Turl changed the topic of #linux-sunxi to: Allwinner/sunxi development discussion - Don't ask to ask. Just ask and wait! - See http://linux-sunxi.org | https://github.com/linux-sunxi/ | Logs at http://irclog.whitequark.org/linux-sunxi | #blameOliver
wingrime has quit [Ping timeout: 272 seconds]
jinzo has quit [Quit: Leaving]
Black_Horseman has joined #linux-sunxi
wolfy has quit [Ping timeout: 252 seconds]
wingrime has joined #linux-sunxi
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
pseudomind has joined #linux-sunxi
wolfy has joined #linux-sunxi
paulk-collins has quit [Remote host closed the connection]
wolfy has quit [Ping timeout: 252 seconds]
enrico_ has quit [Quit: Bye]
wingrime has quit [Ping timeout: 252 seconds]
popolon has quit [Quit: Quitte]
\\Mr_C\\ has quit [Quit: .]
<Turl> dmesg
<Turl> hm, this is not my terminal :|
Nyuutwo_ has joined #linux-sunxi
Nyuutwo has quit [Ping timeout: 244 seconds]
<pseudomind> lol
<pseudomind> If I were a little faster... I would have copy pasted a bit of the dmesg output on my system ;-)
Nyuutwo_ is now known as Nyuutwo
wingrime has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
hipboi_ has joined #linux-sunxi
wingrime has quit [Read error: Operation timed out]
wingrime has joined #linux-sunxi
hipboi has quit [Ping timeout: 264 seconds]
<Skaag> arokux, the 3.4 kernel works great
<Skaag> is it possible to disable GPU acceleration, to reclaim the RAM it uses?
<Skaag> thanks, checking it out
<Skaag> hm. how do I set kernel boot parameters with uboot? I put the line in a boot.scr file in /boot/ ?
<Skaag> I don't currently have a boot.scr in there, all I have is uImage and script.bin
<Turl> yeah, boot.scr or uEnv.txt
<Turl> but it's not just dumping them on there, it takes some extra stuff
<Turl> something like "setenv extraargs <insert your stuff here>" should work
<Turl> inside of a uEnv.txt file
<Skaag> I put this in /boot/uEnv.txt: setenv extraargs console=ttyS0,115200 sunxi_ve_mem_reserve=0 sunxi_no_mali_mem_reserve panic=10
<Skaag> doesn't seem to free up memory, but maybe that's still wrong
<Turl> Skaag: can you paste the dmesg output somewhere?
<Skaag> I don't see that it found it
<Skaag> I must have done it wrong
<Turl> yeah it's not on the cmdline
<Turl> Skaag: did you put uEnv.txt on the first partition's root, next to the uImage and script.bin?
<Skaag> uImage and scrip.bin are in /boot/ not in the root folder
<Skaag> should it be in the root folder instead?
<Skaag> didn't help
Sonicadvance1_ has joined #linux-sunxi
Sonicadvance1 has quit [Remote host closed the connection]
Sonicadvance1_ is now known as Sonicadvance1
<Turl> Skaag: ah, the wiki env is outdated
<Turl> it needs to be in the root if it's a fat partition, and either root or boot/ if it's ext
<Skaag> it's ext
<Skaag> but is the name correct? uEnv.txt or boot.scr?
<Skaag> the /boot partition is vfat
<Skaag> but the / partition is ext3
<Turl> it goes on your first partition on the card, whatever that is
<Skaag> the boot partition is /dev/mmcblk0p1
<Turl> ok then, there, on the root of that filesystem
<Skaag> so it's the first
<Skaag> ok it's there. and the name is right? uEnv.txt?
<Turl> yes
<Turl> see if uboot prints "Loaded environment from uEnv.txt"
<Skaag> rebooting
<Skaag> reboots do not always work, it seems
<Skaag> I have to power it off/on
<Skaag> it doesn't say that
<Skaag> the kernel params are still the standard: [ 0.000000] Kernel command line: console=ttyS0,115200 root=/dev/mmcblk0p2 rootwait loglevel=8 panic=10
<Skaag> those files seem to be ignored
<Turl> Skaag: stop your uboot at the console and run printenv, paste it somewhere
<Skaag> ok
<Skaag> ok bootscr=boot.scr
<Skaag> and root=/dev/mmcblk0p2
<Skaag> so I guess that's where it will be looking for it
<Skaag> right?
<Skaag> brb
<Turl> not necessarily
<Turl> see bootenv and loadbootenv
Benn_2013 has joined #linux-sunxi
Benn_2013 has quit [Quit: -a- Android IRC 2.1.3]
<Skaag> you're right, bootenv=uEnv.txt
<Skaag> there's a bootenv=uEnv.txt and bootscr=boot.scr
<Skaag> and the loadbootenv= line has an || with boot/${bootenv}
<Skaag> so it should in theory try there too, right?
<Skaag> ext2load mmc 0 $scriptaddr boot/${bootenv}
<Skaag> still nothing, I put it in both partitions' root folder
<Turl> try running the pieces manually on the console
<Turl> eg "run loadbootenv;"
<Turl> see what it says
<Skaag> reding uEnv.txt
<Skaag> 66 bytes read
<Skaag> I don't see extraargs set
<Skaag> it should have been set, right?
<Turl> you need to import it
* Skaag is a uBoot noob!
<Turl> env import -t ${scriptaddr} ${filesize}
<Skaag> :-\
<Skaag> in the script or in uBoot interactive shell?
<Turl> on the shell
<Skaag> I see
<Turl> if that doesn't define extraargs, try editing uEnv.txt to look like
<Skaag> anyway I tried adding those mali no reserve parameters in the extraargs manually and it does not like it a little bit... console is screaming bloody murder
<Turl> extraargs=blahblah
<Skaag> ok
<Turl> you're not loading mali are you? :) you must not load it if you set that
<Skaag> ah
<Skaag> however sunxi_ve_mem_reserve=0 worked nicely
<Skaag> now instead of 330mb I have 413mb
<Skaag> that's already a huge improvement
<Skaag> i'll try to put mali in comment in /etc/modules
<Skaag> so x11 graphics will work without mali, right? mali is only 3d stuff?
<Skaag> all I need is simple 2d graphics
<Skaag> awesome
<Skaag> with extraargs= it works
<Skaag> in the /boot/uEnv.txt file
<Skaag> and x11 graphics works, with 2d, and I have now 478mb ram
<Turl> yes, mali is 3d (gles) only
<Skaag> this is fantastic beyond words, thank you so much for guiding me through this
<Turl> if you're not using g2d acceleration you can disable that too
<Skaag> I am assuming the rest of the memory between 478mb and 512mb is used by the 2d engine?
<Turl> you're welcome :)
<Skaag> hm. I don't see any g2d module loaded
<Turl> that's a 34M difference
<Turl> wiki says you may lower the fb memory size too by 16M, have you done that?
<Skaag> no
<Skaag> doing that now :)
<Skaag> I also did sunxi_g2d_mem_reserve=0
<Skaag> I believe I am not using 2d acceleration
<Skaag> maybe I should?
<Skaag> 494mb!
kz1 has quit [Quit: kz1]
kz1 has joined #linux-sunxi
<Skaag> thanks :)
[7] has quit [Disconnected by services]
TheSeven has joined #linux-sunxi
FDCX has quit [Remote host closed the connection]
FDCX has joined #linux-sunxi
FDCX has quit [Read error: Connection reset by peer]
eebrah_ has joined #linux-sunxi
FDCX has joined #linux-sunxi
eebrah_ has quit [Ping timeout: 245 seconds]
eebrah_ has joined #linux-sunxi
tzafrir has quit [Ping timeout: 252 seconds]
Quarx has joined #linux-sunxi
Black_Horseman has quit [Remote host closed the connection]
tzafrir has joined #linux-sunxi
tzafrir has quit [Ping timeout: 252 seconds]
wolfy has joined #linux-sunxi
wolfy has quit [Ping timeout: 252 seconds]
tzafrir has joined #linux-sunxi
<plaes> has anyone ever successfully sent an email from non-gmail account to linux-sunxi group?
n01 has joined #linux-sunxi
AreaScout has joined #linux-sunxi
wingrime has quit [Ping timeout: 248 seconds]
Soru has quit [Read error: Operation timed out]
Soru_ has joined #linux-sunxi
shineworld has joined #linux-sunxi
Soru_ has quit [Read error: Connection reset by peer]
Soru_ has joined #linux-sunxi
shineworld has left #linux-sunxi ["Leaving"]
Soru_ has quit [Read error: Connection reset by peer]
Soru_ has joined #linux-sunxi
eebrah_ has quit [Read error: Connection reset by peer]
<mripard> n01: in a private branch that I'll rebase/publish when 3.13-rc1 will be out
enrico_ has joined #linux-sunxi
eebrah_ has joined #linux-sunxi
<n01> mripard: ok
<mripard> n01: so, something like in a week or so
<n01> mripard: I was looking at your at91_adc.c. It is not clear to me how do you decide which parameters put in .data field of of_device_id and which put in dts
ZetaNeta has joined #linux-sunxi
<mripard> there's only your RTC DT patch in there for now though
<mripard> n01: at91_adc is a mess from that regard.
