deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
popolon has quit [Quit: Quitte]
egbert has quit [Disconnected by services]
egbert_ has joined #linux-sunxi
rz2k has quit []
awafaa has quit [Ping timeout: 240 seconds]
awafaa has joined #linux-sunxi
mnemoc has quit [Ping timeout: 248 seconds]
mnemoc has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
eebrah|noPC has joined #linux-sunxi
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
eebrah|noPC has quit [Ping timeout: 240 seconds]
dwilkins has quit [Ping timeout: 246 seconds]
dwilkins has joined #linux-sunxi
deasy has joined #linux-sunxi
<hno>
oliv3r, that the chip have its reset "button" active all time, constantly being reset.
<hno>
i.e. not capable of doing anything at all.
wingrime has joined #linux-sunxi
drachensun has quit [Ping timeout: 264 seconds]
discopig has quit [Read error: Connection reset by peer]
drachensun has joined #linux-sunxi
<deasy>
plop
eebrah|noPC has joined #linux-sunxi
eebrah|noPC has quit [Ping timeout: 245 seconds]
vicenteH has quit [Ping timeout: 256 seconds]
stekern has quit [Ping timeout: 256 seconds]
stekern has joined #linux-sunxi
<q3k>
oliv3r: Well. I tried to restore a image yesterday, but it all went bonkers (wouldn't even complete, got a nice chinese error message from PhoenixSuit at ~90%) and the flash MBR was completely toast.
<q3k>
oliv3r: So, I ended up copying the MBR from the extracted android image, at least to I can format the NAND into one big root partition for Linux, and I'm keeping u-boot on a small SD Card.
<q3k>
Hoewever, in other news: when I try to set the display node on the kernel command line (eg disp.screen0_output_mode=EDID:1280x1024p60), one of my screens doesn't sync to it (and displays that the resolution is 1282x1025, wtf), and another one display a few pixels of pink margin around the image.
<q3k>
A20, you so silly.
<q3k>
Or maybe I shouldn't try a kernel from Git.
wingrime has quit [Ping timeout: 252 seconds]
<q3k>
(it wouldn't even compile with the default PM settings, so it might not be the highest quality kernel tree ever :V)
discopig has joined #linux-sunxi
discopig has quit [Read error: Connection reset by peer]
rellla has joined #linux-sunxi
discopig has joined #linux-sunxi
wingrime has joined #linux-sunxi
discopig has quit [Ping timeout: 240 seconds]
discopig has joined #linux-sunxi
jemk has joined #linux-sunxi
arokux2 has joined #linux-sunxi
<arokux2>
mripard: hi there
jemk has quit [Ping timeout: 264 seconds]
wingrime has quit [Read error: Operation timed out]
<deasy>
hi mr arokux1
<deasy>
arokux2*
<arokux2>
hi deasy
<deasy>
popolon have do some test with linaro he say most things work
<deasy>
someone have improve the "mmc" driver?
<deasy>
he had trouble just a little time before with it
Soru___ has quit [Read error: Operation timed out]
vicenteH has joined #linux-sunxi
soul has joined #linux-sunxi
soul is now known as Guest75459
<jemk>
rellla: i wanted to try to make vdpau work with xbmc, but it looks like xbmc doesn't even support vdpau with gles. did you have any success there?
Soru has joined #linux-sunxi
Guest75459 has quit [Ping timeout: 245 seconds]
<hramrach>
anyone has schematic for cubietruck?
<rellla>
jemk: i don't exactly know, how rendering in xbmc works together with vdpau... but it seems, that vdpau requires some gl/glx.
<focus_it>
hramrach: me want too
<focus_it>
i order 2x cubietruck yesterday - shipping starts october 15
<focus_it>
may be get it by end of month in uk
<hramrach>
ok, there is no schematic yet :/
Soru__ has joined #linux-sunxi
Soru has quit [Read error: No route to host]
leviathanch has joined #linux-sunxi
Soru has joined #linux-sunxi
Soru__ has quit [Ping timeout: 260 seconds]
<focus_it>
hramrach: anyone, any idea when schematic might get released? I want to get to a new SoM board design with 2GB RAM and 8GB flash. Friday i deliver first prototype using cubieboard2 - would have done it with eoma but i need the 2x eoma for working on
<focus_it>
100% gpl'd 2GB RAM with 8GB flash in kicad is what i be doing
<hramrach>
olemix is going to make an a20 SoM
<hramrach>
*Olimex
<focus_it>
i note CN manufacturing has gone up one step and do 0.1mm tracks and clearances as routine - this means the new board i can do with up to 2 tracks between ball grid pads :)
<hramrach>
hehe, olemix sounds cool too
<focus_it>
it might be possible to shrink the 2GB RAM/ 8GB flash down to eoma size :)
<hramrach>
well, if you want to do it just for the fun of doing it go for it
<focus_it>
as before i put 200 pin SODIMM on the board on one side
<hramrach>
otherwise Olimex has one in testin which should be ready in a month or so
<hramrach>
well, sodimm might be nicer format than lots of pins on the *top* of the board
<hramrach>
but w/e
<hramrach>
you did an a10 SoM or something before?
