ChanServ changed the topic of #linux-sunxi to: Allwinner/sunxi development discussion - Don't ask to ask. Just ask! - See http://linux-sunxi.org | https://github.com/linux-sunxi/ | Logs at http://irclog.whitequark.org/linux-sunxi
torqu3e has quit [Quit: torqu3e]
mdfe has quit [Remote host closed the connection]
techn__ has quit [Read error: Connection reset by peer]
<libv> mdfe just bought a MID7042, should be there on friday
<libv> we will take pictures, and fill out the wiki page
<libv> and i just found this on the fcc website, a full set of pictures of the device torn down: https://apps.fcc.gov/eas/GetApplicationAttachment.html?id=1673045
<Turl> mnemoc: I think I figured out the missing alloc bit
<Turl> hopefully I'll try tomorrow on device and see if it works :)
<Turl> libv: stop linking us to privileged stuff :P "You are not authorized to access this page."
<libv> Turl: really?
<Epsylon3> weird, ive no idea from where come these bad alsa names ;)
<Epsylon3> but i fixed it
<Epsylon3> open tv use the same ones as your 3.4
<libv> there are 10 testpoints on the bottom of the pcb...
<libv> ..**.
<libv> .....
<libv> *.**.
<libv> .*..*
<libv> now the bottom 5 must be jtag :)
<libv> and the upper two are quite likely going to be rx/tx :)
<Turl> libv: what device is it?
<libv> MID7042
<libv> coby kyros
<libv> ithamar already provided .fex and u-boot
<Turl> isn't that like ancient and with resistive screen?
<libv> 1GB, capacitive
<libv> mdfe was very specific about getting 1GB for plasma active
<libv> 800x480 though :(
<Turl> ah, not the tablet I was thinking of
<Turl> a friend has one that iirc is from coby
<libv> coby does more than 1 tablet :p
<Turl> but it's resistive and slow, and has GB
<Turl> libv: how is lima doing these days btw?
<Turl> are you still working on the research driver, or have you moved on to mesa and the like?
<libv> now cleaning up the crap that i wrote for FOSDEM, finally
<libv> and then i will give mesa a kicking
<Turl> awesome :)
* Turl needs a blog of sorts to drop brain dumps on
<Epsylon3> :)
<Epsylon3> google+ :p
<Epsylon3> nah, its a big tweeter
<Epsylon3> too noisy
<Turl> I think I'll set sth up with Nikola
<Epsylon3> drupal is nice for that
<Turl> drupal is too big :)
<Epsylon3> lol, audio works in winamp but...
<Epsylon3> sound is really weird
<Epsylon3> like a dj scratch
<Epsylon3> with volume
<Epsylon3> 03-21 09:55:26.286 D/WifiHW ( 140): faied to read proc/net/wireless
<Epsylon3> fayed :p
egbert_ has quit [Read error: Operation timed out]
egbert has joined #linux-sunxi
hramrach has quit [Remote host closed the connection]
focus has quit [Ping timeout: 245 seconds]
hramrach has joined #linux-sunxi
egbert has quit [Ping timeout: 246 seconds]
egbert has joined #linux-sunxi
focus has joined #linux-sunxi
<Epsylon3> drivers/usb/gadget/f_fs.c:2423: internal compiler error: in dwarf2out_finish, at dwarf2out.c:18906
<Epsylon3> so the dwarf is in ffs
<Epsylon3> ive seen asm stuff named ffs about dwarf :)
torqu3e has joined #linux-sunxi
Epsylon3 has quit [Quit: ( GuikScript 1.2 )]
hipboi has joined #linux-sunxi
ZaEarl has quit [Ping timeout: 264 seconds]
rellla has joined #linux-sunxi
shineworld has joined #linux-sunxi
bsdfox_ has quit [Ping timeout: 276 seconds]
rz2k has quit [Read error: Connection reset by peer]
hansg has joined #linux-sunxi
traeak has quit [Ping timeout: 245 seconds]
anunnaki has quit [Remote host closed the connection]
<shineworld> in wiki linux-sunxi.org/FirstSteps what is sense of : mkdir -p lib/modules
<shineworld> rm -rf lib/modules/
<shineworld> cp -r <PATH_TO_KERNEL_TREE>/output/* .