<n01> ok, so how can I decide where to put specific parameters? it seems to me that I can use both: .data and dts
<mripard> n01: what I do and appreciate is whatever is board specific goes into the DT, whatever is SoC specific goes into .data
_BJFreeman has joined #linux-sunxi
<n01> it seems reasonable
<mripard> so, things like enabled features on a given generation of the SoC goes into the data section
<mripard> while which voltage it's connected to on that given board comes from the DT
_BJFreeman is now known as BJfreeman
<n01> ok ... blurred line anyway
paulk-collins has joined #linux-sunxi
<plaes> argh
<plaes> eventually all 3 emails arrived :S
maz_ has joined #linux-sunxi
rz2k has joined #linux-sunxi
BJfreeman has quit [Quit: had a good time]
deasy has joined #linux-sunxi
<n01> jeeez, aw docs sucks badly
<mnemoc> n01: benn (from cubietech) was asked by aw about how to improve their docs, and he asked us for feedback. you can mail him about it too....
NotRelevant has joined #linux-sunxi
<NotRelevant> hi guys. I am trying to access my 3.5" HDD from Android (on cubieboard2). I already have cubian installed on the HDD (and booting from the microSD card) and it's working as expected. I feel like I am one mount away on Android. Here are the logs
<NotRelevant> root@android:/ # dmesg | grep -i sata <6>[ 1.065715] ata1: SATA link up 1.5 Gbps (SStatus 113 SControl 300) root@android:/ #
<NotRelevant> root@android:/ # dmesg | grep -i sda <5>[ 1.173267] sd 0:0:0:0: [sda] 312581808 512-byte logical blocks: (160 GB/149 GiB) <5>[ 1.173642] sd 0:0:0:0: [sda] Write Protect is off <7>[ 1.173658] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00 <5>[ 1.173791] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA <6>[ 1.219591] sda: sda1 <5>[
<NotRelevant> this part is a little bit strange because I have a GPT and several partitions (5)
<NotRelevant> but it seems that only one is visible
<NotRelevant> _etc/vold.fstab
<NotRelevant> contains
<NotRelevant> dev_mount sata /mnt/sata auto /devices/platform/sw_ahci.0
<NotRelevant> but /mnt/sata is empty
<NotRelevant> and mount doesn't show the HDD
<Dreadlish> meeh
<Dreadlish> you have to mount it before it would be accessible
<Dreadlish> mount /dev/sda1 /mnt/sata
<Dreadlish> if you have there gpt - you should have it enabled in kernel
<NotRelevant> There is no /dev/sda (or sda1), but there is /dev/block/sda1. And the mount
<NotRelevant> root@android:/data # mount /dev/block/sda1 /mnt/sata Usage: mount [-r] [-w] [-o options] [-t type] device directory 1|root@android:/data #
<NotRelevant> There should be ext4 partitions, but
<NotRelevant> root@android:/data # mount -t ext4 /dev/block/sda1 /mnt/sata mount: Invalid argument
eebrah_ has quit [Ping timeout: 264 seconds]
<n01> mnemoc: ok, the problem is that just the register description is not enough. I would expect at least a description of how the peripheral is supposed to work
<NotRelevant> Dreadlish: dd if=/dev/block/sda1 of=file.bin bs=1024 count=1 works, for example
<bfree> NotRelevant: does your kernel have CONFIG_EFI_PARTITION?
<mnemoc> n01: sure, but tell that to benn, not to me :p
<NotRelevant> bfree: I have no idea. I am using cb_a20_android42_v1.05-8188eu-en_US.img. Is there a way to find out from the installed Android?
<n01> lol ok
<NotRelevant> bfree: there is no /proc/config*
<mnemoc> those images use the kernel from allwinner's SDK. not the community one
eebrah_ has joined #linux-sunxi
<NotRelevant> is anyone here that can access a SATA drive from Android?
torbenh3 has joined #linux-sunxi
<NotRelevant> bfree: I have an empty HDD and I will try with MBR
NotRelevant has left #linux-sunxi [#linux-sunxi]
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
mkutsevol has joined #linux-sunxi
<mkutsevol> Hello.
<mkutsevol> Are compiled fex files passed to the sunxi-kernel or are the directly executed by bootloader?
jemk has joined #linux-sunxi
arokux2 has joined #linux-sunxi
<arokux2> mripard: may I ask you what do you mean by " because of a minor thing in the compatible-related thing (on sun4i (Allwinner A10) the reset parts are in the same registers as clock control ones, so we won't be able to use this driver in that case, which makes the compatible a bit different)"
<arokux2> mripard: I have successfully used your driver for USB resets.
<arokux2> Turl: PLL6 is not in the tag https://github.com/mripard/linux/commits/sunxi-clk-for-3.13 which is requested for pull by mripard :(
<mnemoc> shouldn't that be merged into sunxi-devel ?
* mnemoc confused
<arokux2> sunxi-devel contains all the patches that are available for mainline afaik
<mnemoc> mkutsevol: script.bin is preloaded by the bootloader and then read by the kernel
<mkutsevol> mnemoc: that means that a mailline kernel will not be able to configure the SoC?
<arokux2> mkutsevol: mainline has nothing to do with script.bin
<mnemoc> mkutsevol: 3.10+ uses dtb instead of script.bin
<mnemoc> but it works in a similar way, the bootloader preloads the .dtb in a given address
<mkutsevol> CAn you give me link to any device tree for a20?
<mnemoc> look at arch/arm/dts/sun7i*
<libv> plaes: i have rewritten your addition to the wiki, as i completely rewrote that section, but i believe i nicely captured what you meant :)
<mkutsevol> thank you.
<libv> do we have a .fex best practice guide btw?
<libv> or do we just accept any old dump from script.bin?
<arokux2> mnemoc: so please extend and maintain this one instead of sending small e-mails that might get lost. also: Benn isn't the only one who works with AW
<mnemoc> libv: sunxi-boards usually takes three commits per board, first the blind import, then the corrected dram_para, and then disabling the stuff which is not really there
<mnemoc> arokux2: fair enough, i didn't know that page
<libv> mnemoc: i am fleshing out http://linux-sunxi.org/New_Board_howto
<mnemoc> :)
<libv> mnemoc: i will mention the two separate steps for the adding the .fex to sunxi-boards
<libv> is creating a dtb a fully manual process?
<mnemoc> currently yes, but part of the sunxi-3.10 idea is to teach fexc to generate .dts files
<arokux2> libv: yes
<libv> or is it just too early days still to automatically create dtbs from .fex with some extra info
<libv> oh, dts, right
<libv> ok
<plaes> libv: script.fex for gemei-g9 and zatab were identical
<mnemoc> sunxi-3.10 is intended to be full featured, even if drivers are crap, that will give a good overview of all we need in the dts model
<libv> plaes: interesting :)
<libv> plaes: some different bits do seem to be bolted to the mb
<mripard> arokux2: what will happen if both your clock driver and reset driver access the same register at the same time ?
<mripard> => crap.
<mnemoc> plaes: is a10meminfo indentical as well?
<libv> plaes: is this the gps and usb you mentioned in the howto?
<mripard> this is what I meant
<plaes> mnemoc: I couldn't find meminfo for zatab
<plaes> and g9 seems to have extra usb hub chip
<libv> plaes: that's part of our u-boot repo
<mripard> arokux2: and I merged in that pull request whatever patches Turl told me to
<plaes> and doesn't have gps
<mnemoc> mripard: is your sunxi-next still maintained?
<libv> plaes: doesn't seem to be in there :(
<arokux2> mripard: so you mean some kind of locking is needed?
<plaes> yeah :S
Gerwin_J has joined #linux-sunxi
<mripard> mnemoc: yes, why?
Gerwin_J has quit [Client Quit]
<mripard> arokux2: yep
<mnemoc> mripard: uhm... because my robot hasn't built anything since oct 16.
<mripard> I haven't merged any patch since then I guess
<mnemoc> i see
<mripard> with the merge window and everything, the various stuff we got merged aren't yet public, so I can't merge them into sunxi-next :)
<mnemoc> fair enough :)
<arokux2> mripard: it would be very nice if the patches you are submitting would also appear in some branch at your github account.
<mnemoc> i thought that was sunxi-next actually
<mnemoc> "accepted but not released yet"
<mnemoc> while sunxi-devel for "not yet accepted"
<arokux2> mnemoc: the patches that are submitted for review. those that are not even accepted.
<mnemoc> ok, so they are in sunxi-devel's realm
<mnemoc> Turl!
<arokux2> mnemoc: yep, but is difficult to maintain such a repo, I think....
<mnemoc> but it's needed to allow normal humans to test the stuff
<mnemoc> and provide feedback before the thing actually hits mainline
<arokux2> in my case I maintain usb host branch everybody can test and there are nightlights of it. this is more spot-of-interest oriented
<mnemoc> that's exactly why the nightlies thing was made for ;-)
<arokux2> mnemoc: no worries :) those errs are not that important, they can be eliminated but there won't be usb support anyway. my first priority is to (at last....) add OHCI and submit all the usb host patches against mainline for review!