<focus_it>
I put 200pin SODIMM on one side and duplicate FPC connectors on the other side - so SoM board can plug into motherboard or if there is no motherboard, can connect usb, ethernet, lcd etc though fpc connectors
<focus_it>
once get it working, it can be as small as needed - about 2 inch x 2 inch - but you got it in kicad - so make your own formats
<hramrach>
why is there that array under the sodimm connector?
<hramrach>
oh, it's not under, it's the board :S
<focus_it>
that is patch panel to wire in any of the board wires to the sodimm - but it lined up with the board edges, so you can cleanly break it off it not needed
<hramrach>
I really like the pattern the traces make on the other side of the sodimm connector ;-)
<focus_it>
the idea is that you got board with patch panel to rapidly configure and test product, and when you are certain of the results, you would go to the kicad, delete the bits you don't need and wire in the bits you need, and voila product ready to go into PCB production in about 2 days
<focus_it>
:)
<focus_it>
anyone can make their own product :)
<focus_it>
its all in kicad and kicad is free / gpl'd
<focus_it>
neither are we in the sense that we won't make any cubie2 or eoma product to sell because we are not that kind of company
<hramrach>
the whole point of SoM is that you have a working hi-density module which the manufacturer debugged
<focus_it>
we make the infrastructure for anyone to make product - that doesn't exist at the moment
<hramrach>
and you just make simple board with whatever connector the SoM has
<hramrach>
and that's the infrastructure for anyone make a product
<focus_it>
existing soms either carry too much or too little of what ever is needed to make product in highly cost sensitive embedded gadgets
<hramrach>
you onl make like dual layer pcb with simple routing, no massive parallel buses
<focus_it>
the bits near the cpu don't try to re do because that is hard once we got it working
<hramrach>
they carry a SoC. that gives you the featureas and you just put the peripherial connections on external connectors
<focus_it>
products we make for example don't need ethernet - but some do, some need lcds, some don't, etc - hard to get anything in the $20 range to do all that
<hramrach>
the only decision to make for a SoM is to pick a SoC and perhaps amount of ram and flash
<hramrach>
focus_it: how much does it increase cost to make the connector long enough to accomodate pretty much everything?
<focus_it>
we make product supported for 20 years - can't rely on other people's designs
<focus_it>
i though of that - use a series of 80 pin fpc - not at all difficult
<hramrach>
you can if the design is opensource like olimex does wehn it comes to that
<focus_it>
100% gpl'd
<focus_it>
and in kicad so your average ee can learn it and modify it
<focus_it>
olimex in eaglecad - expensive if you need the full features to do up changes
<hramrach>
yeah, they say so. did not read the fineprint because I don't do hardware so don't really care
<hramrach>
still if their design works you likely won't need changes
<focus_it>
if they fail as a company, we fail with them :( - so we never build anything that doesn't come with a circuit diagram and software sources
<focus_it>
this company has been going since 30's with that philosophy
<focus_it>
so it works :)
<focus_it>
since 1930
eebrah|noPC has quit [Remote host closed the connection]
<hramrach>
well, it does not work for me. never used your product :p
<hramrach>
and, srsly, you are talking rubbish. if you do your own design even if they fail as company their design is free to use for you forever
<hramrach>
it would be challenging for me since I would have to figure out how to arrange for the boards to be made by somebody else but you don't have that problem
jinzo has joined #linux-sunxi
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
<focus_it>
hramrach: no - if they fail as a company they take with them their know how. no bom means almost no way to make a duplicate, and no software means total dead end
<focus_it>
we have problems when staff leave to duplicate a product - the know how goes with them and then it has to be relearned by some one else - takes months if not years - cheaper some time to make a duplicate product and re use whatever that was