hipboi_ has joined #linux-sunxi
hipboi has quit [Ping timeout: 245 seconds]
<oliv3r> i opened up my tablet :D i found some unmarked pads, no clue whats on them. silently, i'm hopeing its usbc1 (and probably tty0)
rellla has quit [Ping timeout: 245 seconds]
<oliv3r> but how to test :S i put the scope up to the pads, but all measured about 250 - 300 mV
<oliv3r> so nothing interesting there (yet)
rellla has joined #linux-sunxi
<oliv3r> shineworld: that sounds tupid :p
<oliv3r> and there's pads for a second RAM chip
<shineworld> just to notice the wiki bug ;)
<oliv3r> all ihave to do, is find a 512 MiB ram chip with same specs and solder it on to make it a 1g tablet :D
<vicenteH> when I connect an usb cable from my tablet to the computer, my computer does not detect anything in syslog.
<vicenteH> I would like to see the console in which u-boot boots,
<vicenteH> should I install some module in the kernel?
<oliv3r> you want to see the boot console over USB?
<oliv3r> don't think that is possible, i think you need a USB rs232 cable for that
<vicenteH> oliv3r: so if tablet does not have rs232 output I cannot see the boot console?
traeak has joined #linux-sunxi
shineworld has quit [Quit: Leaving]
<Turl> mnemoc: congratulations on your shiny CoM :)
<Turl> libv: any idea what might this mean? [SurfaceView] error creating EGLImage: 0x3003
hipboi_ has quit [Quit: Leaving]
<oliv3r> vicenteH: all have it though, if not via the internal 'pads' there's a micro-SD -> uart breakout board :)
<vicenteH> thanks oliv3r
gzamboni has joined #linux-sunxi
ganbold_ has joined #linux-sunxi
calris has joined #linux-sunxi
<mnemoc> Turl: hopefully, thanks to this module, I'll be able to use A10 for $work$
<mnemoc> on din-rail stuff
Dave77 has joined #linux-sunxi
<Turl> mnemoc: nice
<Turl> mnemoc: what do you build for $work?
<Turl> mnemoc: I got the stuff not segfaulting anymore now, but it doesn't decode either
<Turl> I'm hoping a proper allocator will do the trick, currently it just gives you random buffers :)
hramrach has quit [Ping timeout: 276 seconds]
Dave77 has quit [Ping timeout: 264 seconds]
n01 has joined #linux-sunxi
hramrach has joined #linux-sunxi
<libv> Turl: BadAlloc, but why this is so, i do not know
<libv> Turl: what is surfaceview?
<Turl> libv: some android thing
vicenteH has quit [Ping timeout: 252 seconds]
<libv> Turl: do you always get this, or only sporadically?
<Turl> libv: always, apparently when trying to copy?/display?/something the cedar decoded video
<Turl> I fail, I left a comment on the olimex website and typoed my own name lol
<libv> well, debug the application on top, and see whether this is the root cause of the issues, maybe it does fail to alloc things properly
<libv> if not, you're twice buggered
<Turl> it might be my uberderpy cedar allocator implementation though
<Turl> I just used what a guy had RE'd and protected it with a lock or two, but it's incomplete and isn't really thought to be used on gralloc
ganbold__ has joined #linux-sunxi
ganbold_ has quit [Ping timeout: 258 seconds]
ZaEarl has joined #linux-sunxi
vicenteH has joined #linux-sunxi
<libv> oliv3r: have you tried finding serial on your yarvik tab260?
eebrah has joined #linux-sunxi
<libv> oliv3r: also, what do you think of the structure of http://linux-sunxi.org/Hyundai_A7HD
<traeak> hmm. auto manufacturer's custom chip?
