Turl changed the topic of #linux-sunxi to: Allwinner/sunxi /development discussion - did you try looking at our wiki? https://linux-sunxi.org - Don't ask to ask. Just ask and wait! - https://github.com/linux-sunxi/ - Logs at http://irclog.whitequark.org/linux-sunxi
apritzel has quit [Ping timeout: 244 seconds]
bwarff has joined #linux-sunxi
curlybracket has joined #linux-sunxi
<wens> vagrantc: bad board design that i want to workaround with a combined regulator
<vagrantc> wens: oh, fun.
sdschulze has quit [Ping timeout: 264 seconds]
mosterta has quit [Ping timeout: 240 seconds]
curlybracket has quit [Quit: Textual IRC Client: www.textualapp.com]
TheSeven has quit [Ping timeout: 250 seconds]
TheSeven has joined #linux-sunxi
cnxsoft has joined #linux-sunxi
ninolein has quit [Ping timeout: 240 seconds]
ninolein has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
<vagrantc> anyone have any idea why linux only sees 1.5GB on a cubieboard4? I would expect something a little closer to 2GB...
<vagrantc> get these messages in the kernel log: [ 0.000000] Ignoring memory range 0x20000000 - 0x40000000
* vagrantc suspects it has something to do with using u-boot 2011.10
tomboy65 has quit [Ping timeout: 264 seconds]
tomboy65 has joined #linux-sunxi
leio has quit [Ping timeout: 244 seconds]
leio has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 260 seconds]
p1u3sch1 has joined #linux-sunxi
reev has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 244 seconds]
p1u3sch1 has joined #linux-sunxi
FDCX has quit [Remote host closed the connection]
p1u3sch1 has quit [Ping timeout: 244 seconds]
p1u3sch1 has joined #linux-sunxi
premoboss has quit [Ping timeout: 240 seconds]
reev has quit [Read error: Connection reset by peer]
JohnDoe_71Rus has joined #linux-sunxi
tomboy65 has quit [Ping timeout: 240 seconds]
tomboy65 has joined #linux-sunxi
reinforce has joined #linux-sunxi
mossroy has joined #linux-sunxi
reev has joined #linux-sunxi
mossroy has quit [Client Quit]
merbzt has quit [Ping timeout: 240 seconds]
merbzt has joined #linux-sunxi
premoboss has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
jernej has quit [Ping timeout: 252 seconds]
yann|work has quit [Ping timeout: 268 seconds]
azend|vps has joined #linux-sunxi
tomboy65 has quit [Ping timeout: 248 seconds]
tomboy64 has joined #linux-sunxi
NiteHawk` has joined #linux-sunxi
pekka10 has quit [*.net *.split]
cajg has quit [*.net *.split]
formruga has quit [*.net *.split]
captainigloo has quit [*.net *.split]
tgaz has quit [*.net *.split]
jero has quit [*.net *.split]
oliv3r has quit [*.net *.split]
vbmithr has quit [*.net *.split]
azend|vps_ has quit [*.net *.split]
NiteHawk has quit [*.net *.split]
ordex has quit [*.net *.split]
longsleep has quit [*.net *.split]
pekka10 has joined #linux-sunxi
formruga has joined #linux-sunxi
captainigloo has joined #linux-sunxi
jero has joined #linux-sunxi
tgaz has joined #linux-sunxi
oliv3r has joined #linux-sunxi
vbmithr has joined #linux-sunxi
longsleep has joined #linux-sunxi
cajg has joined #linux-sunxi
domidumont has joined #linux-sunxi
premoboss has quit [Quit: Sto andando via]
ordex has joined #linux-sunxi
