>
I'd be happy to host some if I can get my cross compiler setup :)
<mnemoc>
hosting is not a problem
Gumboot has quit [Ping timeout: 272 seconds]
<mnemoc>
but at this stage things change fast, and the nightly generation needs to keep the pace
Xlab has joined #arm-netbook
<Xlab>
root@allwinner10:~# cat /dev/fb1
<Xlab>
cat: read error: No such device
<Maqs>
from 2012.10.10 on there are no more builds?
QingPei has left #arm-netbook [#arm-netbook]
<bsdfox_>
Maqs, they've been failing
<Maqs>
can't you just remove "-msoft-float" from PLATFORM_RELFLAGS in arch/arm/cpu/armv7/config.mk (uboot)?
<bsdfox_>
Maqs, that's why I was asking who runs it :P should be an easy fix
ibrah has joined #arm-netbook
<mnemoc>
installing gcc-multilibs solved the -msoft-float problem for me
<mnemoc>
gcc-multilib
<Maqs>
what's the point in installing something when you can just remove something smaller? ;-)
<mnemoc>
i suppose u-boot isn't behaving well in hf and that's why soft-float is forced
<Maqs>
there are a few thousand (bit exaggerated) tutorials out there telling users to remove -msoft-float :-)
<rz2k>
maybe we should add mali x11 drivers with includes fixed and libump fixed for dri2 loading too?
<mnemoc>
+1
<shivansps>
that means working 2D acceleration?
<rz2k>
3d
<mnemoc>
shivansps: mali400 doesn't do 2d
<shivansps>
so it will never work for xserver?
<rz2k>
2d for x11 is done by EXA software acceleration
<mnemoc>
shivansps: yes, after libv blesses us with a exa driver using g2d
<rz2k>
driver is needed for DRI2 integration
<shivansps>
andddd thats the good part of x86 and windows... *runs .exe* all done :P haha
<rz2k>
lol
<mnemoc>
seriously? you can pass *hours* installing drivers on that junk
<mnemoc>
even to have very basic hardware support
<rz2k>
techn_: also could you please somehow mark rel1 and rel2 and whatever revision of libs
cat_n9 has quit [Ping timeout: 260 seconds]
cat_x301 has quit [Ping timeout: 255 seconds]
<shivansps>
actually on x86 ive having more trouble with linux than on windows, but it depends
<rz2k>
because if we will have r3p1-rel1 they will be entirely incompatible with rel0 drivers (see my messages about DRM drivers being coded by guys who apparantly have time machine)
<rz2k>
also there is two revisions of r2p4..
<shivansps>
it was a joke anyway haha
<rz2k>
but as far as I know they are compatible with each other.
alcides has quit [Quit: > SELECT * FROM users WHERE clue > 0]
<shivansps>
oh btw, i was able to boot linux on my jz4770, but the "pisces board" u-boot configuration only see 64mb of ram :/ here is where ainol u-boots sources could come in handy
<techn_>
rz2k: It could be that DRM stuff was accedently in that rel1 delivery
bsdfox has quit [Ping timeout: 252 seconds]
bsdfox has joined #arm-netbook
bsdfox has quit [Changing host]
bsdfox has joined #arm-netbook
<rz2k>
yeah but I've tested r3p1-rel1 x11 on r3p1-rel0 libs and secure_id is failing
<rz2k>
err, not secure_id, pixmap allocation
<rz2k>
it gets first buffer ok, then creates framebuffer fails and shouts in log with unable to create pixmap
<rz2k>
so the order of things changed somehow.
<rz2k>
time to write post to their forum
von_fritz has quit [Quit: vonfritz leaves, don't panic]
cat_n9 has joined #arm-netbook
rsalveti has quit [Ping timeout: 246 seconds]
cat_n9 has quit [Ping timeout: 264 seconds]
Fab_FR has joined #arm-netbook
<Fab_FR>
Hello there
<Fab_FR>
I created a page for my A13 tablet on linux-sunxi.org
<techn_>
Fab_FR: nice.. I should do same of mine
<Fab_FR>
And now I will start to hunt the pins that would manage the HDMI :D
cat_n9 has joined #arm-netbook
<rz2k>
there is no hdmi in a13
<mnemoc>
his a13 tablet has hdmi connector :|
<rz2k>
wtf
<rz2k>
:/
<shivansps>
i bet its a reused A10 PCB
<mnemoc>
rz2k: indeed, but in capitals
<mnemoc>
shivansps: that would be valid IF A10 and A13 had the same footprint
<mnemoc>
but they are completely different
<shivansps>
omg... i was expecting the A10 and A13 to be compatible pin to pin....