<libv> traeak: more likely a cheap chinese trademark violator looking to take advantage of a known brandname
<traeak> heh
<traeak> forgot about that little problem in china
<traeak> wife says they're going to instate a 20% capital gains tax on profits of property sales very very soon
<traeak> not sure what that will do to the chinese stability
plan_b has joined #linux-sunxi
<hramrach> it will make things very stable ;-)
<hramrach> no sales = no change = stable ;-)
wingrime has joined #linux-sunxi
plan_b has quit [Ping timeout: 240 seconds]
Dave77 has joined #linux-sunxi
<slapin_nb> A7HD is very good device
<slapin_nb> I'd say it is best A10 device I ever seen
<slapin_nb> it is not cheap either
<slapin_nb> and site is quite Hyundai'ish http://www.hyundai-digital.cn/ShowProduct.asp?ID=145
<slapin_nb> and here http://www.hyundai-digital.cn/aboutus.asp it is told they are part of big Hyundai or something.
<slapin_nb> and considering the quality of device, I don't have any reasons to not beleive
shineworld has joined #linux-sunxi
eebrah has quit [Read error: Connection reset by peer]
<libv> yeah, it is nice
<libv> but mine is pretty ruined :)
<libv> the previous owner moved the wifi antenna (the big issue with this device)
<libv> and the change in cabling made for uneven backlighting
<libv> i then added more cables for serial :)
<libv> so the display, 1024x600 IPS, is now _very_ unevenly lit
<slapin_nb> libv: mine is ok, I've managed to get serial out very accurately via tiny 4pin connector, I've made small hole in one of sides and glued the connector there
<libv> right, you learned from my mistake :)
<slapin_nb> backlight is ok with mine
<slapin_nb> libv: I did it before you bought one :-P
<libv> really?
<libv> ah, right
<slapin_nb> the cables are fragile, yes, but this case is with all screen modules
<libv> i remember talking to you about it like 6m ago
<slapin_nb> libv: yeah
<slapin_nb> only problem I see with this device is during charge sometimes screen flickers, probably due to noise on backlight from current consumed by battery (flickering depends on how much your battery is discharged). probably some tracks are not fat enough, but it is harmless.
<libv> reasonably ok for a .cn device, yes
<slapin_nb> libv: people do mistakes
<slapin_nb> libv: very little of devices I have don't have design mistakes (looks at n900 on table).
<libv> hehe
<libv> i should still get mine repaired
<libv> slapin_nb: on the n9 all the n900s mistakes were corrected!
<libv> (sadly, a lot more mistakes were made in the process :p)
rz2k has joined #linux-sunxi
<slapin_nb> not to mention devices we have making here, most of which work by miracle or two (loke 2 powering errors compensate each other or software lithium charger implemented in u-boot).
<slapin_nb> so I can't blame neither n900 nor A7HD designers too much.
<slapin_nb> they both work perfectly well for all my use cases
* slapin_nb crosses fingers
Dave77 has quit [Ping timeout: 252 seconds]
rellla has quit [Remote host closed the connection]
ganbold has quit [Remote host closed the connection]
hansg has quit [Quit: Leaving]
<mnemoc> Turl: car wash tunnels, lubes and self service car/dog washes. currently using ugly "industrial" atom/via PCs
eebrah has joined #linux-sunxi
ganbold__ has quit [Remote host closed the connection]
paulk-desktop has joined #linux-sunxi
<Turl> self service dog washing o.O
<mnemoc> you swipe a CC and then use a dial to enable certain things (water, shampoo, air, ...) and when you are done I bill you
<mnemoc> but car wash tunnels are clearly more fun
<mnemoc> from a software perspective I mean, reading the license plate, measuring the vehicle, scheduling things, ...