<mnemoc> :)
<libv> plaes: i have flesh out step 2, what do you think about it?
<libv> i think that there will be some opposition to my "directive" there, but imho we really should start tightening up things, to make sunxi much more accessible for new users
fredy has quit [Excess Flood]
<plaes> step 2 looks good
fredy has joined #linux-sunxi
<libv> plaes: how do you feel about this mid7042 page?
<plaes> it should mention hw
<libv> plaes: since you are new to sunxi, would this have been useful to you (if it contained any actual info)?
<plaes> not at first, but eventually it would make it easier to figure out what drivers it needs
<libv> plaes: i mean, would the mid7042 structure have been useful to you?
<plaes> yes
<libv> as opposed to just listing device marketing, dmesg, detailed hw
<plaes> I used other pages as examples
<plaes> s/pages/devices
eebrah_ has quit [Ping timeout: 252 seconds]
<plaes> though, I still think the marketing specs are useful, when one considers getting a device
chkktri has joined #linux-sunxi
<chkktri> hello, are there any known issues with sound on cubieboard1 using latest kernel? Using archlinuxarm and there are no sound devices detected by ALSA
<arokux2> archlinuxarm uses not the latest kernel
<jelly-home> chkktri: if you can't see either the hdmi device or the analog one, something is wrong with your kernel
<arokux2> chkktri: ^
<chkktri> arokux2: i know it does not so i replaced it with one from nightly builds
<jelly-home> chkktri: what does uname -a say?
<chkktri> which might not be a nice idea, i know
<jelly-home> are you using sunxi-3.4 or mainline or what?
<chkktri> linux-sunxi-stage-3.4-sun4i-latest.tar.xz
<chkktri> this package from nightly builds
<arokux2> chkktri: on the contrary, it is a very good idea, since archlinuxarm has an outdated kernel
<chkktri> i see
<arokux2> chkktri: you did a good choice. now, nobody complained about sound so far....
<chkktri> but currently i'm using kernel built on may 23, from hwpack which is also outdated, i know
<arokux2> Skaag uses sound too
<chkktri> hmm
<arokux2> chkktri: so try it out and tell us how it works :)
<chkktri> also it's kind of confusing that lsmod shows no loaded modules, but that's because of only built one are user, right? (sorry about being newbie here, started using linux platform about 2 month ago)
<chkktri> ok, lemme switch the kernels
<jelly-home> chkktri: was your hdmi monitor connected and powered on at boot time?
<chkktri> i don't have on hdmi monitor, ised ssh to connect to cubie and then dmesg to see what happened
<chkktri> last time i checked ALSA reported no found devices
wolfy has joined #linux-sunxi
<chkktri> okay, kernel Linux dabyu 3.4.67+ #9 PREEMPT Tue Nov 5 21:45:56 CET 2013 armv7l GNU/Linux
<chkktri> there's only timer device in /dev/snd
<chkktri> and lemme look at dmesg
<mnemoc> check if /proc/config.gz has it enabled first
<chkktri> [ 3.713272] ALSA device list:
<chkktri> [ 3.718612] No soundcards found.
<chkktri> oh.
<chkktri> mc
<mnemoc> or linux-sunxi-stage-3.4-sun4i-latest.config.gz
<arokux2> mnemoc: do you know what exactly should be enabled?
<mnemoc> nope :( I don't even use sound on my laptop....
eebrah_ has joined #linux-sunxi
<chkktri> http://paste2.org/Fd778KZk and this is what i have
<chkktri> maybe, comparing this to may kernel is a good idea
<arokux2> let me ask you, did the sound worked before?
<chkktri> yes, it worked on old kernel which came with arch and kernel from hwpack
<arokux2> chkktri: what was the kernel in hwpack?
<chkktri> let me boot it
<chkktri> >Linux dabyu 3.0.76 #8 PREEMPT Thu May 23 14:03:25 UTC 2013 armv7l GNU/Linux
<Skaag> arokux2, right, i'm using sound
<Skaag> good morning people
<arokux2> hi Skaag, have you managed to got it working with sunxi-3.4?
<Skaag> it works pretty good, too, especially since Turl helped me last night with freeing up more memory by disabling memory reservation for mali and g2d
<Skaag> yes, I am on 3.4 now
<Skaag> very happy with it, it went pretty smooth I have to say
eebrah_ has quit [Read error: Connection reset by peer]
<Skaag> It feels more 'complete' than the 3.0 branch somehow
<chkktri> neat, and you're using cubieboard 1?
jinzo has joined #linux-sunxi
<arokux2> Skaag: awesome, chkktri now has some problems with it
<Skaag> no I'm on the A13 by Olimex
<chkktri> ah, okay
<Skaag> I would love to play with the cubieboard but I can't afford to do it right now (time wise)
<arokux2> Skaag: did you need to enable something in config?
<Skaag> I compiled the kernel with high resolution timers, and I enabled alsa high resolution timer feature
<Skaag> this improved stability I believe
<Skaag> I only did this because it was complaining about it in syslog
<Skaag> (alsa was complaining)
<plaes> libv: what should I do next? I can start dismantling the device only tomorrow (serial port + pictures)
<arokux2> Skaag: would you please paste you dmesg?
<Skaag> chkktri, do you have a pcm device under /dev/snd/ ?
<arokux2> dmesg | curl -F 'sprunge=<-' http://sprunge.us
<Skaag> sure
<arokux2> Skaag: <chkktri> there's only timer device in /dev/snd
<chkktri> Skaag: no, i don't
<Skaag> interesting
eebrah_ has joined #linux-sunxi
<Skaag> http://sprunge.us/UDWW <- my dmesg
Black_Horseman has quit [Remote host closed the connection]
<chkktri> on a side note: where do i look for disabling memory cutting by mali and maybe something else?
<Skaag> this is what I used
pacopad has joined #linux-sunxi
<Skaag> in /boot/ put a file called uEnv.txt and this is what I use on my platform: extraargs=sunxi_ve_mem_reserve=0 sunxi_no_mali_mem_reserve sunxi_fb_mem_reserve=16 sunxi_g2d_mem_reserve=0
<Skaag> I mean you put that line in uEnv.txt in /boot/ and reboot
<Skaag> you can also try this out with interactive uBoot, just stop the boot process countdown, and type that in
<chkktri> i thought uEnv is put on that FAT partition
<Skaag> on my system /boot is the vfat partition (and the first partition)
<Skaag> and / is an ext3 partition, and the second partition
<chkktri> ah, that's wise
<Skaag> I'm not sure I even had a say in it, I think I followed some guide and that was the outcome :)
<libv> plaes: the marketing details are not more important than the real info that people want when they look up a device on our wiki though :)
<arokux2> chkktri: can you try to modprobe sunxi-codec ?
<libv> plaes: but sadly, that is usually how this thing works, people create an initial wiki page, then lose interest
<chkktri> sec
<libv> plaes: as for disassembly, be very careful, but then, you have seen the slatedroid pics
<arokux2> chkktri: disregard this, this should be built in
<libv> plaes: it might make more sense to start working through a full build
<libv> plaes: creating an SD card from scratch, while adapting uboot
<libv> plaes: also, do you have a device capable of reading 3.3V serial?
<libv> plaes: a standard pc 12X serial -> usb does not work
soul has joined #linux-sunxi
soul is now known as Guest84939
<arokux2> chkktri: please post the full output from the dmesg with non-working kernel
<arokux2> chkktri: dmesg | curl -F 'sprunge=<-' http://sprunge.us
<chkktri> Skaag: did you modify ftab to achieve that?
<chkktri> okay
Soru_ has quit [Ping timeout: 272 seconds]
<plaes> libv: yes I do
<plaes> I have worked with uart before
<chkktri> crap
<chkktri> wrong pc
<libv> plaes: cool :)
<chkktri> http://sprunge.us/GMKL here we go
<arokux2> [ 2.588235] sunxi-codec: probe of sunxi-codec failed with error -16
<chkktri> just noticed that
<chkktri> what does error code -16 means?
<mnemoc> #define EBUSY 16 /* Device or resource busy */
<arokux2> mnemoc: that wasn't taken into stage, right? http://thread.gmane.org/gmane.comp.hardware.netbook.arm.sunxi/3937
<mnemoc> not yet
<arokux2> mnemoc: but that seems to be a bit different issue..., still look if it should be taken to stage.
<mnemoc> actually never saw it :(
<arokux2> mnemoc: no worries, bugs will always find us
eebrah_ has quit [Read error: Connection reset by peer]
<mnemoc> :p
Guest84939 has quit [Read error: Connection reset by peer]
<arokux2> chkktri: do you have some time to help with debugging of this issue?
