<rz2k>
since you are interested in mainline, no one will ever accept emulation layer in mainline, so we actually need an mtd driver to have nand support in mainline
BJfreeman has quit [Quit: had a good time]
<arokux2>
yeah, I've assumed that
tinti has joined #linux-sunxi
arokux2 has quit [Remote host closed the connection]
arokux2 has joined #linux-sunxi
arokux2 has quit [Ping timeout: 256 seconds]
Gerwin_J has quit [Quit: Gerwin_J]
<akaizen>
Hi, whats the difference between sugar_ref001 and sugar_evb and wing_k70?
<akaizen>
I've become a pariah :(
mcbrick has quit [Remote host closed the connection]
\\Mr_C\\ has quit []
rz2k has quit []
\\Mr_C\\ has joined #linux-sunxi
vinifr has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
Black_Horseman has quit [Ping timeout: 264 seconds]
<drachensun>
turl: I'm here
<Turl>
drachensun: just fyi, the mail I got from you was marked as possibly "forged" by thunderbird
Tsvetan has quit [Ping timeout: 252 seconds]
<drachensun>
hmmm we are using mailchimp, so its not coming from our normal pengpod server
<drachensun>
I wonder if that is why
<Turl>
drachensun: may possibly want to review SPF on pengpod.com then
\\Mr_C\\ has quit []
vinifr has quit [Quit: Saindo]
\\Mr_C\\ has joined #linux-sunxi
tinti has quit [Ping timeout: 264 seconds]
ykchavan has joined #linux-sunxi
[7] has quit [Disconnected by services]
TheSeven has joined #linux-sunxi
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
ykchavan has quit [Ping timeout: 245 seconds]
JohnDoe_71Rus has joined #linux-sunxi
<steev>
WarheadsSE: is that what i want _that_ for, being what we talked about? no. that's my own little side project
dwilkins has quit [Ping timeout: 245 seconds]
dwilkins has joined #linux-sunxi
Soru has quit [Read error: Connection reset by peer]
ykchavan has joined #linux-sunxi
ykchavan has quit [Ping timeout: 260 seconds]
Tsvetan has joined #linux-sunxi
<eebrah>
Just came across this demo of Ubuntu Touch running on the PengPod 1040 [ Allwinner A31 ] http://youtu.be/ypqjfOYiyCk, yay?
wingrime has joined #linux-sunxi
wens has quit [Quit: Changing server]
<Tsvetan>
Downloading now A10S SDK 1.6 (Android 4.0.4 and Lichee 3.0.8) do you guys have interest in it? the Kernel is old, so is the Android, Allwinner seems not interested to develop forward A10S like they do with A13 and A20
<Tsvetan>
also they made the price of A10S higher than A10 which shows that they want to move the customers away from this chip :-)
<rm>
interesting
wens has joined #linux-sunxi
MadSpark has quit [Ping timeout: 252 seconds]
Soru has joined #linux-sunxi
wingrime has quit [Read error: Operation timed out]
ZetaNeta has joined #linux-sunxi
<oliv3r>
Turl: i know how to diff thsoe files :p i ment, what's the easiest way to identify whcih one i'm using :)
<oliv3r>
hugobo_ you can't do what you want yet, ujnless you hack in the mtd drivers. the boot0 and boot1 bootloaders are as you said before the MBR. only boot0 and boot1 can boot from nand right now, you can't change that
BluesBoy has joined #linux-sunxi
rellla has joined #linux-sunxi
atiti has quit [Remote host closed the connection]
rellla2 has joined #linux-sunxi
rellla has quit [Ping timeout: 246 seconds]
tzafrir has quit [Ping timeout: 264 seconds]
n01 has joined #linux-sunxi
<n01>
moin
FR^2 has joined #linux-sunxi
<oliv3r>
rz2k my tablet sometime randomly starts to have corrupted shit even with the original kernel and ext4; so i'm not 100% sure even their code works 100% well
<oliv3r>
Tsvetan: could you get me a list of revisions that do and do not have a GPIO connected to the 5V sata power output? I think revA and revB do not have a gpio for this and are always on; revC and rev D do have the gpio
<oliv3r>
Tsvetan: i tried to hook the gpio in code but i think my board (revA) doesn't have it so a) can't test it, b) have to figure out how to handle those differences in the dts
<Tsvetan>
oliv3r for which board?
<Tsvetan>
A20?
rellla2 is now known as rellla
<oliv3r>
Tsvetan: yeah, but the board is also used for A10 right?
<oliv3r>
the olinuxino micro
<oliv3r>
arokux1: good point olimex image generator vs sunxi-bsp on ML +1
<oliv3r>
Turl: also pong
bamvor has quit [Remote host closed the connection]
shineworld has joined #linux-sunxi
<Tsvetan>
oliv3r rev.A and B are produced just 50 pcs IIRC this was the developer edition
<Tsvetan>
then rev.C and D are same and the only difference is that we experimented with the DDR serial resistors, C have D have no resistors
<oliv3r>
Tsvetan: that might be the reason the 5V isn't controlled via gpio :)
<Tsvetan>
and in Rev C/D we add SATA power on/OFF with PB8
<oliv3r>
yeah
<oliv3r>
so only rev C and revD have that
<Tsvetan>
when you make PB8 in 1 the SATA is enabled
<Tsvetan>
only REV.D is produced now
<oliv3r>
well the dts will work on C and D
<oliv3r>
so i'll make a dts for olimexino and one for -revCD
<Tsvetan>
it will work on A/B too as there SATA is always ON
<Tsvetan>
:)
<oliv3r>
yeah, but i can't 'claim' PB8 on those
<oliv3r>
is PB8 connected at all on revAB?
<oliv3r>
if it's N/C, maybe we can just turn it on/off and pretend its there :)
<oliv3r>
i'll have to think about that some more :)
<Tsvetan>
PB8 in rev A/B go to GPIO connector
<shineworld>
I've a dubt... I've got some problems with lichee 3.3 and I would like to know if my toolchain is right for that version: arm-linux-gnueabihf-gcc (crosstool-NG linaro-1.13.1-4.7-2013.04-20130415 - Linaro GCC 2013.04) 4.7.3 20130328 (prerelease)
<Tsvetan>
in Rev C/D it also go to this GPIO connector but also is connected to enable SATA
<shineworld>
what cross tool do you use to build linux-sunxi kernel ?