ordex has joined #linux-sunxi
domidumont has quit [Remote host closed the connection]
_massi has joined #linux-sunxi
mcan has quit [Changing host]
mcan has joined #linux-sunxi
domidumont has joined #linux-sunxi
reev has quit [Ping timeout: 248 seconds]
paulk-collins has quit [Ping timeout: 260 seconds]
reev has joined #linux-sunxi
marble_visions has left #linux-sunxi [#linux-sunxi]
cosm has quit [Ping timeout: 260 seconds]
yann|work has joined #linux-sunxi
asmir has joined #linux-sunxi
sdschulze has joined #linux-sunxi
NiteHawk` is now known as NiteHawk
vagrantc has quit [Quit: leaving]
asmir has quit [Quit: WeeChat 1.0.1]
PaddyF has joined #linux-sunxi
sdschulze has quit [Quit: Verlassend]
Net147 has quit [Quit: Quit]
Net147 has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
lemonzest has joined #linux-sunxi
Net147 has quit [Quit: Quit]
Net147 has joined #linux-sunxi
cptG_ has joined #linux-sunxi
cptG has quit [Ping timeout: 268 seconds]
cptG_ is now known as cptG
iamfrankenstein has quit [Quit: iamfrankenstein]
_massi has quit [Remote host closed the connection]
avph has quit [Ping timeout: 246 seconds]
iamfrankenstein has joined #linux-sunxi
avph has joined #linux-sunxi
mpmctoo has joined #linux-sunxi
ricardocrudo has quit [Remote host closed the connection]
ricardocrudo has joined #linux-sunxi
cosm has joined #linux-sunxi
leio has quit [Ping timeout: 246 seconds]
leio has joined #linux-sunxi
punkjazz has joined #linux-sunxi
maz is now known as Guest16678
punkjazz is now known as maz
<keesj> Does the a13 load u-boot / kernel from the area between the MBR and the first partition or not?
<keesj> my u-boot output shows http://paste.ubuntu.com/15383453/ but I don't know what prints Found U-Boot script /boot.scr (the rom or something else)
matthias_bgg has joined #linux-sunxi
<KotCzarny> probably spl
<KotCzarny> spl loads uboot, uboot loads boot.scr
p1u3sch1 has quit [Ping timeout: 244 seconds]
<keesj> sorry I was talking about the a10 (for the a13 I previously written some notes http://keesj.github.io/A13-OlinuXino-playground/ )
<keesj> but so is it the same ?
p1u3sch1 has joined #linux-sunxi
enrico_ has joined #linux-sunxi
mzki has joined #linux-sunxi
iamfrankenstein has quit [Quit: iamfrankenstein]
clonak has quit [Read error: Connection reset by peer]
clonak has joined #linux-sunxi
FDCX has joined #linux-sunxi
clonak has quit [Ping timeout: 240 seconds]
<keesj> any clue on that is failing for the ex4 read in the above pastebin? 3182952 is probably the kernel and 17883 I think the dtb
iamfrankenstein has joined #linux-sunxi
<keesj> (I have an ext2 partition fot he bootloader/kernel and and ext4 for the root file system
<KotCzarny> hmm, check on another machine that files are ok
cosm has quit [Ping timeout: 276 seconds]
<ordex> any suggestion how I could boot my orangepi_mini via tftp ?
<ordex> maybe by flashing something adhoc on the microsd ?
NiteHawk has quit [Ping timeout: 240 seconds]
<keesj> (most probably it can't read the initrd the other files are ok) (I am trying to create a process to clone my sd-card (I don't want to install 5 times :p)
<keesj> so I do thinkg the files are OK
<KotCzarny> dd ?