<focus_it>
modern products are so complex its hard to know what magic is inside it - even a complete 100% duplicate with the blessing of its original manufacturer who sold us designs often never works for years on end
<focus_it>
because they don't have the 100% documentation or staff that made their product who may have left to rely on to give us the support
zetaneta has joined #linux-sunxi
zetaneta has joined #linux-sunxi
<hramrach>
well, the olimex boards run Linux so you have all the software
<hramrach>
you are just suffering from NIH here
<focus_it>
that is idea, and potentially the hardware because at least its in eagle, but we fret over what might happen to eagle if they are not around in 20 years
<focus_it>
not bothered by nih - we buy in 90% of the stuff - so long as it includes hardware and software
<focus_it>
when you have to fork out $1m to remake a product - you will really really understand the power of 100% owning hardware and software documentation really means :)
<focus_it>
once we get a product certified, it cannot be easily changed - that adds to the complications
<focus_it>
we can't change out a transistor for example without getting the whole thing re-tested
<focus_it>
at least $5k
<focus_it>
:)
<hramrach>
well, I don't have the resources to design a board so not thrilled by the procpect of re-designing one
<focus_it>
hramrach: no worries - you be software engineer
<focus_it>
you just need hardware that works
<hramrach>
or at least pretends well
<hramrach>
arokux2: did somebody write it or does it come from the Chinese?
tymbarq has joined #linux-sunxi
tymbarq has quit [Client Quit]
<hramrach>
well, from the path it looks like it's from Chinese
<arokux2>
hramrach: the chinese have written it...
<arokux2>
hramrach: in _their_ english
<hramrach>
bah, I can't speak chinese so don't know how they are thinking
<hramrach>
but since it's written there it probably means the driver should pull it down. or it reads its state and this is the branch of code to take when low
<hramrach>
if it was irrelevan for the driver even Chinese would not write it there
<hramrach>
hmm, looks like the earlier
jemk has quit [Ping timeout: 264 seconds]
<hramrach>
and probably means whoever was writing hte driver used docs that had wrong naming of pins nad did not bother to update it
<hramrach>
because if you use the pin its not reserved ffs
<arokux2>
hramrach: what is the meaning of reserved?
<arokux2>
hramrach: like claimed by the driver or what?
<hramrach>
it usually means not assigned yet - unused but not to be touched
<hramrach>
but it might refer to the fact that the driver just claimed the pin and be in really horrid chenglish :s
<arokux2>
hramrach: yeah, that what I'm thinking
<arokux2>
hramrach: without that call the driver will still work. was I lucky?
<hramrach>
who knows
<arokux2>
hramrach: I better do how chinese do, right? :)
<djahma>
Hi hramrach, thanks for putting me on the right track yesterday, about the combo wifi/BT chip. I realized my tablet uses bcm40183 module which apparently is packed into bcm4330. When I try to add this last module I get the following output which I dont understand: http://pastebin.com/RVsfNVLh
<arokux2>
I was also on that list, wonder where is mine. ppl say it could have been split into 2 batches
<djahma>
hramrach: script.bin is correct, I mean it is the one from nanda partition from the manufacturer
<hramrach>
djahma: you need to know what bcm40183_wl_regon is
<hramrach>
djahma: it may be hardcoded in hte driver, too
atsampson has quit [Ping timeout: 264 seconds]
<djahma>
but what is hte?
<djahma>
:-/
atsampson has joined #linux-sunxi
<hramrach>
*the
<djahma>
ha! sorry
<djahma>
ok, I return into my investigations
<arokux2>
focus_it: I wonder if you noticed my private message.
wingrime has quit [Ping timeout: 245 seconds]
rz2k has joined #linux-sunxi
eagles0513875_ has quit [Changing host]
eagles0513875_ has joined #linux-sunxi
eagles0513875_ has quit [Changing host]
eagles0513875_ has joined #linux-sunxi
<oliv3r>
hramrach: i can boot the onboard android and adb shell in; why whats wrong?