<oliv3r>
shineworld: 3.3 isn't really supported by us
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
Black_Horseman has joined #linux-sunxi
gzamboni has joined #linux-sunxi
<oliv3r>
Turl: os-release :D your v3 is version_id=2013.08-git
<shineworld>
I know... what's happen is simple ... I've build the cubiebook android 1.03 SDK using upon toolchain for kernel and livesuite output image doesen't work getting me a kernel panic so I guess is a problems with generated code from compiler
<shineworld>
JohnDoe_71Rus, in the cubiebook upon link none info about required toolchain version
<JohnDoe_71Rus>
shineworld: in the tarball all need for build. exept buildessential+ and jdk
Soru has quit [Read error: Connection reset by peer]
<shineworld>
JohnDoe_71Rus, you are right in the package I've found an older version (missed in previous find) ... I will try with that
<JohnDoe_71Rus>
shineworld: good luck
<shineworld>
crossing fingers
mturquette has quit [Ping timeout: 260 seconds]
mturquette has joined #linux-sunxi
_enrico_ has joined #linux-sunxi
notmart has joined #linux-sunxi
<arokux1>
johns9872003@yahoo.co.uk here?
<rellla>
arokux1 is nagging at him ;)
<arokux1>
rellla: I'm really not sure what he said. I think he wanted to say "I'd" instead of "I'll"
<rm>
he said he will read-only from now on
<rm>
not posting anymore
<rm>
which is fine by me
<rm>
if a person is a retard who can't grok a couple of simple rules of communication
<rm>
I am happy if they rather refrain from communicating
<arokux1>
ah.. I thought he wanted to say that I better read instead of replying to him
<arokux1>
couldn't believe he meant himself.
bamvor has joined #linux-sunxi
<binaryferret>
When attempting to use fel to grab boot info etc has anyone else had issues with 'libusb usb_bulk+send error -7' which is time out error code?
<binaryferret>
I've grabbed the version info, but then when i try to do the read to boot1.header it seems to time out, and then also fel version times out as well.
<ykchavan>
binaryferret, I got that error.
<ykchavan>
I added that in wiki. Let me find it
<binaryferret>
Cheers ykchavan
<ykchavan>
in short, it depends on which way you are going to FEL
<ykchavan>
Each FEL door entry is at different point of initialization.
<binaryferret>
Ahh ok. I'm going the 'FEL dump with no serial console access required (default) (bootinfo + script.bin)' way.
<arokux1>
oliv3r: he will not resign.. I won't reply to him, hopefully he'll only read-only from now..
<oliv3r>
arokux1: I don't think he's really old probably some kid
<oliv3r>
arokux1: you replied perfectly well
<oliv3r>
i need 96 hour days
<oliv3r>
or, i need to learn to live without sleep
<oliv3r>
that would work even beter
<oliv3r>
never be tired
<andoma>
both would be awesome!
<oliv3r>
arokux1: JINX!
<oliv3r>
andoma: 96 hour sleepless days? wow
<oliv3r>
al the hacking we could do in that time
<oliv3r>
speaking off, lunch time :p
<oliv3r>
then i'll reflash a new SD card with a new bootable image
<oliv3r>
so much to do again :S and there's also $work!
<oliv3r>
ugh
<andoma>
mhmm..
FergusL has quit [Ping timeout: 264 seconds]
<binaryferret>
ykchavan: Hi, sorry just nipped out. I'm trying to get linux running on a cheap generic tablet. I wasn't sure which U-boot config to use, so I'm trying to get as much info as I can. Not sure how to get console access to the tablet will read up on that after dinner.
<binaryferret>
cheap geneic a13 tablet
<binaryferret>
generic*
<binaryferret>
I have the script.bin/fex, version info, and a bunch of other system info just require to get boot info now.
<ykchavan>
what is name of the tablet and manufacturer?
<oliv3r>
binaryferret: any of the u-boot configuration that comes close will probably work; a13_mid porbably
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
<binaryferret>
model SoftwinerEvb, product.name nuclear_evb
<oliv3r>
binaryferret: taht's the generic evaluation board info
<oliv3r>
binaryferret: you need to run a10_meminfo from android (sstock rom)
<binaryferret>
Ok will get that onto the device and run it
<oliv3r>
that will give you your memory timings indicating which u-boot is required. generally one that matches your memory size is usually sufficient
<binaryferret>
Cool ok.
<oliv3r>
as for script.bin, if you copy that from your tablet, to your SD card, your good to go
<oliv3r>
as the script.bin defines your tablet to the kernel
<binaryferret>
Yeah that's one thing I have.
<oliv3r>
those 2 datum are the important things
<binaryferret>
Excellent. Thanks guys.
<oliv3r>
now if all a13 u-boots fail, tyhen we have to add your tablet to u-boot ;)
<oliv3r>
binaryferret: for starters however, its best to try hansg's fedora 19 image
<binaryferret>
hehehe. How can i tell if u-boot fails, if the screen remains black/off
<oliv3r>
Turl: v2 is 08, v3 is 2013.11-git :) how you got a november version in september I don't know, but they all 3 work fine (with my 3.11 SID branch)
<binaryferret>
hansg's fedora 19 image. Will give that a whurl as well and see what happends.
Black_Horseman has quit [Remote host closed the connection]
<speakman>
I'm porting an old device driver written by me long ago which make use of gpio pins (and their interrupts). It's many years old and I guess much have changed in how to make use of GPIO in an "generic" manner.
<binaryferret>
so I'd be able to connect one end ot my pc and the other to the standard a13 tablet micro usb slot?
<speakman>
oliv3r: hm, not sure. Which is most "stable" for an A20 board (OLinuXino) ?
<binaryferret>
Feel free to say 'No you silly sausage'
Soru has quit [Read error: Connection reset by peer]
<oliv3r>
binaryferret: the microUSB slot isn't GPIO controlled though, the mUSB has dedicated pins from the SoC
<binaryferret>
Ahhh okdokey.