<mnemoc> self service are dumber
Epsylon3 has joined #linux-sunxi
<Epsylon3> got the bug
<Epsylon3> maybe that can be applied to usb only
<mnemoc> Epsylon3: that solves your gcc crash on gadget/android.c ?
<slapin_nb> Turl: I remember phone serviced public wash machines
<Epsylon3> in fact its in f_fs.c
<Epsylon3> which was included by android.c
<Epsylon3> i tried to build f_fs separated
<Epsylon3> removing static prefixes before
<Epsylon3> but i heard on mailing lists the dwarf error is related to the -g
<Epsylon3> and this patch remove the -g
<slapin_nb> Turl: and this can be used for everything, especially in front door of brothels, nise feature it seems and should be very popular
<Epsylon3> only on CONFIG_DEBUG_INFO_REDUCED
<Epsylon3> with that, that crash on first asm file :p
<Epsylon3> so weird
<Epsylon3> i think there is some conflicting symbols
<Epsylon3> ive seen in dwarf stuff some FFS defines
<Epsylon3> an asm feature, not remember what exactly
<Epsylon3> dont*
<oliv3r> libv i just tested the pads with my scope, i'll need to let a serial program output 1010101 and see if I can spot that
<libv> oliv3r: do you really have to go through such lengths?
<Epsylon3> i suspect google to have made something to force use of gcc 4.6 ;)
<libv> oliv3r: just attach gnd, and then poke rx around until you see something
<libv> and spewing noise is good, but 01010101 is not too helpful
<Epsylon3> or gcc which is not happy of the google fork
<Epsylon3> :p
<libv> some actual characters would be better :)
<libv> at least, that is my plan :)
<Epsylon3> LD .tmp_vmlinux1
<Epsylon3> yep all was built this time
<Epsylon3> now i need to remove my static changes
<Epsylon3> :p
<oliv3r> libv na your board doesn' even come close to mine :( (see pics)
<libv> oliv3r: i will be attempting the same thing with a coby kyros mid7042 on the weekend
<libv> 10 testpoints
<libv> but it seems pretty obvious which of those are rx/tx
<libv> (from the fcc pics i have seen)
<Epsylon3> 3.4 built on android repo
<Epsylon3> with opentv 3.0.62 defconfig minimized + leds + gpiolib
<Epsylon3> + ext4 security
<Epsylon3> Target boot image: out/target/product/cubieboard/boot.img
<Epsylon3> my linux build server is crazy
<Epsylon3> it doesnt boot if i plug a android device on usb :p
<Epsylon3> disks enumeration seems made on usb before sata
<oliv3r> libv: right now, I only measure 250 - 300 mV on all 5 pads
techn_ has joined #linux-sunxi
<oliv3r> there's a group gnd +2 pads, and 3 seperate pads (without gnd next to it)
<Epsylon3> hmmm wtf
<Epsylon3> Starting download of 5 MB
<Epsylon3> downloading 1 MB ...
<Epsylon3> lock
<Epsylon3> in fastboot
<oliv3r> but spewing out, i think it was the letter G? should give you some nice 1010101
<Epsylon3> i ordered a baseboard
<Epsylon3> just hope my nand will stay alive until i get it :p
<oliv3r> i'll edit my PCB screenshot and mark some things I found. IF the interesting pads are on the bottom, i may be screwed, don't wanna disassemble it that far :S (it still works fine)
<Epsylon3> look like ive some problems
<oliv3r> poor Epsylon3
<libv> oliv3r: do you need serial?
<Epsylon3> that lock at 4MB now
<Epsylon3> after some retries
<Epsylon3> is there a maximum size for the boot image ?