<KotCzarny> :)
NiteHawk has joined #linux-sunxi
<keesj> I am doing dd + other stuff but the cards have slightly different sizes hence fdisk + mkfs and a little dd magic to copy the SBL sounds like a great plan to me
<KotCzarny> yeah. i personally leave ~1.6GB for os and mount the rest as /home
<KotCzarny> that way you can have one 'install' for all boxes and just create additional partition for user
<keesj> hmm that might have worked (I started with everything on /) and quite a huge swap because android's fastboot somehow likes loading files in memory or similar
<keesj> I cheched the boot.src and the problem is indeed loading the ramdisk (e.g. the thing it does after loading kernel + dtb
<keesj> I copied the bootloading like http://paste.ubuntu.com/15383617/ e.g. dd the first 2047 blocks and dd if back to a different card while skipping the first block (MBR)
Amit_T has joined #linux-sunxi
ninolein has quit [Ping timeout: 268 seconds]
clonak has joined #linux-sunxi
ninolein has joined #linux-sunxi
leio has quit [Ping timeout: 244 seconds]
leio has joined #linux-sunxi
avph has quit [Ping timeout: 246 seconds]
vishnup has joined #linux-sunxi
avph has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 276 seconds]
<keesj> KotCzarny: is it possible that u-boot 's ext2 support is sub optimal? I only did a cp ramdisk.img new.img ; mv new.img ramdisk.img and u-boot was now able to read the full content of the file
p1u3sch1 has joined #linux-sunxi
<KotCzarny> as i said, maybe earlier file was damaged (maybe badblock on the card?)
reev has quit [Ping timeout: 248 seconds]
<keesj> certainly not (it never left the sdcard I just did a copy) hence can't be a back block nor anything else just u-boot and some magic happening
<keesj> I can also be something with the file system (e.g. not probably umounted)
JohnDoe_71Rus has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
<keesj> the sd copy gods are not with me http://paste.ubuntu.com/15383895/
<keesj> so the blkid to boot from is hardcoded somewhere else
<keesj> (the fstab of the root file system) .. can mount root but can find the fstab
orly_owl has quit [Ping timeout: 268 seconds]
<codekipper> ordex: if you have a booting sdcard then all you need to do is change the boot.scr
<ordex> codekipper: mhhh how ?
<codekipper> 2 secs I'll find my settings....the only thing I found is that I couldn't boot over GB on my network
<ordex> thanks !
<ordex> I just to avoid burning my sdcard due to too many reflash
<codekipper> here is mine for my Itead Ibox http://pastebin.com/9Lq0n06j
<codekipper> I'm currently booting rootfs from the sdcard but there is settings there for also booting over nfs
<codekipper> are you still using openwrt?...I'm struggling to build an image with alsa-libs....it has worked recently but not now
<ordex> codekipper: yes, still on openwrt - but I am reflashing my sdcard too many times and i broke it (it was probably cr0ppy)
<ordex> so now i wanted to go with tftp or anything else that i can change on the fly and reboot
<ordex> codekipper: does 0x46000000 depend on the device ?
<codekipper> no that should be device related. I've used the same script on other devices that I've net booted.
<codekipper> s/should/shouldn't
<ordex> oh ok
<codekipper> ordex: could you check if enabling alsa-libs/utils on your build works?
<ordex> codekipper: first I have to make sure it can boot via tftp, otherwise I can't try much :D
<codekipper> I'm desperate for a small rootfs that I can test audio with
cnxsoft has quit [Quit: cnxsoft]
cnxsoft has joined #linux-sunxi
<ordex> codekipper: where do you specify the kernel in your script ?
<codekipper> Ahhh..I've got another script for building the kernel which also copies the files that are required for netbooting to the tftp dir
<ordex> nop
<ordex> let me dig into that :)
<ordex> thanks !
cnxsoft has quit [Quit: cnxsoft]
clonak has quit [Ping timeout: 240 seconds]
<codekipper> one thing to note is that tftp doesn't seem to translate symbolic links which is why I copy the files I need to the tftp dir
<wens> chrooted?
JohnDoe_71Rus has joined #linux-sunxi
<KotCzarny> for nfsroot you only need nfs client enabled in kernel with bootp/dhcp and nfs rootfs somewhere
<KotCzarny> and kernel can be on sdcard, no need for tftpd
afaerber has joined #linux-sunxi
<ordex> codekipper: btw I managed to boot the kernel by following http://linux-sunxi.org/How_to_boot_the_A10_or_A20_over_the_network, but I do not understand how to provide a root image. Must I use NFS ?
reinforce has quit [Quit: Leaving.]