<oliv3r>
speakman: sunxi-3.4 i suppose :)
<Turl>
binaryferret: ah, is it a tablet?
<binaryferret>
Yeah
<oliv3r>
ohh now i'm confusing people :D
<speakman>
oliv3r: ok, then it's drivers for 3.4. ;)
<binaryferret>
It's a chepo a13 generic tablet.
<Turl>
binaryferret: you'll either have to open it up, find the uart pins and solder
<speakman>
Or; how do I write a generic driver which utilizes two GPIO pins for some bit bangning (but only reading though)?
<oliv3r>
speakman: for 3.4 not sure, mainline uses the dts mostly to setup pins afaik
<binaryferret>
OOooooo
<Turl>
binaryferret: the only trouble is that you use up the uSD port and you can't put an SD on it
<Turl>
so you'll have to boot from nand
<Turl>
oliv3r: can give you the details :)
<oliv3r>
wiki
<oliv3r>
:p
<oliv3r>
you need to build u-boot from the lichee-dev branch, as a target you need the _sd ones
<oliv3r>
you need to modify your script.bin as per wiki
<binaryferret>
Ahh ok. I think i'll stick to booting from an sdcard at the moment rather than the nand then.
<oliv3r>
binaryferret: have you booted from SDcard so far?
<binaryferret>
Not yet. Put a few different things on the sdcard but it'd never progress from a black screen.
<binaryferret>
Went into the rootfs but no logs were seemingly made.
<binaryferret>
So not sure where it was falling down. It's why I took a few steps backwards and decided to make sure that I was getting u boot correct first.
<steev>
does the a20 (or any allwinner) support thermal/hwmonitoring?
<binaryferret>
It's fun either way, as I'm learning lots as I go along.
<oliv3r>
binaryferret: try hansg's Fedora 19 image, it's a perfect 'first time thing'
balage has joined #linux-sunxi
<arokux1>
binaryferret: what is your ultimate goal?
<balage>
oliv3r, Hi, i was away in the last 3 days. did you have the chance to test your tablet for the wifi issue I have (lsusb can't see wifi, and actually usb keyboard doesn't work either) on my Hackberry?
<balage>
oliv3r, fyi I recompiled my kernel and disabled OTG (I had different settings with 3.0) but did not help.
<Montjoie>
steev, for thermal monitoring I found somethin in the axp209, but I must code the driver
<oliv3r>
balage: i have not :) i DID manage to boot 3.4, but my initramfs was bad, something about not able to execute init
<oliv3r>
balage: so i gave up for the time being
<balage>
oliv3r, okay, no problem, is there anything I can do for troubleshoot?
\\Mr_C\\ has joined #linux-sunxi
<steev>
Oliv3r: did you pass init= on the command line? If not just symlink it from where it is in the initramfs to the root of the initramfs
<oliv3r>
steev: i used the exact same initramfs so it 'should' in theory work I thought :)
<oliv3r>
i extracted the boot.img from my android partition
<oliv3r>
extracted the initramfs and put it straight into my kernel
<oliv3r>
and booted that :) so maybe it was that, but i hadn't looked yet to be honest
<oliv3r>
btw, what was the advantage of using boot.img over just putting the files on the fat partition? i forgot
<steev>
Less chance of an issue with it when the file system gets un mounted prematurely I'd guess
Black_Horseman has quit [Remote host closed the connection]
shineworld has joined #linux-sunxi
<arokux1>
ganbold: hey, were you the guy which runs some *BSD?
ganbold_ has joined #linux-sunxi
tinti has joined #linux-sunxi
<binaryferret>
arokux1: Ultimate goal is to get an android build booting from an sdcard
<binaryferret>
My thinking was if I could get the kernel built, and a linux rootfs booting on a sdcard then I can then make the next step.
<binaryferret>
More steps in getting the parittioning correct etc for android, so that's why I've opted for the linux rootfs first.
<arokux1>
binaryferret: I see. Would be nice to have you as a kernel hacker :)
<mouchon>
hello
<mouchon>
i would like to know if some one is working on spi for the A20 ?
<binaryferret>
I'm afraid I'm still learning alot, I work as a junior embedded engineer but it's mostly been with atmegas at the moment. I use linux at home and at work but just starting to get through this side of linx/kernel. It's a lot of fun.
<n01>
mouchon: you? :)
<oliv3r>
Turl: yay, i got a working defconfig now that gives me console. so now i'll compare and strip
<mouchon>
yes i would like to try
<binaryferret>
Might be awhile before I can start properly hacking the kernel and actually knowing what I'm doing.
<oliv3r>
binaryferret: ther's a howto for exactly that on the wiki
<n01>
good, welcome on board
<oliv3r>
mouchon: spi is being worked on in mainline by mripard
<oliv3r>
mouchon: didn't you send an email to the ML?>
<binaryferret>
oliv3r: Yep. The whole wiki is very useful.
<oliv3r>
mouchon: i think i2c was yours wasn't it
<mouchon>
yes for i2c
<mouchon>
i just posted the clean (i hope patch)
<oliv3r>
mouchon: yeah your patch looks good I suppose, but I don't have i2c hardware to properly test it with :)
<mouchon>
yes i have
<mouchon>
i have logicial analyser
<oliv3r>
but spi for mainline is being done by mripard
<oliv3r>
mouchon: for 3.4 it should be there allready
<oliv3r>
mouchon: otherwise it's just a port away from lichee-3.4 -> sunxi-3.4
<mouchon>
oliver for i2c i tested with different frequency at boot time and when loading as modules. after i checked with some i2c device i own
<andoma>
any of the cedar reverse engineering peeps in here?
<mouchon>
oliver i am currently working with stage-3.4
<arokux1>
binaryferret: it is easy to start with sunxi-3.4 and try to unify things, once you are done with ur android build.
<mouchon>
and on stage-
<mouchon>
the spi for a20 is not working
<oliv3r>
mouchon: then it needs to be copied from lichee-3.4
<oliv3r>
and unified to sunxi-spi
<oliv3r>
and/or
<binaryferret>
what's the difference between lichee and sunxi repositories?