<chkktri> yep, whole night
eebrah_ has joined #linux-sunxi
Soru has joined #linux-sunxi
<Skaag> chkktri, sorry, had to shower... my /etc/fstab: http://sprunge.us/NPiX
<chkktri> oh, thanks
<Skaag> and the only modules in /etc/modules are 8192cu and sun4i-gpio
<Skaag> and videodev and uvcvideo for my usb camera stuff
<arokux2> chkktri: ok, first we need to know where the error is coming from, I have added outputting of some additional information. you have linux-sunxi git repo cloned, right?
<chkktri> yes, i do
<arokux2> chkktri: it is the easiest if you can add my branch with modifications. proceed as follows: git remote add aroku https://github.com/arokux/linux.git
<arokux2> chkktri: git fetch arokux
<arokux2> chkktri: git checkout -b sunxi-sound-bug arokux/sunxi-3.4-sound-bug
<arokux2> chkktri: please make sure you have this commit (I've just added it) 2fe3a406eddf
<arokux2> chkktri: you can do: git show 2fe3a406eddf
<chkktri> mm, i get fatal error on first line
<chkktri> which is " git remote add aroku https://github.com/arokux/linux.git"
<arokux2> chkktri: show the full error plz
wolfy has quit [Ping timeout: 252 seconds]
<arokux2> (I misspelled, change aroku to arokux, this shouldn't have cause any error though, but next commands are relying on arokux)
<chkktri> uh, right
<chkktri> okay, done
<arokux2> git show 2fe3a406eddf <--- showed something for you?
<libv> getting fel mode does not sound like it will be suited for average joe user
<chkktri> yep, some more code added
<chkktri> guess, now i need to compile the kernel?
<arokux2> chkktri: good, yep, now compile and boot it. I need to make sure that was the place the error is coming frome
<arokux2> from*
<chkktri> :)
<chkktri> where do i change used compiler? arm-linux-gnueabihf-gcc is used but i have arm-linux-gnueabihf-gcc-4.7
<chkktri> or maybe i just create symlink..
<arokux2> chkktri: that depends how you use it.....
<chkktri> nevermind that, forgot to instal package
eebrah_ has quit [Read error: Connection reset by peer]
<chkktri> ll*
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
<arokux2> jemk: you there? are you happen to work on porting GMAC to sunxi-3.4?
tzafrir has quit [Read error: Operation timed out]
<jemk> arokux2: gmac for sunxi-3.4 doesn't need much porting, it's in the cubieboard repo.
rellla has joined #linux-sunxi
<arokux2> jemk: yep, but has anybody done the effort of "copying" it to the sunxi-3.4 and testing if it actually works?
<jemk> arokux2: i copied it and it compiled, tomorrow i hopefully have access to the boards and can test it
<arokux2> jemk: awesome, would you please then submit the patch against stage/sunxi-3.4 to the mailing list?
<arokux2> jemk: hm.. if you have your work public / or get get it public now I can test now.
<jemk> arokux2: i only cherry-picked the commit and it seems to work (i hope i didn't miss something)
<arokux2> jemk: alright, will give it a shot.
rellla has quit [Quit: Nettalk6 - www.ntalk.de]
<torbenh3> how is the status of a20 in mainline ? i just got a cubietruck, and would like to have an overview of what needs to be done...
<arokux2> torbenh3: this is an up-to-date page
<arokux2> torbenh3: GMAC could be added to mainline...
<torbenh3> i saw a few claims, that its an IP, similar to some other thingy, thats already supported. is that true ?
<arokux2> torbenh3: "that is similar" -- yes, but nobody knows to which extent
tzafrir has joined #linux-sunxi
<jemk> torbenh3: yes, the ip i nearly identical, only totaly different glue around, but it doesn't look like this existing driver is easy to reuse
<chkktri> ..great, got make error №2, and have no clue what exactly triggered it. well, LD maybe
<torbenh3> jemk: which driver is that ?
<jemk> torbenh3: stmmac
<arokux2> chkktri: always just paste the error message
<arokux2> jemk: I guess in that case stmmac should be rewritten to support both...?
<chkktri> just this
<torbenh3> arokux2: if they are very similar, thats likely to be requested.
<jemk> arokux2: the dwmac (the ip related part) in stmmac could be reused if it can get split out somehow.
<arokux2> chkktri: do not mess around with sunxi-bsp, you have your rootfs already on the sd-card, right? so you just need to recompile the kernel and write it to the sd-card. so cd to the directory with the kernel and do "make magic" there.
<jemk> arokux2: but the stmmac driver contains many more stmicro specific things, but i didn't look too deep into this, so no guaranty
<chkktri> er, okay
<arokux2> chkktri: also, may I ask you about your general goals with cubieboard/sunxi? if your intentions are to hack around the kernel it is worth investing your time into nice sd-card free set up, loading kernel over network and rootfs on NFS mount share. there are nice docs how to do it.
<torbenh3> jemk: does there exist docomentation for the gmac ?
<torbenh3> is there u-boot support for the gmac already ?
<jemk> torbenh3: no and no
<torbenh3> :S
<arokux2> torbenh3: so you see how important it is...
<chkktri> arokux2: i wanted to try messing with GPIO so i could play music using old sound generators from arcade machines and consoles, but i got stuck at stage where i need TTL <-> CMOS converter. Currently i just want to set up pulte audio/JACK server and mpd so i can listen to music while nothing else is powered up
<arokux2> chkktri: I see. you could go with an old kernel of course, but helping sunxi.org to find the bug in sunxi-3.4 would be awesome!
<arokux2> chkktri: since you are not going to do any kernel development, setting up NFS is probably an overkill right now.
<chkktri> i think putting rootfs on internal nand will be enough for me, but only when things get more or less stable
<chkktri> sftp is enough for me
<torbenh3> arokux2: mmm... hacking u-boot without jtag is not so much fun...
<arokux2> torbenh3: then start with the kernel
eebrah_ has joined #linux-sunxi
Gerwin_J has joined #linux-sunxi
Gerwin_J has quit [Client Quit]
<arokux2> hno: :(
<arokux2> torbenh3: you can find easier tasks if you like
<torbenh3> does that mean, mainline doesnt even have smp support ?
<arokux2> torbenh3: it is something that some individuals work on
<arokux2> torbenh3: this is the last news I think: http://lwn.net/Articles/562118/
geecko has quit [Quit: Quitte]
geecko has joined #linux-sunxi
geecko_ has joined #linux-sunxi
<arokux2> torbenh3: do you want to get something specific working, or just hack? :)
<arokux2> mnemoc: do you know how to generate "As of 2013-07-19 the list contains" at https://github.com/linux-sunxi/u-boot-sunxi/wiki
<arokux2> mnemoc: ?
<torbenh3> arokux2: actually just hack ;)
<torbenh3> arokux2: but hacking kernel without tftp is annoying.
<arokux2> torbenh3: exactly!!! :)
<arokux2> torbenh3: that is why I'm going to test gmac with sunxi-3.4 now
<torbenh3> well... i could live with kexec, though :)
<arokux2> torbenh3: how could that help?
<mnemoc> arokux2: grep sun.i boards.cfg | awk '{ print $7; }' is what I use in https://github.com/linux-sunxi/sunxi-nightly/blob/master/build_u-boot.sh
<mnemoc> but it could be a single awk call
<torbenh3> arokux2: you create a kernel with an initrd, which brings up the network (since gmac in kernel seems to be working, then you load the dev kernel, and kexec it....
<arokux2> torbenh3: ok, maybe I need to try this.
Soru__ has joined #linux-sunxi
<arokux2> torbenh3: do you thing gmac could be something too complicated for you to start with?
<mnemoc> arokux2: awk '/sun.i/ {print $7;}' boards.cfg
<torbenh3> arokux2: no. my dayjob is getting linux to run on embedded hardware....
<arokux2> mnemoc: :) it will output capitalized names.. however on make all lowercase names should be passed. so my question is... should we change everything to lower case in boards.cfg?
<mnemoc> arokux2: | tr 'A-Z' 'a-z'
<mnemoc> it was hno's choice to use capitals, I won't comment on that
<torbenh3> arokux2: its just that i dont have so much freetime in front of computer.
Soru has quit [Ping timeout: 264 seconds]
<arokux2> mnemoc: hno: shouldn't we change everything in boards.cfg to lower case? (you cannot use a board name with upper case letters with make)
<torbenh3> arokux2: btw... if you try to use kexec with 3.4 ... you will need a few patches, to get it running with devicetree.
<arokux2> torbenh3: the point is all the stuff that is left is not small, afaik
geecko has quit [Ping timeout: 244 seconds]
<mnemoc> arokux2: make FooBar_config works fine
<libv> hah
<libv> coby is defunct
Gerwin_J has joined #linux-sunxi
<torbenh3> ARM: 7539/1: kexec: scan for dtb magic in segments
<torbenh3> ARM: 7540/1: kexec: Check segment memory addresses
<torbenh3> ARM: 7555/1: kexec: fix segment memory addresses check
<arokux2> thanks torbenh3 if I use it I'll document it on the wiki
<arokux2> torbenh3: don't be scared by the time, just pick something and work on it as much as you can...
<chkktri> so, i ran kernel & module build procedure but i can't find neither uImage nor modules folder anywhere
<chkktri> am i missing something again?