<ordex> mh axtually it gets stuck at Starting kernel ...
<ordex> and the eth led blinks regularly ..
<ordex> likea kernel panic
<ordex> mah mah
<codekipper> ordex: do you have a seperate kernel git or are you just using what's in openwrt
<ordex> codekipper: what's in openwrt right now
<ordex> re-using the same image
<ordex> I'd expect some output at least
<ordex> not stuck right away
<codekipper> 2secs I'll dig out a log
<ordex> :D
<ordex> da man of logs!
<codekipper> old man of memory loss...http://pastebin.com/kbabHk4e
<ordex> mh the eth port is a bit like cr0p, it does not work at 1Gbps - i think this was the warning on the sunxi
<ordex> website
<ordex> I see
<codekipper> If it doesn't find the tftp then it's a good chance that it could be a Gbps issue
<codekipper> I've got a cheap GL.iNet running openwrt to drop 1Gbps down to 100Mbps
maz has quit [Quit: Leaving]
<ordex> nope, tftp is recognised
<ordex> that work
<ordex> now it works
<ordex> now the problem is hostapd - what driver do I use to bring it up ?
<ordex> the next question is, why did I choose a board with such ugly wifi chip ? :D
<codekipper> I never got that far with openwrt...does your eth work?
vishnup has quit [Ping timeout: 260 seconds]
<ordex> codekipper: yup
<ordex> I was ping da internet
<ordex> I was just fighting with hostapd, because it does not want to make the wext driver work
<ordex> *pinging
bwarff has quit [Ping timeout: 260 seconds]
Guest16678 is now known as maz
reinforce has joined #linux-sunxi
JohnDoe3 has joined #linux-sunxi
orly_owl has joined #linux-sunxi
<KotCzarny> ordex, in armbian there is a hostapd compiled for realtek
<KotCzarny> try grabbing the package and see if it works
<KotCzarny> lamobo/banana pi r1
JohnDoe_71Rus has quit [Ping timeout: 250 seconds]
<ordex> mh, ok, I will have a look
<ordex> thanks !
cosm has joined #linux-sunxi
reinforce has quit [Quit: Leaving.]
vishnup has joined #linux-sunxi
<AA781> I hope you folks don't mind if every now and then I ask, what maybe perceived as a stupid question?
<AA781> trying to get myself up to speed kernel development wise for arm boards
jernej has joined #linux-sunxi
<AA781> hey longsleep have you and apritzel been the only ones working on the kernel fixes for the a64?
<AA781> pine specifically I mean
<AA781> let me know
<ordex> AA781: working with the pine64? :)
zuikis has joined #linux-sunxi
<AA781> will be when I get my batch in
<AA781> just wrapping my head around the road ahead
<AA781> x86 amd64 has been my background thus far
domidumont has quit [Read error: Connection reset by peer]
jukivil1 has quit [Ping timeout: 240 seconds]
<AA781> are you working on the pine64 ordex?
ilmaisin has quit [Remote host closed the connection]
gusenkovs_ has joined #linux-sunxi
<gusenkovs_> Patch Applay Sensor not work
mzki has quit [Quit: leaving]
khuey|away is now known as khuey
p1u3sch1 has quit [Ping timeout: 244 seconds]
p1u3sch1 has joined #linux-sunxi
vishnup has quit [Remote host closed the connection]
cosm has quit [Quit: Leaving]
nove has joined #linux-sunxi
p1u3sch1 has quit [Ping timeout: 246 seconds]
p1u3sch1 has joined #linux-sunxi
zuikis has left #linux-sunxi [#linux-sunxi]
bonbons has joined #linux-sunxi
mossroy has joined #linux-sunxi
zuikis has joined #linux-sunxi
pmattern has joined #linux-sunxi
pmattern has quit [Quit: Genug für heute.]