<Epsylon3> mine is 5214KB
<Epsylon3> :/
<Epsylon3> dont had any problem to fastboot yet
<Epsylon3> and im still on the same kernel
<mnemoc> http://imgur.com/a/TlADp <--- pictures of BDD's A10 so-dimm kit.... the module has a battery for the RTC :) and the production device will come with a eeprom for the mac address
<Epsylon3> oups, uboot
<Epsylon3> mini pci ? :)
<Epsylon3> no, so dimm ok
bsdfox has joined #linux-sunxi
bsdfox has joined #linux-sunxi
<oliv3r> libv: actually, was hoping that one of those pads was USBC1 :D
<Epsylon3> oh
<mnemoc> Epsylon3: it exposes all of the pins of the A10, include csi and i2s stuff
<mnemoc> including*
<Epsylon3> damned
<Epsylon3> just got the baseboard
<Epsylon3> well... ordered
<Epsylon3> :p
<Epsylon3> now she need to plane over atlantic :p
<Epsylon3> are all those 0.1 DINS ?
<oliv3r> Libv if I can find the pads for usbc1 (usbc2 is wifi, usbc0 well OtG) i can hook up one of those mini bluetooth modules
<Epsylon3> where did you get that ?
<mnemoc> Epsylon3: you mean the board for this so-dimm? yes, all normal 0.1"
<Epsylon3> yep
<Epsylon3> :)
<oliv3r> bluetooth? dealextreme sells those for 1.80 USD
<Epsylon3> its a prototype ?
<Epsylon3> from you ?
<mnemoc> Epsylon3: pre-production kit. I got it for $work$ ... hopefully my boss will accept to make a custom board for it for our products
<Epsylon3> :)
<Epsylon3> else ask to open source the schematics ;)
<Epsylon3> maybe we can make it with PCBBench or such companies
<mnemoc> i think the guys who made the module will open source the reference design for the carrier board
<Epsylon3> right, hmm both are needed
<mnemoc> for the module I got the "pin out"
<Epsylon3> do you have a mini pci wifi card to compare ?
<mnemoc> it's 6.77cm x 5.21cm
<mnemoc> the module
<Epsylon3> the height is not important
<mnemoc> http://linux-sunxi.org/BDD_A10_CoM <--- pinout, work in progress
<Epsylon3> hmm, no there is too much pins after the hole
<mnemoc> what connector is that?
<Epsylon3> mini pci
<Epsylon3> double face
<Epsylon3> of pins
<Epsylon3> smd has 2 sides too
<Epsylon3> one under the board
<libv> mnemoc: cool, so bdd is opensource friendly, nice
<libv> oliv3r: sounds like you have no other option than to access the bottom of the pcb
<mnemoc> libv: next goal is lima-powered laundrymats :p
<libv> mnemoc: running q3a timedemo all day long
<mnemoc> :)
<libv> at least it beats watching laundry go round
<rm> interestingly enough
<Epsylon3> hmm
<rm> the stock hostapd from debian seems to work fine
<rm> not requiring a hacked version
<Epsylon3> where is stored the temporary fastboot image ?
<Epsylon3> in ram ?
<Epsylon3> (to understand if its a usb problem or nand one)
<Epsylon3> nevermind... fastboot flash recovery out/target/product/cubieboard/boot.img
<Epsylon3> lol
<Epsylon3> nah... im damned, same
n01 has quit [Quit: leaving]
<Epsylon3> somebody really dont want i work on the device :p
<Epsylon3> 2 gcc failures
vicenteH has quit [Ping timeout: 264 seconds]
gzamboni has quit [Ping timeout: 260 seconds]
<wingrime> Epsylon3: how about a13 suppotrt
christopher has joined #linux-sunxi
gzamboni has joined #linux-sunxi
<oliv3r> Libv probably, i have a semi-foto of the bottom. i'll open it up again, d/c everything and take a foto of the bottom
eebrah has quit [Ping timeout: 245 seconds]
<wingrime> mnemoc: I tryed to diff sun4i_usb and sun5_usb
<wingrime> mnemoc: It looks like hardware same
<wingrime> mnemoc: 100 %
<wingrime> mnemoc: but sun_5i have differene code version
<techn_> wingrime: Epsylon3: cmw10 a13 support is wip by me
<wingrime> techn_: what statys
<wingrime> ?