<djahma>
hramrach: this investigation is beyond my skills! However, I notice in sunxi-3.4 that there is support for "broadcom 4330 wireless cards" (the one currently selected as module) and also support for "broadcom 4329/30 wireless cards" which require a firmware path and a nvram path. Should I select this last one instead? But how to set these paths properly?
rellla has joined #linux-sunxi
<oliv3r>
q3k: i'm curious as to maybe your mbr is completly broken. on the bright side, you have a prime example on working on the mtd stuff ;)
eebrah|noPC has joined #linux-sunxi
rellla has quit [Remote host closed the connection]
\\Mr_C\\ has quit []
\\Mr_C\\ has joined #linux-sunxi
Soru has quit [Read error: Operation timed out]
<djahma>
compiling with bcmdhd=m gives the following error in sunxi-3.4: ERROR: "register_inet6addr_notifier" [drivers/net/wireless/bcmdhd/bcmdhd.ko] undefined!
<ssvb>
focus_it: please define "not much is happening", how do you know that you actually have a problem?
<ssvb>
focus_it: ah, I guess that's modesetting again
<focus_it>
ssvb: glxgears same speed as software emulation
Gerwin_J has joined #linux-sunxi
<focus_it>
ssvb: popolon in #cubieboard mention CONFIG_SUN4I_G2D=m needed in .config about 2 days ago (he plan to blog it soon) and also /dev/g2d needs to be created about - not sure if i need to do both
wingrime has joined #linux-sunxi
wingrime has quit [Read error: Operation timed out]
[7] has quit [Read error: Connection reset by peer]
TheSeven has joined #linux-sunxi
<ssvb>
focus_it: try es2gears instead of glxgears
<ssvb>
focus_it: when using mali binary drivers, glxgears will always run as software emulation
<ssvb>
focus_it: but if you are *really* interested in running glxgears with hardware acceleration (and some of the old OpenGL 1.x games), then you can try https://github.com/lunixbochs/glshim/
wingrime has joined #linux-sunxi
<hramrach>
oliv3r: sunxi 3.4 kernel locks up with two penguins showing on the fb console
<focus_it>
ssvb: installed mesa-utils-extra to get es2gears - it crawls slower than glxgears :(
<focus_it>
ssvb: one thing i notice i do different is i did all the actions as normal user, and then did only the make install as root - will that be ok? or do i need it do it all as root?
<focus_it>
i believe i did that yesterday - checking...
<rz2k>
glxgears are ok to have software
<rz2k>
mali cant do GLX, it can do only EGL
<rz2k>
check if you have /dev/ump
<rz2k>
may be you dont even have mali loaded/working
eebrah|noPC has quit [Ping timeout: 256 seconds]
<focus_it>
i did the 50-mali.rules in /etc/udev/rules.d/
<focus_it>
no /dev/ump ... doing it up now
<focus_it>
sory /dev/ump exists
<rz2k>
are you sure you have the rights to write to /dev/ump then?
<rz2k>
and udev set them
<rz2k>
maybe your user is not in video group?
<focus_it>
rz2k: ack - what do i do to fix all 3 issues?
<focus_it>
cat /etc/group - not in video :(
<focus_it>
groups jm - video not lists - going to add it now
<hramrach>
oliv3r: it's the disp/lcd sections
<hramrach>
using the cubieborad's works but the CT ones break mmc and the root fs never comes online
<rz2k>
focus_it: for one time testing, just do 777 for /dev/mali /dev/ump
<rz2k>
if it does work, fix your udev rules
<focus_it>
r2k: yeeeeeeeeeeeeehhhhhhhhaaaaaaaaaaaaaaa!!! es2gears 135 fps small - 25 fps large
<focus_it>
:)
<rz2k>
ok
<rz2k>
thats accelerated enough, I guess
<rz2k>
you can make 350 or around if disp layers will be used
<rz2k>
check the Xorg log
<rz2k>
it will tell you what you need to have zero-copy
<rz2k>
you can also install glmark2-es2 to test the fancy stuff
<focus_it>
wow! got a rotating horse!
<focus_it>
3D
<rz2k>
there's 15 or so test
<rz2k>
with various graphics
<focus_it>
now a texture rendered box in b/w
<rz2k>
s/test/tests/
<focus_it>
a 3d blue cat
<focus_it>
asteroid?
<rz2k>
switch back to where you've launched the glmark2 to see the results, it should be around 50-80
<focus_it>
fps is 75 when it starts up
<focus_it>
gelly fish nice
<focus_it>
i'm logged into unity window manager - but unity panels all disappeared
<focus_it>
the desktop icons are there and working
<rz2k>
compiz crashed?