domidumont has joined #linux-sunxi
domidumont has left #linux-sunxi [#linux-sunxi]
enrico_ has quit [Quit: Bye]
JohnDoe3 has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
JohnDoe_71Rus has joined #linux-sunxi
Netlynx has joined #linux-sunxi
afaerber has quit [Quit: Ex-Chat]
gusenkovs_ has quit [Quit: Page closed]
ricardocrudo has quit [Remote host closed the connection]
tkaiser has joined #linux-sunxi
<tkaiser> ordex: Before you try out an Armbian image to fix WiFi or hostapd problems on your Lamobo R1 better read through the forums to get the idea how crappy the R1's WiFi is especially if you try to use it with hostapd.
<tkaiser> The best known way to make the R1 into an access point is to either replace it with something else, desolder the WiFi chip and replace it with something more reliable or use an USB dongle.
<KotCzarny> tkaiser, you didnt get the context
<KotCzarny> he doesnt have bpi-r1
<KotCzarny> he just need realtek patched hostapd
<KotCzarny> *needs
<tkaiser> Ok, but if it's 8192 then it's bad anyway ;)
<KotCzarny> i think its 8188eu
<KotCzarny> but i might be wrong
<zuikis> 8188eu is pretty bad, i managed to scan with it, but paring failed. On the other hand i just might have been doing something wrong...
apritzel has joined #linux-sunxi
kaspter has quit [Ping timeout: 248 seconds]
<AA781> realtek 8188 and 8192 are pretty dodgie in x86 aswell
<AA781> good to see some things stay the same no matter the architecture
apritzel has quit [Ping timeout: 244 seconds]
diego71 has quit [Remote host closed the connection]
diego_ has joined #linux-sunxi
<longsleep> AA781: i have received Kernel fixes from two other people, see the commit logs in the pine64 specific branches
ricardocrudo has joined #linux-sunxi
<tkaiser> longsleep: Care to enable a few USB input devices with your next kernel release? I wanted to test GUI stuff just to realize that none of my peripherals (all from Apple) work connected to Pine64 ;)
<AA781> will take alook
<longsleep> tkaiser: sure, if you can come up with the names the the modules
<longsleep> to many "the"s
<longsleep> one of them is supposed to mean "of"
<AA781> what types of inputs
<longsleep> mice, keyboards the sort i would guess
<longsleep> maybe touch
<AA781> I see
<longsleep> who knows what apple devies tkaiser has :)
<longsleep> +c
<AA781> mind if I pick your brain a moment longsleep
<longsleep> i have a funny problem with audio decoding, it does not work well - in fact Pine64 decodes h265 better than any audio
<longsleep> AA781: sure go ahead
<tkaiser> longsleep: I'll send a PR
<longsleep> tkaiser: ok great
<longsleep> tkaiser: i can roll a build quickly
<tkaiser> Containing a few more HID devices that user might have lying around (@Armbian we went through the same procedure a few weeks ago)
<longsleep> well yes, but most impoartantly bring up the CPU in some state, bring up RAM in some other state, set addresses and other stuff whatever required for the SoC and then run u-boot
lemonzest has quit [Quit: Leaving]
<AA781> ok got it longsleep
<AA781> I'll continue here longsleep
<AA781> reading the link
<longsleep> AA781: arm actually provides reasonable documentation, see https://github.com/ARM-software/arm-trusted-firmware as well for some arm64 related fun
<AA781> like I said new to this
dev1990 has joined #linux-sunxi
<AA781> unlike standard pc scape
<longsleep> AA781: pine64 uses bl31 in between boot0 and u-boot somewhere
<AA781> I will read all this before disrupting you any further
cosm has joined #linux-sunxi
<longsleep> its not that i am the expert, things are actually pretty straight forward once you got the basics
<AA781> and that is what I am lacking for the moment
afaerber has joined #linux-sunxi
<longsleep> AA781: do you know the basics of BIOS or UEFI in x64 land? Its actually similar but diffrent :)
<AA781> that wahts I am trying to learn at this moment
<AA781> I see similarities from looking at things
iamfrankenstein has quit [Ping timeout: 248 seconds]
<AA781> bios is more robust I will say that
<longsleep> AA781: have you ever seen BIOS source code?