<chkktri> > make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- uImage modules
<chkktri> this command
<mnemoc> modules_install ?
<mnemoc> and dtbs if 3.10+
<arokux2> chkktri: uImage should be at arch/arm/boot/uImage
<chkktri> it's noth there
<chkktri> not*
<arokux2> chkktri: mm... you could have set some O=... by sunxi-bsp
<arokux2> so the output is at some other location
<chkktri> well, even find gives nothing
<chkktri> but i guess i should make clean in bsp directory and try again?
geecko_ is now known as geecko
<arokux2> chkktri: make clean in linux-sunxi
<chkktri> done
<torbenh3> arokux2: what branch do you suggest for a start ?
<torbenh3> arokux2: i see sunxi-3.10 ...
<torbenh3> you guys seem to be talking about sunxi-3.4 all the time.
<mnemoc> sunxi-3.10 is still an experimental project
<mnemoc> mixing the mainline work, and the crap drivers, into a feature full dtb based stable kernel. but far from there yet
<torbenh3> mnemoc: where are the crap drivers sitting ?
<mnemoc> 3.4
popolon has joined #linux-sunxi
<libv> torbenh3: getting this kernel into usable shape is a gigantic project. constantly chasing the upstream version only leads to further breakage and will not solve anything real. it's like running inside a hamsterwheel
<arokux2> libv: "this" -- which?
<libv> anything sunxi
<libv> our full featured kernel(s) are far from usable as is
<libv> they usually just pass the "it kind of works, a bit" mark
<libv> and i am really rather tired of people whining about "why are you not upstream" or "why are you not running this kernel version which was just released"
<libv> i will face similar useless crap with lima when i release the mesa driver
<libv> i will get a load of people whining about this code not being part of mesa or the kernel side not being in linus' tree
<arokux2> libv: hm.. I'm not sure I understand what you suggest? :)
<arokux2> libv: if we are lucky some "whining" ppl help us squash bugs... if not, we are just working slowly towards perfection for our own satisfaction
<libv> while those same whiny folk will either never be running my code, or, they wouldn't bother upgrading their kernel or significant parts of their installation, to get my code
<libv> my strategy for not blindly chasing upstream will mean that i, in contrary to the other open ARM GPU projects, might actually attract some users
<libv> arokux2: i guess i am just more generally whining myself, or anti-whining, which is essentially the same
<arokux2> libv: you've been in oss project way to longer then me, you should have got used to this kind of situation! :)
<torbenh3> where does sunxi-3.4 create all the platform devices ? i only see it creating 4 in sw_pdev_init() and init_machine is not calling more...
<libv> arokux2: where on this page does it describe what works and what doesn't?
<libv> arokux2: because if that is not described, then you're just misleading and quite quickly, discouraging, people
<mnemoc> torbenh3: currently each driver checks the script.bin on load, and fail if the device is not enabled there
<arokux2> libv: here you can find it here what works, etc: http://linux-sunxi.org/Linux_mainlining_effort
<torbenh3> OMG :)
<arokux2> libv: and thanks to mnemoc we have nightlies!
<mnemoc> torbenh3: but i'm working on turning them into more decent platform_device/platform_driver approach processing the script.bin once at the begining to create the platform_devices there
<libv> arokux2: please make it clear that that page contains the current status
<arokux2> libv: you seem to know better how to say it, so please go ahead: http://linux-sunxi.org/index.php?title=Linux_mainlining_effort&action=edit
<torbenh3> mnemoc: where does the script.bin come from ?
<arokux2> libv: not that I'm forwarding work to you, it is just not quite clear for me what are you talking about.
<arokux2> torbenh3: from <board_name>.fex file
<mnemoc> and, like the .dts files, describes what's available on that particular board
<arokux2> torbenh3: then fex2bin utility is used and script.bin is born
<torbenh3> and u-boot passes it on kernel loading using atags ?
<mnemoc> but it's allwinner specific, not universal as DT
<mnemoc> torbenh3: preloaded on a blessed address
<jelly-home> hmm, no fex for mele m5 yet
eebrah_ has quit [Ping timeout: 272 seconds]
<arokux2> jelly-home: m5 being A31-board?
<libv> arokux2: hrm so mmc does not work...
<libv> nor does nand
<jelly-home> arokux2: A20
<arokux2> libv: yes, the only possible way is USB HDD right now
<libv> nor does sata.
<libv> nor does sound
<libv> or disp
<libv> anyone who accidentally ends up on http://linux-sunxi.org/Mainline_Kernel_Howto will be rather unpleasantly surprised
<arokux2> libv: sata is something what is worked on
<mnemoc> arokux2: anything to include in the nightlies?
<arokux2> libv: what have you expected? we are all unpaid starving hackers without any documentation!
<libv> i am not complaining that it doesn't work yet, i am complaining that the howto doesn't warn average joe user about this
<libv> but i will fix this now
<mnemoc> average joe should just use 3.4
<arokux2> libv: somebody who is taking a look at that page isn't supposed to be average joe :)
<mnemoc> google can lead joes there
<mnemoc> and poor joe will be awfully frustrated
<arokux2> mnemoc: as to the hard core book on particle physics, so what?
<mnemoc> *g*
<mnemoc> die joe die
<arokux2> mnemoc: so each page of the hard core book on particle physics shouldn't warn you about how complicated it is, should it? :p
<arokux2> hno: how about returning to the sunxi family...? u-boot misses your love.
<mnemoc> maybe we should do a template to warn about the intended audience of those pages
<mnemoc> like "HOT!!!" in the coffee cups
<libv> arokux2: it's a wiki, and it should collect information for everyone, and should present some information so that average joe user can get his/her device working without losing all of his/her hair in the process
<mnemoc> and maybe a category related to the hardcoreness level
<mnemoc> (templates can add cateries)
<libv> arokux2: you know what i just did, i typed "kernel" in the search box
<libv> that's what normal people do when they see the great mass of unnavigatable pages listed on linux-sunxi.org
<libv> "mainline kernel howto" pops up twice, and most closely resembles what people would be looking for
<arokux2> libv: you should have input "Linux" :p
<libv> on a website called linux-sunxi...
<arokux2> libv: and then you clearly see [PREFERRED]
<arokux2> and "See also Linux mainlining effort." under the table
<libv> arokux2: why does the "Linux" page not match a search for kernel?
<arokux2> libv: no idea
<mnemoc> because the kernel word is nowhere on it's content?
<libv> maybe a rename of this page to linux_kernel is in order
<mnemoc> or a kernel disambiguation page...
<mnemoc> considering freebsd also has sunxi support
<arokux2> libv: basically you are trying to fix the situation where a random user inputs a random search string. and you want to show him the most relevant page...
<mnemoc> and the wiki also works as http://sunxi.org
<libv> arokux2: "kernel" is not a random search
<libv> arokux2: "linux" equals "ubuntu" in many peoples heads
<arokux2> The first steps: Getting u-boot, a kernel, and a rootfs on an SD card.
<arokux2> this is from the main page
<libv> yeah, that one also needs to be properly fixed
<mnemoc> add a kernel page and a bootloader page, with an small explaination and links to the real places
<arokux2> i'm thinking maybe all those details about different SoCs should go to their own page.
<libv> rm made it massively confusing by insisting he put his freshly made images on top, while this page really just needed a rename
<arokux2> instead we could use the front page to outline the current situation like sunxi-3.4 vs mainline stuff
servili007 has joined #linux-sunxi
<libv> anyway, one thing at a time. first i finish the warning in the mainline howto
<plaes> how big is the possibility that booting from uSD works?
<plaes> for tables with A10 SoCs?
<arokux2> plaes: 100% :)
<plaes> nice
<arokux2> mnemoc: what about moving section "Allwinner SoC's family" to the separate page?
eh2scqw has joined #linux-sunxi
<arokux2> mnemoc: together with the table
<arokux2> mnemoc: this info is useless to joe.
geecko has quit [Read error: Connection reset by peer]
<eh2scqw> something is flooding the syslog of my cubietruck with these messages:
<eh2scqw> CFG80211-ERROR) wl_escan_handler : Couldn't find P2PIE in probe response/beacon
geecko has joined #linux-sunxi
<plaes> now I only need to find my uSD -> SD-card adapter :S
<arokux2> eh2scqw: that something is the driver for WLAN adapter
<plaes> u-boot only requires dram-specific config?
<mnemoc> arokux2: it's a wiki, be bold! :)
<libv> plaes: yup
<eh2scqw> arokux2: thank you. I will have a look at the code of the module
<libv> plaes: i was just about to starting writing up step 5 :)
<mnemoc> plaes: there are prebuilts of every supported board at http://dl.linux-sunxi.org/nightly/u-boot-sunxi/u-boot-sunxi/u-boot-sunxi-latest/
<libv> but wanted to add some links to step 4 first, and then ran into the next inconsistency
<arokux2> mnemoc: ok.. do not tell me I haven't warned you :p
<plaes> mnemoc: I already got mine built (with my board specific dram)
<mnemoc> once a new board gets added, it will show up there in less than 1h
<mnemoc> arokux2: i'm not the owner of the wiki, we all are ;-)
pacopad has quit [Quit: pacopad]
<chkktri> http://paste2.org/1gHma7YZ is it just me or something is missing in there?