<wingrime> status
<techn_> wingrime: trying get kernel to boot
<techn_> seem to work now
<mnemoc> wingrime: that happens a lot, that's why we try to unify the drivers *before* starting fixing/cleaning/improving
eebrah has joined #linux-sunxi
ssvb has quit [Quit: Leaving]
gzamboni has quit [Read error: Connection reset by peer]
<hramrach> wingrime: about the erroors I am getting with the kernel panic on usba cable reconnect
<wingrime> ?
<hramrach> I get both ep already claimed warning and cleaning all urb even when using the adb
<hramrach> so that's probably normal or not an issue
<hramrach> but the problem is that on driver reload and irq reset tehre is some list corrupted
ssvb has joined #linux-sunxi
<hramrach> the driver says cleaning all urbs and crashes dereferencing null in a list function
<wingrime> mnmeoc: I need docs for rewriting about usb on sunxi
<hramrach> figuring out what corrupts it might be a bit tricky because at that time it's already corrupted :s
<wingrime> hramrach: sunxi-usb look like pice of shit
<wingrime> hramrach: I realy Hard to understand relation between code
techn__ has joined #linux-sunxi
techn_ has quit [Ping timeout: 252 seconds]
<hramrach> and the gadget code does look better but is not really easier to understand
<hramrach> not much time for debugging sutff at the moment
gzamboni has joined #linux-sunxi
<mnemoc> wingrime: the lack of documentation is the main reason why we prefer to refactor the code instead of rewrite
<mnemoc> err
<mnemoc> wingrime: for the for unintended highlighting
<mnemoc> f*
<mnemoc> wingrime: sorry the for unintended highlighting
wingrime has quit [Ping timeout: 256 seconds]
<gzamboni> wooot
<gzamboni> very nice
<gzamboni> let me read it ;)
<mnemoc> also got their schematics for the vga/db9 if you want it
<mnemoc> gzamboni: the post is mostly for the pictures :p
<gzamboni> i will be back in France Monday, i will add my cubie ext board to the wiki, and send you one also.
<gzamboni> Are they selling it already ?
<gzamboni> i guess you are a beta-tester
<mnemoc> gzamboni: mine is pre-production, but I think they are already taking pre-orders
<gzamboni> they are from US ?
<mnemoc> yes
<mnemoc> but built in shenzhen.... obviusly
<gzamboni> ;)
<mnemoc> they also said allwinner had delayed the A20 another week :<
<gzamboni> i have only the camera sensor to go. i have talked to drachensun about it
<gzamboni> i saw someone selling it at alibaba.com
<gzamboni> or maybe it was the A31
<gzamboni> dont remember
<mnemoc> i think I have an csi patch from drachensun pending.... .oO
<gzamboni> he told me there are several modifs in your personal github
<gzamboni> i will check it all this evening when i will get back to the apartment
<gzamboni> im in Brazil, came here for the work
<mnemoc> :)
<mnemoc> please don't hesitate to poke me about forgotten patches
gzamboni has quit [Ping timeout: 260 seconds]
bsdfox has quit [Ping timeout: 260 seconds]
gzamboni has joined #linux-sunxi
shineworld has quit [Read error: No route to host]
rz2k has quit [Read error: Connection reset by peer]
<oliv3r> mnemoc: forget my patches, all of them; i'll redo them all
<mnemoc> oliv3r: [unstared] :)
<oliv3r> the touch driver seems to work-ish i guess
<oliv3r> it loads and shows interrupts
torqu3e has quit [Read error: No route to host]
<oliv3r> but i have no way to test it does anything
<oliv3r> fedora just says '/device/virtual/input or something silly
<oliv3r> and i don't have a regular /dev/input/ node
torqu3e has joined #linux-sunxi
<oliv3r> which is why i need turl and Epsylon3 to hurry with their android thing ;)
<oliv3r> also, i need to find out how i best 'swap' the kernel out
<oliv3r> i guess the 3.0 one should be 'compatible'
<oliv3r> wingrme! argh he's gone again
<oliv3r> mnemoc: zeitecsemi (zet6221) has sent me a zip with a reference driver for the A10
<oliv3r> mnemoc: also, some of these drivers depend heavily on the A10 :S
<oliv3r> also solomon systech responded to my datasheet request and sent it to me; (dl it allready)
<hramrach> :o
<oliv3r> not good?