<focus_it>
sudo ccsm to enable it
arokux2 has quit [Remote host closed the connection]
<focus_it>
but no panels as yet
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
popolon has joined #linux-sunxi
djahma has quit [Ping timeout: 252 seconds]
jemk has joined #linux-sunxi
djahma has joined #linux-sunxi
<hramrach>
is gmac supposed to work?
<focus_it>
rz2k: thanks for the help, lot of problems with unity and compiz; can't type anything into terminal, other window managers work like xfce, lxde gnome-fallback - just going to reinstall on another partition and try again noting all the detailed steps
<hramrach>
ok, I have USB ehternet
geecko has joined #linux-sunxi
<rz2k>
focus_it: i dont use unity so no idea
<rz2k>
maybe it is bugged with egl
<rz2k>
or mali is bugged, no idea.
<geecko>
mali :3
<geecko>
leaked android 4.3 from samsung, drivers still r3p1. i wiping my tears right now
<geecko>
i'm*
<rz2k>
be sure to send kudos to samsung for drivers not being r2p1
<rz2k>
because this guys actually can pull that off
<focus_it>
rz2k: one time it briefly came up, screen went blank, it disappeared when screen came back, then it disappeared, screen went blank and it disappeared again when screen came up and thereafer terminal will start up but can't type anything into it to try out more stuff :(
<rz2k>
you need to debug these guys: nux, compiz, unity
<rz2k>
there's even a script called nux_graphics_check or something like that that will tell you if you can run unity on this sytem
<focus_it>
i just ran kicad with with full 3D rendering of PCBs and components (older .brd files) :D
djahma has quit [Remote host closed the connection]
<focus_it>
that was on the eoma68 - practically identical except for script.fex
<focus_it>
and boot.cmd
rellla has joined #linux-sunxi
<focus_it>
rz2k: got another job on, it be a while before i do debugging - pick low hanging fruit for now :)
<focus_it>
doing all this now in weekend spare time
<focus_it>
anyway thank you rz2k, very pleased I go this far today
atsampson has quit [Read error: Operation timed out]
atsampson has joined #linux-sunxi
deasy has quit [Ping timeout: 264 seconds]
<focus_it>
torcs - got 0.2fps - other drivers finish before i got to first corner :D
<djahma>
in the script.bin file the part about [sdio_wifi_para] contains a value for sdio_wifi_mod_sel which goes from 0 to 10, but in sunxi kernels, apparently, only values up to 7 are managed, whereas mine is sdio_wifi_mod_sel=8
<djahma>
really??! I've spent so much time looking for it! I can't believe it. I try it now!! cheers
<oliv3r>
is the wifi+bt connected via sdio or usb?
<oliv3r>
(i haven't looked)
rz2k has joined #linux-sunxi
<hramrach>
sdio
<djahma>
sdio
<oliv3r>
awesome
<hramrach>
yeah, AW's take on sdio is ttly awesome
<djahma>
nop, not working. this module was actually already compiled in the kernel with MMC_SUNXI_POWER_CONTROL=y but when I switch the tablet on, no device is recognised
Guest90865 has quit [Read error: No route to host]
<djahma>
Looking into dmesg, there's something working, but power is turned off immediately during boot: http://pastebin.com/GYX98NMR
Soru has joined #linux-sunxi
<djahma>
would you know how to use this sdio card once the system has booted?
<arokux2>
what is going on guys?
zetaneta has quit [Quit: Leaving]
Seppoz has quit []
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
Soru has quit [Read error: Connection reset by peer]
arokux2 has quit [Remote host closed the connection]
Soru has joined #linux-sunxi
jemk has quit [Remote host closed the connection]
eebrah|noPC has quit [Ping timeout: 268 seconds]
geecko has quit [Ping timeout: 264 seconds]
arokux2 has joined #linux-sunxi
djahma has quit [Ping timeout: 240 seconds]
djahma has joined #linux-sunxi
Soru_ has joined #linux-sunxi
Soru has quit [Read error: No route to host]
Soru_ has quit [Read error: Connection reset by peer]
Soru_ has joined #linux-sunxi
wingrime has quit [Ping timeout: 264 seconds]
leviathanch has quit [Remote host closed the connection]
Soru___ has joined #linux-sunxi
Soru_ has quit [Ping timeout: 264 seconds]
leviathanch has joined #linux-sunxi
Seppoz has joined #linux-sunxi
akaizen has quit [Remote host closed the connection]
akaizen has joined #linux-sunxi
ganbold_ has quit [Remote host closed the connection]