<oliv3r>
sunxi is 'ours'
<oliv3r>
lichee is allwinners (sdk) dump
<mouchon>
ok will try
<oliv3r>
we started with the 3.0 dump and started to port the whole thing to 3.4, fixed tons of stuff etc
<oliv3r>
AW first took their drivers from 3.0, dropped them into 3.3; ignored out changes
<oliv3r>
now they've released 3.4 sdk, and took their crap from 3.3 and put it into 3.4
<oliv3r>
all fixes, improvements, merges etc are completly ignored
<oliv3r>
so many old bugs are being resurfaced from users trying these AW kernels
<binaryferret>
Ahhhh that must be very annoying.
<arokux1>
oliv3r: you mean they have ignored our fixes?
<oliv3r>
arokux1: absolutly
<oliv3r>
they kept using their stuff and kept mulling forward
<arokux1>
oliv3r: :) why did you think they won't?
<oliv3r>
our kernels are far superior to their
<oliv3r>
s
<arokux1>
oliv3r: they probably are not following our tree at all
rellla has joined #linux-sunxi
<arokux1>
oliv3r: what is the status of the 3.10?
<arokux1>
I might want to try my usb driver in there, because I know rtlwifi from 3.10 works. the could have been a regression between 3.10 and torvalds tree, though unlikely.
tinti has quit [Quit: Leaving]
<oliv3r>
arokux1: it boots :D
<oliv3r>
arokux1: only a10 for now, turl has been very busy backporting a lot of stuff
<oliv3r>
his stuff is in wip/experimental/3.10
<oliv3r>
but he's awol so I might test it and push it into non-wip
<arokux1>
oliv3r: I have A10 too ;)
Soru has joined #linux-sunxi
<oliv3r>
good :)
<oliv3r>
i'll push a new sunxi_Defconfig soon
<oliv3r>
testing it still
<arokux1>
oliv3r: let me please know once you backported sunxi-clk.c (or was it ckl-sunxi.c ...)
shineworld has quit [Quit: Leaving]
<oliv3r>
Turl: what happened to your sram driver
<oliv3r>
arokux1: turl is working on that ;)
<arokux1>
oliv3r: cool
<Turl>
oliv3r: sram? it's already on mainline :p
<arokux1>
Turl: I hope u'll submit pll6 clock to mainline soon :p
<Turl>
mnemoc: I backported all the things (see wip/exp/3.10) but it doesn't boot on my A20 :p
<Turl>
need to investigate further
<arokux1>
mnemoc: yeah.. you see, you are bad manager :p
<mnemoc>
Turl: maybe importing everything in one pass was too greedy
<Turl>
mnemoc: but hey it builds :P
<mnemoc>
=)
<oliv3r>
Turl: merged and commited allready? never saw the results of that; what can we use it for etc?
<Turl>
oliv3r: the driver was already on mainline :p
<oliv3r>
Turl: i thought you merged a*0 support into it
<Turl>
oliv3r: I should probably send patches to add the dt nodes
<Turl>
what support is there to add? it's just ram :p
<oliv3r>
Turl: but you where hacking on it like mad a few weeks ago!
<Turl>
I'd say it was more than weeks :p
<n01>
Turl: returning -ENOMEM from a *_probe with dev_err message in redundant?
<mnemoc>
arokux1: today I'm going to measure my next root and se how much space I really have, and the buying bed, desk, etc to be able to work productively on not-$work$ stuff
<Turl>
oliv3r: that was PM though
<Turl>
n01: if you mean print("No mem!"); return -ENOMEM, don't
<mnemoc>
after almost a full year in limbo....
<Turl>
just return, if there's no mem an error will be printed already
<arokux1>
mnemoc: you can try to find some second hand furniture shop
<n01>
Turl: ok, so it is redundant
<arokux1>
fairkauf or smth, you can buy good stuff there cheaply
<oliv3r>
Turl: Power Managment? I don't recall that at all; i remember you hacking on the sram driver!
<mnemoc>
arokux1: yes, but it's an attic.... so dimensions are very irregular
<n01>
whereas it is important for stuff like -ENODEV I presume
<Turl>
oliv3r: well, you need to copy the code to SRAM
<Turl>
maybe that's what you recall? dunno
<mnemoc>
arokux1: need to properly diagram it first
<oliv3r>
Turl: let me whitequark you
<n01>
mnemoc: did u find a home?
<Turl>
n01: hm -ENODEV?
* Turl
is not too familiar with probe funcs :p
<n01>
:) ok
<mnemoc>
n01: big room in an attic, shared house
<n01>
gratz :)
<mnemoc>
thanks :)
<mnemoc>
now I need to make it usable :p
<n01>
you can host me then if I drop by ;)
<oliv3r>
Turl: ok i can't find it, maybe i'm imagining things
<mnemoc>
n01: possible :)
<Turl>
oliv3r: you need more sleep time
<Turl>
:p
<oliv3r>
Turl: i'd be happy with more time generally
<mnemoc>
stupid God making days only 24h long
<n01>
God? which God?
<mnemoc>
don't know. or I have already written a complain
wingrime has quit [Ping timeout: 240 seconds]
<n01>
yeah, bugzilla right?
<binaryferret>
MEANWHILE IN Valhalla 'Another complaint about the hours in a day Odin'
<oliv3r>
Long live the gods
<mnemoc>
the gods should have a bug tracker for world, universe and life issues. and solve their f* mistakes
<binaryferret>
If this is all a simulation, perhaps they do.
<mnemoc>
=)
vicenteH has joined #linux-sunxi
<binaryferret>
#432 Humans: The entertainment facilities are near the waste disposal, we should do something about this.
panda84kde has joined #linux-sunxi
<arokux1>
the day is actually longer now than it used to be when the earth was rotating faster
<oliv3r>
oh interesting new driver possibilty? GPIO power off
<oliv3r>
'Board level reset or power off'
<oliv3r>
and 'restart power off driver'
wingrime has joined #linux-sunxi
akaizen has quit [Ping timeout: 248 seconds]
drachensun has quit [Quit: Leaving]
akaizen has joined #linux-sunxi
leowt has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
ykchavan has joined #linux-sunxi
Soru has quit [Ping timeout: 256 seconds]
BluesBoy has quit [Quit: HydraIRC -> http://www.hydrairc.com <- Like it? Visit #hydrairc on EFNet]
<choko>
Hi. I have problem with usb detection. I must unplug and plug any usb device after booting of system. If i boot system with connected usb device it don't detect it. My device is mk802 with a10s chipset, kernel: 3.4.61.sun5i+,u-boot:mk802_a10s
<choko>
any sugestions ?