<shivansps>
that HDMI n A13 works?
<mnemoc>
shivansps: the a13 can even be soldered by hand
ibrah has quit [Quit: ChatZilla 0.9.89 [Firefox 16.0.1/20121011121529]]
<mnemoc>
shivansps: by someone skilled enough
<Fab_FR>
For the moment nope, but I'm looking for which pins could be interesting to try on
<mnemoc>
shivansps: it's a completely different package, made for cheap-as-hell tablets
<shivansps>
mmmm i bet that is to reduce the number of layers on the pcb
<Fab_FR>
As for my PCB It has many layers I guess
<rz2k>
no, both a10 and a13 can be layed out on 4layer, problem is bga fine-pitch pick and place machines
<rz2k>
which china seem to lack
cat_n9 has quit [Ping timeout: 255 seconds]
<Fab_FR>
On the back of my PCB is written HDSI, so... that's strange
<Fab_FR>
HDSI stands for High density semiconductor interconnect?
<rz2k>
also bga can be checked only with xray, when tqfp can be checked with eye sight
<shivansps>
still some times i dont understand the companys... did you see the A10 netbooks? most of the does not have HDMI, none has eSATA or even a empty internal 2.5" disk bay, thats a job for the A13, not the A10... and there is not A13 netbooks, when they should compite with VIA 8650 in price...
<rz2k>
I think guys who make a10 products just dont get that they can sell them.
<ZaEarl>
sun6i would be awesome in a laptop
<rz2k>
(...if we will have kernel sources, u-boot sources and mali libs)
mysteryname has joined #arm-netbook
<ZaEarl>
I have word from Eva that they will release the source
<mnemoc>
see to believe
<ZaEarl>
it's the best we have
<shivansps>
i hate to say it... but Windows sells, Linux do not, sun6i need to be fully compatible with Windows RT if it has to have any chance to be sucessfull in sells for netbooks and notebooks
<shivansps>
i hate windows rt/8 anyway
<shivansps>
there is any info con the sun6i yet?
<ZaEarl>
most likely quad-core A7 with Mali T6xx
<rz2k>
ZaEarl: seriously? that the first time I hear that they dont say "this is our cool technology, we are open to requests but not for opensource"
<mnemoc>
ZaEarl: nah, mali 450. ARM won't launch their 8xT6xx with a chinese manufacturer
<shivansps>
but they have 2 T6xx
<ZaEarl>
mnemoc, I see the 450 can handle 8 cores, so, yes, it could be that.
<mnemoc>
but it's cool anyway :)
<ZaEarl>
a good boost to performance, sounds good to me
<shivansps>
no opengl support :(
<ZaEarl>
We're going to see a lot of 2048x1536 displays. they'll need the boost.
<rz2k>
shivansps: if 450 is based on 400 we have a chance
alcides has joined #arm-netbook
alcides has joined #arm-netbook
alcides has quit [Changing host]
cat_n9 has joined #arm-netbook
<shivansps>
just opengl es, T6xx should have the desktop OpenGL... mmmm there is 5 T6xxx now.... maybe there is a chance for a T604...
<ZaEarl>
I don't see OpenGL listed on the T6xx specs. Just OpenGL ES.
<shivansps>
i still hoping that to be a mistake, the mali competitos have support for DX9 and well all OpenGL
<shivansps>
ARM is listing just DX11...
<shivansps>
http://pastebin.com/mJtSvqh8 First boot of the JZ4770 with the CP2102, using some demo rotfs from ingenic, im gona try a debian... debootstrap should do the trick