<AA781> for ami/phoenix
<AA781> dsdts
<AA781> apci stuff in the past
<longsleep> well, all the things vendors can hide or ship closed source have the potential to be pretty horrible
tkaiser has quit [Quit: jIRCii - http://www.oldschoolirc.com]
cosm has quit [Ping timeout: 240 seconds]
iamfrankenstein has joined #linux-sunxi
<longsleep> ARM boot is very flexible, mainline U-Boot has improved a lot over the years
<longsleep> its the closed source blobs which are scary
<AA781> especially recent bios with ms only friendly acpi
<AA781> loading debian and fine tuning it to a toshiba laptop has never been fun
cosm has joined #linux-sunxi
<AA781> h20 insyde bios
<AA781> is what I have dealt with the last 3 years
<AA781> toshiba garbage
JohnDoe_71Rus has quit [Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/]
<longsleep> AA781: well then you should get it quickly - just dont think complicated as it really is not
<AA781> but I will read what you sent me and I will get back to you or anyone once I am read up on all this
<AA781> thanks for the resouces mate
<AA781> got another month to get the hang of all this
<longsleep> AA781: well, just get another device to play, i have a whole collection of boards most of them pretty useless :)
<longsleep> RPi3 is sold all over the place at the moment
<AA781> I ordered my first pi 2
<longsleep> but more interesting is ODROID-C2 if you can get this where you are
<AA781> can get em cheap since 3 was just released
<AA781> thats the s905 from amlogic correct?
<longsleep> mhm define cheap. RPi3 is 44EUR at the shop next to my office
<longsleep> AA781: yes
<AA781> 39.99 canadian with case ;)
cosm has quit [Ping timeout: 246 seconds]
<longsleep> thats cheap if you ask me :)
<AA781> it is
<AA781> my first developer board
<longsleep> well amazon sure, i was more thinking get to the shop now and try within the hour :)
<longsleep> ah and thats a Pi2
<AA781> not much around where I am at
<AA781> I can wait the 2 days
<AA781> I get those s805s in 2 days aswell
<AA781> got 2 untis I want to experiment on
<longsleep> fair enough, but note that RPi does not use U-Boot by default
<AA781> s805s
<AA781> but the s805 does ;)
<longsleep> they have their own reduced bootloader which can run linux directly
zuikis has left #linux-sunxi [#linux-sunxi]
<longsleep> but thats something you can try to lead the stuff, i think rpi2 support is in u-boot mainline
<AA781> I want to understand the entire spectrum
<AA781> u-boot to kernel
<AA781> I figure the kernel portion should be easy enough
<longsleep> you will want an uart USB adapter and probably some wires to connect to header pins
<AA781> still awaiting on that
<AA781> should be arriving on the 18th for me
mosterta has joined #linux-sunxi
<longsleep> yeah, you need that to do anything. Also get some switches so you can turn it on and off without unplugging all the time
<AA781> most things arm wise are not within the city
<AA781> and I live in Toronto
<longsleep> mhm there must be a hackerspace in Toronto, no?
<AA781> very limited
<longsleep> shame, where do all the nerds go there?