<arokux1>
choko: not yet, but if you post your dmesg (for the boot with usb device in) there could be some
<choko>
ok but where, pastebine ?
<arokux1>
choko: just do this: dmesg | curl -F 'sprunge=<-' http://sprunge.us
<choko>
i'm from windows ;)
<arokux1>
choko: ok.. use whatever you like but plz give a link to raw data
<choko>
so you have only two: one mini usb and normal usb
<arokux1>
choko: now, I do not know why normal (which is better referenced to as otg) usb port works so strangely, but can you try to use that second one?
<arokux1>
choko: because I kinda know more about it and (maybe) can fix problems.
<choko>
i don't have mini usb/usb adapter right now but i test it
<choko>
i thinks problem is in u-boot
<arokux1>
choko: i'm not sure. why do you think so?
<choko>
because everthing works normal if you boot system first and then connect usb device
<arokux1>
choko: does it work with stock android (or whatever)?
<choko>
yes
<arokux1>
choko: did it work previously with our u-boot/linux?
<choko>
i tested other u-boots like: A10s-OlinuXino-M and r7-tv-dongle same results
<arokux1>
choko: you shouldn't randomly test u-boot for other boards..
<arokux1>
choko: so this is your first try of our kernel, right?
<steev>
yeah... u-boot doesn't quite work like that
woprr has quit [Remote host closed the connection]
<arokux2>
akaizen, I'm on mainline
<arokux2>
on my desktop I see a whole lot of stuff in /dev/input/
<arokux2>
but not the case with sun4i, I wonder, is there kernel config that should be enabled
<arokux2>
USB_HID is on
<akaizen>
arokux2: I see sry but I dont know
<akaizen>
You're much further along than I am :)
<wingrime>
oliv3r: u-boot page: U-Boot Initially developed by Tom Cubie and now developed and maintained by Henrik Nordström
<oliv3r>
yeah sounds about right
<oliv3r>
tom had access to boot0 and boot1
<oliv3r>
sdram.c is almost a direct copy
<wingrime>
oliv3r: U-Boot tom's home-made project or so
<wingrime>
oliv3r: but page suggest that
<oliv3r>
u-boot was tom cubies work initially yeah
<oliv3r>
and then when aw allowed it to be released; henrik and him worked on it outside
<wingrime>
oliv3r: you can't say that tom started u-boot intself, but you can say that tom do initial port to sunxi arch
<oliv3r>
ohhh
<Turl>
wingrime: add "sunxi" at start
<Turl>
:p
<oliv3r>
sunxi support was added to u-boot initially by ...
<oliv3r>
yeah chinglish
<oliv3r>
Turl: your tree also stops on a10 with ecc disabled
<oliv3r>
well the wip tree
<oliv3r>
i'll do it 1 patch at a time
<drachensun>
wingrime: have you seen this touch driver before?
<drachensun>
ft5x06_ts
<drachensun>
The newer A20 device I have gotten have it
<Turl>
oliv3r: on A10?
<arokux2>
drachensun, have you googled? there is some code
<oliv3r>
Turl: yeah
<drachensun>
arokux2: I looked about a month ago, I hadn't since, thanks
<Turl>
oliv3r: notice that I used plain 3.10
<Turl>
oliv3r: what compiler are you using?
<oliv3r>
turl i just commited something on your wip tree (locally only) witha s title 'merge me' cause i need to merge it on a diff branch; but git log won't show it at all?
<oliv3r>
Turl: gcc 4.6.3
<wingrime>
drachensun: I had some table with ft5x
<Turl>
oliv3r: git log branch?
<oliv3r>
i'm on the wip branch
<wingrime>
drachensun: but ft5x06 if I correct was in a20 sdk....
<drachensun>
wingrime: yeah most of mine have that ft5x06_ts is different, the ft5x driver wont work with it
<wingrime>
drachensun: a20 sdk,
<wingrime>
drachensun: wait a second I check it myself
<oliv3r>
Turl: i did git add path/to/file; git commit -m 'merge me'
<oliv3r>
then git checkout experimental/sunxi-3.10
<oliv3r>
and it properly said 'your branch is ahead/behind by 1 cmmit' when i switch between
<oliv3r>
but i do'nt see the commit :S
<Turl>
then git log wip/experimental/... should show it?
<oliv3r>
so can't find the hash so can't cherry pick
<Turl>
oliv3r: otherwise just use git reflog
<oliv3r>
ok
<oliv3r>
found it with reflog
notmart has quit [Quit: notmart terminated!]
<oliv3r>
but identical commit hash?
<oliv3r>
19fb844 HEAD@{1}: checkout: moving from experimental/sunxi-3.10 to experimental/sunxi-3.10
<oliv3r>
19fb844 HEAD@{2}: commit: merge me
<oliv3r>
what does that mean?
<drachensun>
wingrime: I had downloaded it but not unpacked it yet, doing so now
<Turl>
oliv3r: git show 19fb844 ?
<oliv3r>
that's the one
<oliv3r>
but 2 entries with the same hash ... puzzelign
<arokux2>
mainline isn't stable. maybe EMAC.. just hangs after some time
<oliv3r>
n01: ohhh i like the sound of that
<oliv3r>
arokux2: emac is bugged in hardware
<oliv3r>
arokux2: the driver is very basic and possibly bugged
<oliv3r>
even in 3.4 its bugged
<oliv3r>
so I doubt emac can ever be stable
<oliv3r>
hence i'm hopeing for gmac on a20
<arokux2>
i'm testing usb2ethernet now....
<oliv3r>
got 2 of those; 1 axis based, the other i don't remember
<drachensun>
Something else funny to me, the display driver seems to create 8 framebuffers, or I guess now SUNXI_MAX, but only calls the next init routines on the first 2
<arokux2>
Turl, do you have a wifi dongle?