<mnemoc> plaes: try to submit the patch so other people with your board has it simpler in future
<plaes> yeah, I have to test it first, though :)
<mnemoc> fair enough ;-)
<arokux2> chkktri: what is this?
<chkktri> it's from the last stage of kernel compiling
<libv> pfff
<chkktri> i'm still getting no files
<libv> the uart page is dead weight as well
<libv> what sort of person spends his days copying bits of wikipedia, and then pats himself on the back for a job well done?
<chkktri> however i can see 47MiB built-in.o file in drivers directory
maz_ has quit [Ping timeout: 245 seconds]
<chkktri> ...and it has ELF header
<arokux2> chkktri: mm.... so you have cleaned linux-sunxi and built is, which what commands?
<arokux2> libv: I had that question too. lots of such "content" was dropped already
<arokux2> mnemoc: why this is not collapsed? http://linux-sunxi.org/Main_Page#Software
<chkktri> arokux2: right
<mnemoc> arokux2: check the categorytree plugin docu
<arokux2> chkktri: you were supposed to run this: make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- uImage modules
<chkktri> and this is exactly what i executed
<arokux2> mnemoc: I'm comparing it with http://linux-sunxi.org/Main_Page#Devices where everything works fine...!
<mnemoc> don't know :|
<arokux2> chkktri: that should have ended with something like this: http://sprunge.us/jaSA ( not exactly this!)
<chkktri> :S
<chkktri> maybe i just clone linux-sunxi into new repo?
<arokux2> chkktri: no, too much time. what is "git branch" saying?
<chkktri> sunxi-3.4
<chkktri> * sunxi-sound-bug
<arokux2> chkktri: perfect, so what is the output of "make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- uImage modules"
<chkktri> thing i pasted was the last line of output (i used V=1 for more detail output)
<arokux2> chkktri: well, V=1 is too detailed for now, what is the output without it
eebrah has joined #linux-sunxi
eebrah is now known as Guest18760
<Turl> plaes: I have
<chkktri> uhh, i have to build it once again, don't i?
<arokux2> chkktri: yes, but it will be quick this time
<chkktri> anyway, it was something like "LD built-in.o"
<chkktri> oh, ok
<arokux2> hi Turl!
<arokux2> Turl: maybe you could include PLL6 stuff into Maxime's pull request or it is too late?
<plaes> Turl: ?
<chkktri> >sound/soc/sunxi/sunxi-codec.c:1491:3: error: implicit declaration of function ‘printf’ [-Werror=implicit-function-declaration]
<chkktri> :P
<arokux2> chkktri: oh, sorry, printf ---> printk
<arokux2> who is Patap?
Soru__ has quit [Read error: Operation timed out]
<Turl> plaes: sent from non-gmail to google groups
<Turl> plaes: sorry, old backlog
<Turl> hi arokux2
<plaes> ah.. ok :)
<Turl> arokux2: the series has some comments I haven't addressed fully yet
<arokux2> Turl: oh, ok
<Turl> and it's pretty late now to be merging stuff, considering we need dt patches for that
soul has joined #linux-sunxi
soul is now known as Guest71790
<Turl> but I'm confident we can get it resolved and queued early for next cycle
Quarx has quit []
Guest71790 has quit [Read error: No route to host]
Soru has joined #linux-sunxi
Soru has quit [Ping timeout: 272 seconds]
Soru has joined #linux-sunxi
<arokux2> checkout the wiki main page
<arokux2> this is an idea right now...
<jemk> arokux2: wow
<arokux2> jemk: so bad?
<jemk> arokux2: looks pretty
<mnemoc> arokux2: now try to open it in a tablet or phone :/
<arokux2> mnemoc: looks nice on my 10" tablet
<mnemoc> impresively it looks decent on my nexus 7 too
<jemk> mnemoc: btw, who is responsible for the wiki software? do a search for "ps/2", it gives massive errors
<arokux2> mnemoc: tables need to be refined, this is the basic idea to have tables
<chkktri> I'n sorry to interrupt you yet again but: http://paste2.org/94YHzMwY
<mnemoc> jemk: looks like a mediawiki bug...
geecko_ has joined #linux-sunxi
<arokux2> chkktri: no problem! show your kernel config
<chkktri> should be arch/arm/configs/sun4i_defconfig if I'm getting you correctly
jinzo has quit [Ping timeout: 252 seconds]
<arokux2> chkktri: ok.. a sec...
<arokux2> chkktri: everything is find for "make sun4i_defconfig" for me
<arokux2> libv: what do you think about the new concept for the main page: http://linux-sunxi.org/Main_Page
<arokux2> libv: i'm not sure I like it
<chkktri> grep -r "make sun4i_defconfig" gives me nothing
<arokux2> chkktri: it shouldn't
<arokux2> chkktri: you run it
<arokux2> make sun4i_defconfig
<chkktri> aah
<arokux2> (be sure to back up your old .config)
<chkktri> :D
<arokux2> chkktri: ^^^
eh2scqw has quit [Quit: eh2scqw]
<chkktri> >*** Can't find default configuration "arch/x86/configs/sun4i_defconfig"
<chkktri> wat
<mnemoc> x86....
<chkktri> actually, there was a crapload of promts for various settings
<chkktri> i'm still doing something wrong
<mnemoc> what abotu ARCH=arm ?
<arokux2> chkktri: make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- <your_SoC>_defconfig
<Turl> jemk: mnemoc should be fixed, but we probably should report it to mediawiki
<mnemoc> Turl: is the auto-updating cron running?
<mnemoc> if so, please be sure to commit the fix :)
<Turl> I dunno mnemoc
<arokux2> Turl: how do you like wiki's main page?
<Turl> mnemoc: it's not the best code ever, but it does the trick http://sprunge.us/jYUh
<mnemoc> yes, update and backup tasks are running
<mnemoc> Turl: looks like the wrong fix.... but thanks for fixing it :)
<mnemoc> but nothing with php can be pretty
jinzo has joined #linux-sunxi
<Turl> mnemoc: wrong fix?
<mnemoc> Turl: no, it's fine.... escaping needs to happen there... when composing the pattern
<mnemoc> interestingly wikipedia doesn't have that issue...
<Turl> maybe they run a newer version of wikistuff
<Turl> mnemoc: see the post arokux2 linked
<Turl> arokux2: not bad, but it looks a bit crammed to me
<arokux2> Turl: yes, this is basic concept so far
<mnemoc> Turl: but we do upgrade hourly...
<arokux2> Turl: the question is if this basic concept is better than what we had
<Turl> mnemoc: to 'the latest' or to 'latest stable' of kinds?
<Turl> arokux2: maybe if you drop the howtos column and change the extra devices category list to something that says 'you can find all the other supported devices in [[the devices category page]]' or sth like that it'd look less crammed
<mnemoc> Turl: REL1_20 branch
<Turl> mnemoc: there's a 1.21 rel now
<mnemoc> any reason for upgrading?
<arokux2> mnemoc: is it possible to remove "Main Page" from main page...?
<mnemoc> arokux2: not that I know :|
<mnemoc> but some people makes templates that do javascript or css voodoo turning the h1 into lowercase for example
<Turl> arokux2: it can be done with 1 css line
<Turl> .page-Main_Page #firstHeading { display: none;
<Turl> }
* mnemoc didn't know the made per-page clases...
<Turl> mnemoc: firebug :)
<arokux2> Turl: put it at the page please
<Turl> mnemoc: where does one add custom css to mw?
<mnemoc> it's a wiki page on it's own iirc
<arokux2> <categorytree mode=pages hideroot=on depth=1>Tutorial</categorytree>
<arokux2> how to make it render its contents in several columns....
servili007 has quit [Quit: Leaving]
<arokux2> mnemoc: can we have something like this? http://www.mediawiki.org/wiki/Extension:DynamicPageList_%28third-party%29
Soru has quit [Read error: Connection reset by peer]
Soru has joined #linux-sunxi
paulk-collins has quit [Ping timeout: 245 seconds]
<mnemoc> arokux2: done
<mnemoc> bbl
<chkktri> finally, arokux2, I need mkimage tool, where is it located?
<arokux2> chkktri: depend on your distribution, some package needs to be installed
<arokux2> chkktri: are you using Arch?
<chkktri> no, it's ubuntu, compiling on Atom based notebook will be too time consuming
<arokux2> chkktri: uboot-mkimage then
<chkktri> didn't expect it to be in repositories, heh
<arokux2> chkktri: why??
<chkktri> dunno, last time i used some toolchain i had to compile gcc under cygwin
<chkktri> (that was rockbox btw)
<chkktri> i mean, gcc for arm cpu, also all of tools had to be compiled
<arokux2> chkktri: how many years ago was this last time? :)
<chkktri> 2 years i think
<chkktri> ah, finally, it compiled
<arokux2> Turl: mnemoc any idea? http://sprunge.us/bijF
<Turl> nope :(
<arokux2> Turl: I do not have permission to edit this: http://sunxi.org/MediaWiki:Common.css
<arokux2> Turl: can you please add to it: body.page-Main_Page h1.firstHeading { display:none; }
<arokux2> Turl: disregard
<chkktri> arokux2: http://sprunge.us/gNRH
<arokux2> The text you wanted to save was blocked by the spam filter. This is probably caused by a link to a blacklisted external site.