<AA781> online
<AA781> that's about it
<AA781> I do a lot of nux stuff for clients
<AA781> simple load blanced web clusters
<AA781> and setup web based applications for business maainly
<AA781> have had to build some kernels here and there
<AA781> mainly for poorly supported laptops cough cough toshiba
<AA781> btw
<AA781> if you need any resources
<AA781> storage space
<AA781> webstuffs
<AA781> let me know
<AA781> was thinking of putting together pine linux site
<AA781> for new comers
<longsleep> well, if you want to help with pine development try to set up a compile platform, document it nicely
<AA781> that's another thing I have been jumping around on
<AA781> reading materials but knowing what is what in what order
<AA781> lol
<longsleep> AA781: yeah, i am not sure about this - i have intentionally kept the docs low to avoid getting unqualified feedback and end user attention :)
<AA781> if I understand you have modified u-boot to more inline with mainline correct?
<AA781> from what I read in the forums I mean
cosm has joined #linux-sunxi
<longsleep> AA781: no, the BSP U-Boot is ancient, i just backported some functionality
<AA781> as I read somehwere the allwinner provided way is somewhat unconventional
<longsleep> mhm, it is pretty normal for vendors in that price segment
<longsleep> one gets bad quality ancient unclean zip files which contain code which does not compile
<longsleep> pretty normal :)
<AA781> the idea of flooding the market combined with better support might have given allwinner and edge
<AA781> an*
<longsleep> well, i am sure they earn money with their current model
<AA781> true
hulu1522 has joined #linux-sunxi
<AA781> but you can get yourself to a state of brand reliaility too
<AA781> reliability
<longsleep> look at RPi how long it took them to get where they are now
<longsleep> and they do not have a language barrier
<AA781> true
<AA781> british
<AA781> lol
<AA781> so
<longsleep> mhm, sound works like shit on my Pine, did anyone try sound on android? I assume it works there
<AA781> I have heard it works well under android
cosm has quit [Ping timeout: 264 seconds]
<AA781> are we tlaking delays in the audio
<AA781> pops hisses?
<AA781> stutters?
<longsleep> stutters ever .5 seconds
<AA781> or could it be the output frequency
<AA781> ok
<longsleep> pulseaudio is unusable, direct alsa works better
<longsleep> i think its related to DMA
<longsleep> so you can watch movies without sound on Pine64 :)
<AA781> no thanks
<AA781> got a j1900 for that
<AA781> lol
diego71 has quit [Quit: leaving]
<longsleep> yeah do not care much either, but i am interested in the why
diego_ has joined #linux-sunxi
<AA781> so you got cedarx running
<longsleep> after all Ubuntu Mate works well enough, except for audio and firefox
<AA781> decoding video
<longsleep> yes
<longsleep> that was easy thanks to the folks here
<AA781> and you have a stutter with the audio
<AA781> I read that
<longsleep> stutter with the audio, even when playing a simple wav file
<AA781> cedar x is for video only correct
<longsleep> uhm, i think so but i did not look into details there
<longsleep> sound also clips the first bunch of samples almost all the time
<longsleep> really strange
<AA781> from what I read there is a CedarA CedarV
<AA781> aspect
<AA781> to the cedarx
<nove> please don't use the word "cedarx" to describe the hardware, "cedarx" is the name of the allwinner proprietary software
<AA781> it looks like it may have ties ins with audio
<nove> check http://linux-sunxi.org/Category:Video_Engine for name questions
<AA781> sorry nove
<AA781> new here
keh has joined #linux-sunxi
<nove> AA781: there is an audio engine, but only in A10 soc (and maybe older socs)
<AA781> ok
<AA781> ACE for a10 and lower
<AA781> quick question
<AA781> why is there a reference to a cedarx coprocessor if it isn't hardware
<AA781> just wondering
<nove> AA781: where?
cosm has joined #linux-sunxi
<AA781> Efficient use of system resources when decoding multimedia, offloading audio and video decoding to CedarX co-processor so that the CPU can be used for other things, (when CedarX is working as it should).