<wingrime>
drachensun: err
<wingrime>
drachensun: only ft5x2 ...
<oliv3r>
Turl: you did some branch merging etc? the end of your ocmmits look a little messy?
<drachensun>
then registers all 8
<wingrime>
drachensun: but may be helpfull
<oliv3r>
Turl: in other words, can I just go down the list of commit hashes and start picking them?
<wingrime>
Tsvetan: are you finished a10s sdk down?
<drachensun>
wingrime: I got some source from the supplier and arokux2 just pointed me to different source
<arokux2>
drachensun, is it different from suppliers?
<drachensun>
wingrime: I had tried to use it before with the first source and it wouldn't work under Linux, not sure about the latest find
<drachensun>
arokux2: yup, looks pretty different
mcbrick has quit [Quit: Leaving]
<arokux2>
drachensun, its developed by the vendor of the touchscreen though
<drachensun>
I've got some smaller volume guys who want the A20 but that is the impediment to offering it
<drachensun>
arokux2: I could be wrong, let me check again
<wingrime>
drachensun: you business is not such size that can change something in a world , even Cannonical can't
<drachensun>
lol
<drachensun>
yes, that would seem so
<wingrime>
massss
<wingrime>
volumeeee
<arokux2>
I'm getting
<arokux2>
dhcp_open: eth1: Address family not supported by protocol
<arokux2>
ideas?
<drachensun>
if I got the press about my crowd funder that Canonical got for there, I could offer something though :-) We are just getting no press reaction this time.
<wingrime>
drachensun: but industrial applications maybe workable
<wingrime>
drachensun: It's maybe possible found some sponsor in that area
<wingrime>
drachensun: black-berry at least alive "somehome"
<wingrime>
*somehow
<drachensun>
heh yup
<drachensun>
I dont know we will see
<wingrime>
drachensun: all ways belong to own R&D, design, but not garantee direct success Example MS surface
<wingrime>
no one need tablet that can't run window apps without MS cert
<drachensun>
well I can probably sell 100 or so right away if I can get this touch screen driver working but I really dont want to take time from trying to promote the crowd funder
<drachensun>
interested it fixing it at taking a share?
<drachensun>
arokux2: you are right its the same driver, I just have version 3 and that is version 1 it seems
<oliv3r>
drachensun: haven't payed attention, but what's the A31 crowdfunding status?
<drachensun>
not good
<drachensun>
21 days left, only 26k of 350k
<oliv3r>
i'm sorry to hear dude
<oliv3r>
oh stillt hat long
<drachensun>
yeah
<wingrime>
drachensun: realy I here only for non-profit fun
<drachensun>
we will still be fine if the press kicks
<drachensun>
in
<drachensun>
but I'm somewhat out of gimmicks to get their attention at the moment
<wingrime>
drachensun: but If you send ".ko" I can take a look with IDA and comparea with sources we have
<drachensun>
wingrime: Ok, I understand
<n01>
Turl: does it make sense memset-ting a dma_handle??
<drachensun>
wingrime: I appreciate the offer, if I decide to go after it next week I'll send that on
<Turl>
arokux2: nope
<Turl>
oliv3r: hm? it's a linear history
<Turl>
n01: memsetting with what?
<oliv3r>
Turl: git: 'sl' is not a git command. See 'git --help'.
<oliv3r>
Turl: your script fails!
<wingrime>
drachensun: this is realy not your work as founder, do such things, you main work- find money by anyway
<Turl>
oliv3r: meh :P
<n01>
Turl: dma_addr_t dma_handle from dma_alloc_coherent()
<Turl>
oliv3r: sl == log --oneline
<Turl>
oliv3r: you gotta get my alias :D
<oliv3r>
ahh ok
<Turl>
n01: no idea
<n01>
ok :(
<Turl>
n01: why do you want to memset it?
<oliv3r>
Turl: thnx works
<wingrime>
drachensun: so, if you sand original driver ".ko" I can find time to look at it, and find whats differs
<arokux2>
Turl, CONFIG_PACKET=y, (Packet socket) (why the fuck they cannot say this in error message?!)
<oliv3r>
so now i'm debating, merge patch by patch, or git bisect your branch
<Turl>
arokux2: ?
<arokux2>
Turl, ah, you replied to usb2wifi?!
<n01>
to zero it. the problem is that it is supposed to be a bus address not a virtual kernel address
<oliv3r>
119 patches
<n01>
and I have a driver that memset it
<n01>
(not written by me)
<wingrime>
drachensun: but , also, you "feedback" is required (so, you have to test findings/fixes in sources we have to find on real hw)
<arokux2>
ok, ppl. another usb device works. usb2ethernet
<arokux2>
Turl, oliv3r mripard ^
<arokux2>
steev, hey, could you test wifi dongle?
<wingrime>
drachensun: thats realy,realy sad that you using your time to fix hw instead finding good sponsor
eebrah_ has quit [Ping timeout: 260 seconds]
<wingrime>
find some angel, or something alike , but I have no exp in such things, I can say only general stuff I think
<arokux2>
wingrime, do you have wifi dongle?
<wingrime>
arokux2: no, but my tablet's have internal usb's
<wingrime>
wifi
<arokux2>
wingrime, internal won't work probably....
<arokux2>
wingrime, although you could try. do you have serial?
<wingrime>
arokux2: I have solder it to testpoints
<arokux2>
wingrime, what sunxi is it?
<wingrime>
arokux2: a13 and a10
<wingrime>
arokux2: in a13 case I know where uart test points
<wingrime>
arokux2: in a10 no
<arokux2>
wingrime, can you test wifi in mainline? :)
<wingrime>
arokux2: you know why I have problems with it, I need solder to testpoints, new sdcard, etc
<Turl>
arokux2: I don't have a USB wifi, dunno what _PACKET is?
<Turl>
oliv3r: add cubieboard into the grep, there's some patches that don't have sun*i on them :(
<wingrime>
arokux2: but if you have patch for sunxi branch
<arokux2>
Turl, I've thought you've replied to a different message of mine.
<wingrime>
arokux2: I can do it easy
<arokux2>
wingrime, no patch for sunxi branch, what do you mean?