<hramrach> what does solomon do?
<hramrach> unusual to get a datasheet when not already published
<oliv3r> hramrach:to be fair it is
<oliv3r> solomon does the ssd2553 touch screen controller
<oliv3r> zeitescemi does a ts too :)
<oliv3r> wingrme was REing one
<oliv3r> but now i received some reference implementation
<oliv3r> so will send that over to him
<hramrach> well, maybe ts is not so secret as other devices
shineworld has joined #linux-sunxi
<oliv3r> oh! int he reference code, is the firmware included :)
<oliv3r> it's an array in a header file, but still, firmware is firmware
<hramrach> that might be quite useful
<oliv3r> probably need format permission to redistribute it in the official kernel
<oliv3r> but i know wingrime was speaking of it beign a missing bit
<hramrach> if the firmware is not present normally or can be erased accidentally
<oliv3r> well a) the C file is GPL; the C includes the H
<oliv3r> then also, header files can't be 'copyrighted' in or something
<oliv3r> and it is included with GPLed code so that makes it automatically gpl?
<oliv3r> i think there's some docs in the linux kernel how to properly mark those
<hramrach> well no
<hramrach> but it's included in gpl licensed file so yes
<oliv3r> so that makes the firmware gpl licensed aswell?
<hramrach> so long as the file is actually gpl licensed
<hramrach> but it also means they don't adhere to gpl when they only provide firmware binary in the form of C array
<hramrach> so long as the firmware does what is should I don't really care
<hramrach> maybe they want to relicense it if they don't want to provide the source for that
<oliv3r> i do care, but firmware for those kinds of things, are too far fetched atm ;)
<oliv3r> well it's two files, a C file and a H file
<oliv3r> the C file is about half the size of the H file, and IS gpl licensed
<oliv3r> the H file has no license at all, only the array
<oliv3r> but the H file is included in said C file
<oliv3r> ideally, we'd request the source code for the firmware :p
<hramrach> that means nothing
<hramrach> since 'headers cannot be licensed'
<oliv3r> hramrach: so what does that mean with regards to the firmware int he header?
<oliv3r> either way, i'm pretty sure we have to obtain official permission from zeitcesemi to redistribute the firmware with the kernel
<hramrach> unknown license
<oliv3r> actually
<oliv3r> it isn't required to redistribute
<oliv3r> since you only flash it into your controller if you break it, it's flashrom, so once in, it should stay there safely
<hramrach> I guess it would be good idea to get clarification of the license
<hramrach> and probably distribute the firmware as a separate .bin
<hramrach> at least it will make it smaller ;-)
<oliv3r> yeah put it as a bin, and host it only on linux-sunxi, it doesn't ever need to be loaded (unless you break it) so having it licensed will help, but is much less a requirement
<oliv3r> now if we could get the SOURCE for it; that'd be much cooler
<hramrach> well, it it works it's ok
<hramrach> sure, you could possibly change th protocol if you had the sauce
<hramrach> and make all ts the same protocol
<hramrach> but meh
<hramrach> and fix firmware bugs if present
<hramrach> I have a Wacom tablet that locks up :S
<oliv3r> OHhhhh noo
<oliv3r> this isn't firmware
<oliv3r> look
<oliv3r> statci unsigned char zeitec_zet622x_firmware[] __initdata = { ... }
<oliv3r> so it could be (maybe it is firmware, i don't know) just maybe be the 'init' blob? setting up stuff?