<arokux2> Turl: ...
<arokux2> Turl: this I get from method 2
<chkktri> arokux2: i guess, i should've seed "failed to register card" somewhere
<chkktri> but there're none
<arokux2> :(
<Turl> arokux2: so I do not add the css then?
<Turl> arokux2: akismet is off so that's from spamhaus xbl or the like
<arokux2> Turl: add it, since I wasn't able to use method 2 too
<arokux2> chkktri: please do: git pull
<arokux2> chkktri: then recompile, boot and show dmesg
<chkktri> mm
<arokux2> chkktri: make sure git show c08592da shows smth
<chkktri> if i delete codec object file and run compile, will it work?
<arokux2> chkktri: why do you ask?
<chkktri> 'coz i don't want to recompile it for 5th time :D
<arokux2> chkktri: not the whole kernel will be recompiled, make will figure out what should be recompiled.
<chkktri> ah, okay
<Turl> arokux2: ok, done
<arokux2> Turl: thanks
<arokux2> chkktri: you could have get some kind of conflict while pull...
<arokux2> Turl: should this die? http://linux-sunxi.org/Mksunxiboot
<Turl> arokux2: is it in sunxi-tools?
<arokux2> Turl: "but already integrated into the mmc branch of u-boot for sunxi."
<Turl> arokux2: last edit is feb 2013 :)
<arokux2> mnemoc should know this, since https://github.com/amery/mksunxiboot/
<mnemoc> that tool was integrated into u-boot-sunxi ages ago
Sonicadvance1 has quit [Ping timeout: 244 seconds]
<mnemoc> the tool still exists, and it's essential to boot from mmc
<mnemoc> just not maintained as a separated tool anymore
<arokux2> mnemoc: so that page can be killed?
<arokux2> mnemoc: is that tool is used by u-boot to do some magic while creating *.bin?
<arokux2> mnemoc: u-boot.bin
<arokux2> mnemoc: also: please take a look at this, http://sprunge.us/bijF
<chkktri> arokux2: http://sprunge.us/LeFJ
iamfrankenstein has joined #linux-sunxi
<arokux2> chkktri: awesome, now we know the error is coming from: http://lxr.free-electrons.com/source/sound/core/init.c#L150
<arokux2> chkktri: that was the easy bit. the complicated one is to understand why.....
<mnemoc> arokux2: the tool converts the spl into a sunxi-bootable spl
<arokux2> mnemoc: so it is currently used by u-boot internally?
<arokux2> mnemoc: my basic question is, should that wiki page be deleted?
<mnemoc> not sure
<arokux2> mnemoc: ok, we'll leave it at that. time will show.
<mnemoc> the fel card also needs it, and any baremetal code you want to boot directly from a card
<arokux2> mnemoc: plz make dpl working.
<mnemoc> can't atm... need to keep prepare for an interview tomorrow morning. sorry :< .... poke Turl
<arokux2> mnemoc: no problem
<arokux2> Turl: how to delete this page? http://linux-sunxi.org/Windowmanager
<mnemoc> done
<arokux2> mnemoc: prepare for the interview!!
Guest18760 is now known as eebrah_
Sonicadvance1 has joined #linux-sunxi
<mnemoc> arokux2: :p
n01 has quit [Ping timeout: 245 seconds]
<arokux2> who the fuck is Rocker and why he dumped so much general stuff to our wiki?????
<moofree> "Nope, can't be bothered with IRC to be honest" lol http://linux-sunxi.org/User_talk:Rocker
<arokux2> idiot ...
<arokux2> Turl: here is the full list of the pages that should be deleted. thanks. http://sprunge.us/SBAR
<arokux2> Turl: if you teach me how to delete a page, I won't bother you again
<mnemoc> mediawiki has tools to deal with people who doesn't know to behave
<arokux2> funny, sound would work on my mele, which is a10 too
<libv> arokux2: the main page seems a little more accessible this way, but it needs further work :)
<libv> but imho it is a step in the right direction
<arokux2> libv: thanks, how-tos are going to be multi column, this will be next improvement
jinzo has quit [Quit: Leaving]
<arokux2> torbenh3: have you picked something to mainline? :)
jemk has quit [Remote host closed the connection]
dapsaille2 has joined #linux-sunxi
geecko_ has quit [Quit: Quitte]
<dapsaille2> Hi
AreaScout has quit [Ping timeout: 245 seconds]
<dapsaille2> it this the good place to ask about .fex options and strange resolution on a13 hardware ? :)
arete74 has quit [Ping timeout: 240 seconds]
<arokux2> Turl: update list of the pages that need to be deleted. http://sprunge.us/iSAV
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
<arokux2> dapsaille2: definitely
bsdfox\ has joined #linux-sunxi
Soru_ has joined #linux-sunxi
<dapsaille2> nice, i'm new to arm/sunxi world and i want to output a 15khz signal trough the vga port of my a13 olinuxino .. something like Modeline "320x224 NTSC (60Hz)" 6.700 320 336 367 426 224 236 239 262 -hsync -vsync
<dapsaille2> i asked on Olimex forum and someone tell me to test some "voodoo" in script.fex .. but my board don't like thoose parameters and say Err04: Violate "(lcd_vt/2) >= (lcd_vbp+lcd_y+2)" with others errors ..
<dapsaille2> is there is a way to convert modelines to script.fex parameters ?
<arokux2> deasy: hi, you there?
Soru has quit [Ping timeout: 272 seconds]
<deasy> arokux2, yup
<arokux2> deasy: you have cubieboard1, right?
<deasy> yup
<arokux2> deasy: there is very strange issue with audio, it won't work on cubieboard1 and nobody knows why. it works on my mele (which is also a10) so I cannot reproduce the problem. can you plz help me to hunt it down?
<deasy> yup
<deasy> i can plug speaker and test
<deasy> i eat something now, wait a bit
<arokux2> deasy: no problem, you'd first need to show me your dmesg and we start from it.
<deasy> who wrote the audio driver :p
<arokux2> Turl: http://sprunge.us/LiLg update list of the pages that need to be deleted.
<mnemoc> arokux2: just ask Turl to give you burocrat permission :p so you can move and delete stuff yourself
<arokux2> deasy: Allwinner engineers
<deasy> an OS driver?
<deasy> or this is from the source without license ?
<arokux2> deasy: all the sources are GPL licenced
<deasy> ok nice
<mnemoc> all code released by allwinner was GPLed
<deasy> since when?
<mnemoc> they initially forgot to add the headers, but they confirmed by mail that everything was GPL and then one @allwinnertech.com sent patches to add the headers
geecko has quit [Ping timeout: 272 seconds]
<deasy> my reading about it say they have release source sometime but ni license...ok nice !
<deasy> so...
<deasy> :D
<dapsaille2> hehe, i agree
* dapsaille2 look at it's mediatek smartphone .... angry ...
<arokux2> dapsaille2: there were some releases of MK code too
<dapsaille2> not the framework for android :x
<dapsaille2> long flaming post on Xda :)
<arokux2> ok...
<deasy> someone had do a note on 10 or 20 or 100 about the open(?) ness? of SoC builder ?
<arokux2> deasy: send me your dmesg please
<deasy> with a table :p
<deasy> wait ! almost finished my fruit
<dapsaille2> does someone can help me to translate a modeline to .fex lcd_ parameters ? :)
<deasy> i don't want juice split kill my board :p
<arokux2> dapsaille2: the guys keen on that part seem to be absent now... they are ssvb wingrime etc
<dapsaille2> ok good to know thanks, for now i've resolved half the magic of translation :)
<arokux2> deasy: yes! we need it for testing!
<arokux2> dapsaille2: maybe you can post your findings to our wiki?
<dapsaille2> when done, with pleasure, i have some friends who are waiting for my tests results
<dapsaille2> we want to use theese boards on our arcade cabs :)
eebrah_ has quit [Ping timeout: 272 seconds]
<dapsaille2> and crappy composite is not an option ^^
naobsd has quit [Quit: Page closed]
<Turl> arokux2: welcome to the bureaucracy
<deasy> lol
<deasy> arokux2, back
<deasy> i plug it
<deasy> i start cubian
<Turl> arokux2: next to the edit/history button you should see an arrow, there's the stuff to delete pages
<arokux2> Turl: i have only watch and move
<Turl> arokux2: try logging off and in again
<Turl> my vps provider is failing again and only let me with v6 connectivity >.<
<Turl> usually it's the other way round
<arokux2> Turl: still the same
<Turl> mnemoc: any ideas? ^
<mnemoc> Turl: I removed ipv6 stuff iirc
<mnemoc> never got it working correctly on lxc
<arokux2> Turl: no network with cubietruck in u-boot, right?