<wingrime>
techn_: no edid, handshake thats must work without , no any things of life on screen
<arokux2>
and show it your love
<wingrime>
techn_: tcon0 is LCD
<arokux2>
Turl, what was the usb device that has broken things?
<Turl>
arokux2: USB soundcard
<techn_>
wingrime: but you just said that color bars are working?
<oliv3r>
Turl: ARM: sun5i: Update the clock compatible strings
<oliv3r>
that's the first one i cp
<techn_>
wingrime: and you set those color bars to come from hdmi regs?
<wingrime>
techn_: when I changed bit with working disp
<wingrime>
techn_: in linux
<techn_>
ah
<wingrime>
techn_: but this important, as we can test hdmi without anything
<wingrime>
techn_: without de_fe and de_be
<oliv3r>
wingrime: that's why it's so important test
<wingrime>
oliv3r: yeax, you not need any frontend and backend code for it
<oliv3r>
yep yep
<drachensun>
wingrime: ? Yeah we can't get that A10 devices anymore thats why I'm looking at an A20 as a stop gap
<wingrime>
drachensun: only one model now?
<drachensun>
i've nothing now
<drachensun>
unless the crowd funder succeeds
<wingrime>
drachensun: how many was selled?
<wingrime>
drachensun: thats cool, you at least selled all
<drachensun>
wingrime: yeah, we could buy those 100 at a time so we didn't have such a risk
<wingrime>
drachensun: how about a13 tablets?
<wingrime>
drachensun: also you have no pre-pay mechanism in you store?
<drachensun>
pre pay how? its pretty much all pre pay through a credit card or paypal
<Turl>
drachensun: why is it a 350k$ crowdfunder btw? do you need to buy like 5k tablets?
<drachensun>
turl: 2k and indiegogo fees and and shipping and shipping out and all that
<Turl>
drachensun: your previous igg campaign was much smaller though, how many did you buy that time?
<wingrime>
drachensun: crowd funds are some kind of pre pay, you can rid of indiegogo fees with prepay
<drachensun>
turl: Around 700 but the devices cost a lot less in that case and the actual minimum on those was only 100 units
Black_Horseman has joined #linux-sunxi
<oliv3r>
and to top it off, some of us still shun the a31 :)
<drachensun>
wingrime: I see what you mean, the problem with trying to do a prepay directly is (1) do people trust me to return it (2) its actually pretty difficult given the rules on paypal and merchant accounts. Paypal tried to suspend my account last crowd funder over that EVEN THOUGH we went through indiegogo
<drachensun>
we waste a week getting the funds released
<wingrime>
drachensun: face-palm
\\Mr_C\\ has quit []
<wingrime>
drachensun: can't calculate you profit loss without anything at store and such
<wingrime>
drachensun: but even without pre pay you can ask people claim (no money!) order and send e-mail when it will aval
<drachensun>
well right now with nothing in the store there is no profit :-)
<drachensun>
yeah
<drachensun>
last time we had about a 30% follow through rate on that
<drachensun>
the problem is capital though
<drachensun>
I just wouldn't have the cash to buy 2k tablets without the crowd funding
<wingrime>
drachensun: just add some pre-claim on store, and ask user for email
<Turl>
drachensun: SMB loan?
<wingrime>
drachensun: when it will be successed you send massive emails to them
n01 has quit [Read error: Connection reset by peer]
n01 has joined #linux-sunxi
<drachensun>
wingrime: yeah, we did that for the 2 months between crowd funding and delivery last time, 1/3 came back and paid
<drachensun>
turl: The lady at the bank didn't laugh at me
<drachensun>
turl: but it was close
<wingrime>
drachensun: otherwice if they claim has a failed you send anouther e-mail with other buy suggeest
<wingrime>
drachensun: also, spam users some time when you have new things at store
<Turl>
time to find another bank :p
<drachensun>
turl: A loan like that might be possible for the second batch was the up side of that though, they thought that 'might' be possible
<drachensun>
man, somebody just sent me one of those 'why not just buy this' aliexpress links
<drachensun>
it actually looks like a pretty good device
<wingrime>
drachensun: sell suport not device in prior
<drachensun>
I thought about that, we might be able to do that if we get this one going
<drachensun>
we will have a full time guy for support and order packing then
<wingrime>
drachensun: also, just to store "will be there soon" sections and get user's emails, and also, send some week-reports to them
<wingrime>
*just add to
<oliv3r>
Turl: how can 2 patches in order, result in a conflict? and did you manually solve it
<oliv3r>
Turl: error: could not apply 442f150... ARM: sunxi: Remove the .map_io function declaration
<Turl>
yeah I did
<Turl>
just nuke both :)
<wingrime>
drachensun: geeks are mostly entusiasits, they pay not for device they pay for future, so don't loose theirs attention, news, blog, mail reports, support, updates,....
<oliv3r>
LOL
<drachensun>
wingrime: Yeah, I've just become a little fatalistic, we might add some of that though
<arokux2>
ok, another USB device is verified to work. this time USB keyboard.
<arokux2>
oliv3r, Turl mripard ^
<oliv3r>
ok i'll test it soon
<oliv3r>
actually i'll compile now
<Turl>
oliv3r: ah, found what was missing
<Turl>
oliv3r: we need bc37324e820e4a23a0bccef79ae797ce4d93 as prereq
<Turl>
"boots" with that
<oliv3r>
Turl: oh that's fast
<oliv3r>
doesn't mention sunxi at all
<oliv3r>
arm though of coruse
<oliv3r>
i'm gonna copy/paste that list of patches
<oliv3r>
that bc37 does that go 'first'?
<oliv3r>
or does it have dependancies
<arokux2>
Turl, you sure your sound card works with mainline?
<arokux2>
Turl, on desktop I mean
<n01>
WASD Keyboards Order 133970 has been shipped HUUU
<Turl>
arokux2: yeah it does, I use it on 3.4 in fact
<Turl>
and my PC detects it just fine too
<arokux2>
Turl, but what is the version of your kernel on desktop?
<Turl>
oliv3r: cherrypick that first, also pick the sched_clock one
<Turl>
then you can pick what the script says
<arokux2>
can anybody please test my tree with wifi dongle?