<oliv3r> nah, it's too big for that
<hramrach> weird naming
<oliv3r> yeah,
<oliv3r> i'll admit i don't understand
<oliv3r> it's static, i get, unsigned, sure, char; yep; firmware[] an array, logical
<oliv3r> but what does __initdata mean there?
<hramrach> wip
<oliv3r> it's not a comment
<hramrach> actually
<hramrach> it's some flag
<hramrach> and if you look for where it is defined you will see what flag
<oliv3r> it's not
<oliv3r> it isnt' defined anywhere int he code
<oliv3r> is it a gcc thing?
<hramrach> or kernel
<RaYmAn> it's some sort of linking flag for gcc iirc
<hramrach> or they forgot to define
<RaYmAn> well, kernel I guess
<RaYmAn> or both :P
<RaYmAn> " Tagging uninitialized global
<RaYmAn> variables __initdata works, and moves the variables from .bss to .data."
<oliv3r> ah ok
<mnemoc> techn__: what bootloader gives you that panic? what device?
<techn__> mnemoc: lichee-dev and sun5i (a13_mid)
<mnemoc> uhm... techn: seems like lichee-dev is not passing the meminfo
<techn__> mnemoc: I had to modify u-boot to get that far.. I'll send patches later
<techn__> but after enabling atag mem it boots much further
<techn__> now it seems to freeze on uart stuff
<techn__> [uart]: used uart info.: 0x03 <- last print
<mnemoc> techn__: that usually means you forgot to pass the console's ttyS baudrate
<hramrach> or does it inint the uart so you stop seeing stuff?
<techn__> mnemoc: it's there
<hramrach> where?
<techn__> console=ttyS0,115200
<hramrach> also the driver might be broken :s
<mnemoc> techn__: bad script.bin info?
<hramrach> so even with the right parameter it does something wrong to the port
<techn__> mnemoc: that could be the case
<mnemoc> techn__: did you adapt it for mmc0's uart?
<mnemoc> if not, your serial output simply started going to a different place
<techn__> mnemoc: yes.. changed uart 0 enabled and to correct pins
<hramrach> maybe try consoly=tty0 to see if that helps
<hramrach> hmm, but would have ro remove serial info from script bin
shineworld has quit [Quit: Leaving]
<hramrach> either way the serial driver would not frob the port and whatever causes output to stop will not happen
<techn__> I'm wondering if I need to disable uart_para1
<mnemoc> you don't lose anything by trying
<techn__> but checking uart driver first
<hramrach> well, try disabling both for starters
<mnemoc> also be sure to not have any pin conflict
<hramrach> no uart driver then ;-)
<mnemoc> you need one for /dev/ttyS0
<hramrach> the kernel is well capable of writing to the port as configured by u-boot
<hramrach> so technically you don't for tx only
<mnemoc> i mean for userspace/login shell
<mnemoc> getty will need it
<oliv3r> Oh, TI AM335X touch screen driver is new in 3.8.3; TI are capable devs, 3.8.3 is relaly new, i shall use that as my template
eebrah has quit [Quit: Leaving]
<techn__> that did the trick :)
<mnemoc> doh
<mnemoc> good catch
<techn__> now I have android booted to ICS :)
<mnemoc> \o/
<techn__> oh.. a13 :p
<techn__> but still black screen :(
<mnemoc> details ;-)
<techn__> and I'm too tired to continue today :p
<mnemoc> :)
<RaYmAn> it seems kind of malplaced to be on the microsd breakout page. Shouldn't it be on the UART page instead?
<oliv3r> wtf is a 'cell' in the mfd driver?
<oliv3r> was looking where sensors should live, looks like 'mfd' was invented for that
Dave77 has joined #linux-sunxi
<Turl> mripard: yay! we got a reply from Mike
paulk-desktop has quit [Quit: Ex-Chat]
Dave77 has quit []
bsdfox has joined #linux-sunxi
bsdfox has joined #linux-sunxi
Wetomelo has joined #linux-sunxi