<Turl> mnemoc: I meant on the wiki stuff :p
<Turl> arokux2: nope
<Turl> mnemoc: apparently bureaucrats can't nuke pages
<mnemoc> :\
eebrah_ has joined #linux-sunxi
<mnemoc> Turl: try to give them delete
<mnemoc> and ... protect?
<Turl> ok, delete given
<mnemoc> funny... bureaucrats can set permissions, but not do editing related changes
<Turl> try now arokux2
<Turl> mnemoc: they also can asksql, that sounds dangerous
<arokux2> Turl: YES!
<arokux2> now I'll delete!
<Turl> arokux2: delete responsibly
<mnemoc> Turl: seems like we want people to be admin, not bureaucrats
<arokux2> Turl: it will be the list I've shown.
<Turl> mnemoc: admin is "wiki root" isn't it?
<mnemoc> seems like admin is content manager
<mnemoc> all their permissions are page-related
<mnemoc> while bureaucrats can do permissions, rename users, disable the whole wiki, and query the db
<Turl> mnemoc: the 'query the db' part scares me
<Turl> that basically means they can do anything :p
<mnemoc> that's why I mean we misunderstood what "administrator" means
eebrah_ has quit [Ping timeout: 265 seconds]
<mnemoc> and bureaucrats are more powerful than admins
<mnemoc> starting with the simple "userrights" permission!
<mnemoc> bureaucrats can change their own permissions
<mnemoc> so that's the risky role, not admin
<Turl> mnemoc: http://linux-sunxi.org/index.php?title=Special:ListUsers&group=bureaucrat
<mnemoc> we should reduce the number of bureaucrats, and turn those wrongly made bureaucrat-only into admins
<mnemoc> because that's the idea. content control. not db, not user permissions
<Turl> +1
<mnemoc> and people who has been inactive for a long time, back to normal users
<Turl> mnemoc: /q
<arokux2> do you know some file sharing service?
<Turl> arokux2: dl.linux-sunxi.org :)
n01 has joined #linux-sunxi
Black_Horseman has quit [Quit: Αποχώρησε]
Black_Horseman has joined #linux-sunxi
geecko has joined #linux-sunxi
Sonicadvance1 has quit [Ping timeout: 264 seconds]
eagles0513875 has quit [Changing host]
eagles0513875 has joined #linux-sunxi
eagles0513875 has quit [Changing host]
eagles0513875 has joined #linux-sunxi
Sonicadvance1 has joined #linux-sunxi
tzafrir has quit [Ping timeout: 272 seconds]
rz2k has quit []
<arokux2> Turl: please ban this bastard in our wiki: http://linux-sunxi.org/User:Rocker
<Turl> arokux2: you are an administrator, you should be able to ban people
<Turl> arokux2: personally I wouldn't go so far though
<arokux2> Turl: why? he dumped the wikipedia to our wiki, do you think it was good intention?
<Turl> arokux2: I checked some of his articles back in the day, they didn't match wikipedia
<Turl> arokux2: I believe it was good intention, just that it was on the wrong wiki
<rm> popcorn.gif
<Turl> elinux might have been a better place
<Turl> rm: haha
<arokux2> Turl: he copied stuff though
<arokux2> and put two lines with sunxi
<Turl> yeah, most contributions weren't really relevant
<Turl> but they weren't spam either
<Turl> he did some categorizing and other stuff too
<arokux2> Turl: ok
<arokux2> I'll keep a close eye on him
wolfy has joined #linux-sunxi
<Turl> arokux2: check the OpenSSL, GnuTLS, LUKS pages, they may be mostly irrelevant too
<arokux2> Turl: :$
<arokux2> Turl: I wonder ppl edited these pages!!!!!!
<arokux2> other smart ppl!
<dapsaille2> damn .. i lost 2 hours for understand the logic of modelines and fex parameters ... now i can see in xorg log that i got the right modeline but .. no picture on crt ...
* dapsaille2 go to bed ....
<arokux2> Turl: deleted them all
n01 has quit [Ping timeout: 244 seconds]
<deasy> arokux2, dictator :p
<arokux2> for deleteing spam?
<deasy> just laughing :)
<arokux2> :p
<arokux2> deasy: do you have a usb sound card?
<deasy> nop
<arokux2> deasy: any other exotic usb devices?
<deasy> hmm
<deasy> bluetooth and sim card reader
<arokux2> sim card reader.. you hacker :)
<deasy> nop
Soru has joined #linux-sunxi
<deasy> one of my customer was not able to use it
<deasy> so he gives me it
<deasy> i have never use it
<deasy> just see than it works
<torbenh3> arokux2: i made gmac driver compile on mainline. need to write code to fill platform data from dt, but i hope, i can take that from stmmac.
<arokux2> torbenh3: does it work?
atsampson has quit [Quit: time to break everything!]
<arokux2> torbenh3: in the shape it is now
<torbenh3> arokux2: without platform driver, it wont work.
<torbenh3> s/driver/device/
<torbenh3> but that should be crated from DT.
Soru_ has quit [Ping timeout: 272 seconds]
<torbenh3> it needs more work... next WE :)
<arokux2> torbenh3: alright :)
<arokux2> torbenh3: can you maybe make your work public?
<arokux2> torbenh3: like put it on github?
<torbenh3> i just messed with it about an hour, and also spent time comparing gmac with stmmac.
<arokux2> torbenh3: but you got it compiled, that is a lot already
<arokux2> I suggest you write about it to ML, ppl will take a look at your code, help etc.
<torbenh3> i dont think thats a lot.
<arokux2> in some cases lots of stuff is removed/changed and it is not to make something compile in 3.12 that was done for 3.4
<torbenh3> arokux2: network didnt change that much. only the dmabuf recycling got removed.
<torbenh3> arokux2: and i could just peek at the stmmac driver.
<arokux2> torbenh3: in any case you should write a line about your effort here, so the others know: http://linux-sunxi.org/Linux_mainlining_effort#Work_In_Progress
Soru__ has joined #linux-sunxi
<torbenh3> arokux2: lets see where we stand after next weekend.
<arokux2> ok
Soru has quit [Ping timeout: 272 seconds]
<torbenh3> arokux2: gmac and stmmac are _very_ similar. maybe its easier, to just try stmmac, and fix things up there, that fallout.
<torbenh3> arokux2: i actually have to try booting sunxi-next on the board :) didnt do that yet.
<arokux2> torbenh3: cubietruck?
<torbenh3> arokux2: yes.
<torbenh3> there is at least a DT for it.
<arokux2> torbenh3: I can do it now
pitillo_ has quit [Ping timeout: 240 seconds]
<torbenh3> be my guest ;) anyways. bedtime for me.
<arokux2> torbenh3: ah, dt is not yet merged, is scheduled for 3.13
<arokux2> torbenh3: good night !
pitillo has joined #linux-sunxi
<arokux2> oliv3r: u-boot shows: DRAM: 1 GiB for my cubietruck, but it seems to have 2GB
<deasy> yes cubietruck had 2GiB
atsampson has joined #linux-sunxi
<deasy> 4x512MiB chip
<arokux2> deasy: have you got it too?
<deasy> nop
<deasy> and 8GB storage
<arokux2> here is says 1 or 2
wolfy has quit [Ping timeout: 252 seconds]
<arokux2> this is wrong...
<deasy> IIR the first batch had 1GiB only
<arokux2> which batch?
<deasy> the first board builds for testing
<popolon> 8cores
<arokux2> popolon: 2 x cubietruck price wise
<popolon> still need to wait the odroid version to have a descent price
<popolon> nice price
<deasy> too expensive :p
<arokux2> popolon: odroid will be even cheaper?
<popolon> each time odroid does cheaper card than arndale
<popolon> but everytime few month after
<Turl> cubietrucks have 2G I think
<Turl> but uboot only supports 1G unless you use the special oliv3r branch
<oliv3r> arokux2: you have toguse my u-boot
<deasy> yup Turl
<oliv3r> u-boot has abug with 2 gb
<arokux2> oliv3r: have you submitted patches?
<oliv3r> upstream yes
<arokux2> oliv3r: I'm sorry for maybe asking you again.. but: there is no network u-boot with cubietruck, right?
<oliv3r> but hno needs to rewiew it. s uboot ml
<popolon> 3GB ram
<oliv3r> nope no gmac driver yet and phy seems incompatob;e with emac
<popolon> usb/eth controler is external chip, not in exynos ???
<arokux2> oliv3r: ok, emac is on, though
<popolon> and 10/100 eth only
<deasy> popolon, i see one octa(big little) for odroid
<deasy> 169 $
<popolon> but that's the 8410
<popolon> (PowerVR)
<oliv3r> ye but emac doesnt speakn rmii
<arokux2> oliv3r: is there a patch I can try?
<arokux2> oliv3r: at your github, I guess? :)
<deasy> yup
<deasy> bad bad
<oliv3r> my github has bran h
<oliv3r> onphone in bed
<arokux2> oliv3r: how difficult is it to add GMAC to u-boot on the scale from 1 to 5