<Turl>
ignore the reboot ones
<Turl>
arokux2: 3.11
<oliv3r>
'the shed_clock one'
<arokux2>
Turl, mainline is one step further.... also there were some problems
<wingrime>
drachensun: thats worse than fatalistic, but just understand , you have do only one thing as start-up leader -- get money,attention==future-money, future customers, investors, there is many people that have mony for some invest , geeks are sometimes are works at big companyes, investment ins't like simple buy, don't simply kick customers form store, get theirs claim, promise and be honest, if you can't do their's claim just suggest some other varian
<arokux2>
Turl, I'm not saying you are affected.. but who knows.....
<oliv3r>
Turl: so what do you recon is better, merge your tree 'as is' or cherry-pick those patches and fix/add missing bits as we find them
<mripard>
Turl: yes ?
<drachensun>
wingrime: We are just trying to direct anyone interested now to the crowd funder. Its basically the same thing as taking a pre-order
<Turl>
arokux2: I built as =y, so not passing anything on index
<drachensun>
wingrime: In the USA they have just legalized crowd funding for actual investments like stock, I'm kind of curious to see what that brings
<Turl>
oliv3r: I'll do yet another repick later today on top of the branch mnemoc prepared, we can push that then
<Turl>
oliv3r: better if we fix pinctrl first tho :)
<arokux2>
mripard, do you have a usb wifi dongle and want to test mainline usb? :)
<mripard>
I have crappy ones, so I'm not sure it'll prove useful :)
<Turl>
mripard: off the top of your head, any ideas why may pinctrl explode in sunxi_pconf_group_set+0x54/0x200?
<wingrime>
drachensun: excuse - is some behavior when you agree with curent state-of-things
<oliv3r>
Turl: well I've started doing the cherry-pick on the experimental tree in order
<Turl>
mripard: we're backporting all the stuff to 3.10
<mripard>
Turl: of course :)
<oliv3r>
Turl: what pinctrl fixing?
<mripard>
I have no idea :)
<Turl>
oliv3r: it explodes awesomely :D
<wingrime>
drachensun: you free find a new ways always, be independed thinker
<mripard>
but it seems you're already seeing the backporting nightmare I was talking about :)
<Turl>
mripard: 90% backports cleanly :)
<arokux2>
mripard, it would be very kind of you if you could test a usb wifi.. :)
<Turl>
10% takes 90% of the time to figure out though :p
<mripard>
yeah, and as usual, those take 10% of the time :)
<mripard>
arokux2: I'll be back home next week, I'll try to find itt
<arokux2>
mripard, I've proven 3 different devices work already: keyboard, usb2ethernet, usb storage
<wingrime>
drachensun: I simply poke some things that can be easy changed -- small detals decide everything
<arokux2>
mripard, ah.. I meant now :p
<arokux2>
wingrime, you are electronics guy, righ?
<mripard>
arokux2: I'm 4000 kms away from my wifi dongle :)
<arokux2>
mripard, oh.. I'm closer to it than you then! :)
<mripard>
indeed :)
<mripard>
I'm in Helsinki right now
<arokux2>
somebody tries to skim the passwords of our university users and I cannot inform anybody.
<arokux2>
should I try police?
<wingrime>
arokux2: some kind of
<arokux2>
wingrime, what do you think can I measure with a multimeter around my wifi module to check if it is operated correctly? i.e. the same way as under some different kernel?
<arokux2>
wingrime, maybe it's a crazy idea..
<oliv3r>
Turl: ok i've stopped at the siscors and testing (just before the pinctrl)
<granite_crusher>
arokux2: I thought maybe it needs to be set at boot, so I reboot and then in U-boot i made "usb start" (with sound card pluged in) and u-boot kept vomiting some repetite messages
<granite_crusher>
*repetitive
<arokux2>
granite_crusher: ah.. you should have saved them somehow...
<granite_crusher>
arokux2: I did
<arokux2>
granite_crusher: cool! show them to me.
<granite_crusher>
this time I set log on putty (I know everybody will start to advice I should use screen :) )
<arokux2>
granite_crusher: no.. I use gtkterm, for example
<granite_crusher>
arokux2: log is very big, I should cut just essential part, and will add comments (with hashes what I did)
alcides has joined #linux-sunxi
<arokux2>
granite_crusher: ok, but feel free just to dump it on me
wingrime has quit [Read error: Operation timed out]
alcides has quit [Remote host closed the connection]
<arokux2>
granite_crusher: sorry, you are probably tired already of this testing
<granite_crusher>
arokux2: I added some comments, its messy, that puTty logging was configured crappy
<arokux2>
granite_crusher: yep, I see, thanks a lot
leowt has quit [Quit: leowt]
akaizen has quit [Read error: Connection reset by peer]
<granite_crusher>
arokux2: there are few things which I think is worth to discuss in short
akaizen has joined #linux-sunxi
<arokux2>
granite_crusher: ok, go ahead
<granite_crusher>
arokux2: first - I tested with voltmeter that USB_EN pin (it fortunatelly is exposed on board as GPIO) and it is 0 with comented out line and 1 normally
<granite_crusher>
arokux2: however in u-boot it is 0
<granite_crusher>
arokux2: so if board would have not this hack of usb always enabled, probablly it would be unable to boot from usb?
<granite_crusher>
by 0 and 1 I mean 0V and
<granite_crusher>
3.3v
<arokux2>
granite_crusher: hm.. you are right, since I've totally forgotten to adjust u-boot for A13, it has different gpios!
<granite_crusher>
well A13-olinuxino in default configuration will have no problem with that
<arokux2>
granite_crusher: .. but we shouldn't rely on this.
<granite_crusher>
arokux2: but if there is other devices based on A13..
<arokux2>
granite_crusher: so I'll adapt u-boot.
<arokux2>
granite_crusher: what else? :)
<granite_crusher>
arokux2: second thing it is possible to mount just two usb sticks at same moment
<granite_crusher>
there is no mounting part in logo
<granite_crusher>
but used blkid which behaves same
<arokux2>
granite_crusher: you mean it is *not* possible?
<granite_crusher>
arokux2: it doesnt mount, it gives error of invalid file or device