<AA781> under - Benefits to Allwinner's CedarX technology and libraries:
<AA781> http://linux-sunxi.org/CedarX#Disadvantages_Allwinner.27s_CedarX_technology_and_libraries:
<nove> AA781: outdated information, that was written already 3 years ago
<AA781> ok
<AA781> good ot know
<AA781> to*
<nove> well its a wiki
mossroy has quit [Quit: Quitte]
cosm has quit [Ping timeout: 264 seconds]
<AA781> where is the goto place to look at all work doen for the pine
<longsleep> AA781: pine forum?
<AA781> I mean sources and work done
<AA781> for collabaration
<longsleep> mhm, closes thing probably this IRC channel
<AA781> ok
Guest57997 has joined #linux-sunxi
<longsleep> i narrowed done the audio problem to mp3 and ogg, wav plays fine
<AA781> so something with a library
<AA781> codec
<longsleep> mplayer says "audio device got stuck" - does that mean something to someone? I am using the sound test files from http://www.kozco.com/tech/soundtests.html - eg. not working is piano2-CoolEdit.mp3
doppo has quit [Ping timeout: 246 seconds]
<AA781> and oyu are using alsa no pulse
<longsleep> pulse, works better with alsa but not perfect
<KotCzarny> longsleep, maybe audiodriver is broken
<KotCzarny> or it tries to do hw decoding
<longsleep> KotCzarny: yeah - i have probably enabled to much in the kernel config
doppo has joined #linux-sunxi
<longsleep> mhm paplay works much better than mplayer
<longsleep> so maybe pulse is doing some remixing
cosm has joined #linux-sunxi
cosm has quit [Client Quit]
apritzel has joined #linux-sunxi
<AA781> somethign to do with pulse
<AA781> and mplayer
<AA781> are oyu playing around with volume levels or is there some sort of eq/audio normalization occuring?
<longsleep> no, i am just trying to play sound without hickups in a default ubuntu mate session
<longsleep> currently rebuilding kernel, i had all kinds of test stuff enabled related to sound - turned that off now
<longsleep> btw, just saw that i have CONFIG_HID_APPLE=y, so what does tkaiser need in addition?
<AA781> he said input devices
<AA781> but nothing specific
<longsleep> ah CONFIG_HID_MAGICMOUSE maybe
<AA781> pissbly
<AA781> possibly
Guest57997 has quit [Quit: http://www.kiwiirc.com/ - A hand crafted IRC client]
<AA781> gonna download bsp 1.2
<AA781> part 1 and 2
<AA781> torrnet
<AA781> 24 gigs is huge
<longsleep> thats so funny, they really could have cleaned the thing up first
<AA781> after I am done I will leave it up
<AA781> got a decent 250/20 connection
<AA781> and soon 1gbit/100mbit coming my way
<AA781> seems liek a lot
<AA781> they have what in there
<AA781> everything for building the 3.10 kernel
<AA781> android?
<longsleep> yes, they have their complete build tree as found on a developer machine including all temporary files, .o files and everything else the compile generates
<longsleep> thats why it is so large
<longsleep> and all the dependencies for android
<longsleep> its like if i would tar up my development folder
<longsleep> which is probably what they did
akaizen has quit [Read error: No route to host]
tomboy65 has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 250 seconds]
akaizen has joined #linux-sunxi
<longsleep> mhm new kernel, still audio device got stuck with mplayer -ao alsa and -ao pulse when playing 44.1KHz wav audio
<KotCzarny> try 48k ?
<longsleep> same
<longsleep> tried that
<AA781> 44100
<KotCzarny> move things to other cores?
<longsleep> well, there is no cpu load visible
<longsleep> i think mplayer is bugged - i have to try this on the C2
matthias_bgg has quit [Quit: Leaving]
<KotCzarny> try mpg321 ?
matthias_bgg has joined #linux-sunxi
<KotCzarny> ffmpeg?
<KotCzarny> sox?
<KotCzarny> load oss compatibility module and try -o oss ?
tomboy64 has joined #linux-sunxi
<KotCzarny> -ao oss
<longsleep> yeah didnt compile that in :D
<KotCzarny> compile as modules and copy
<KotCzarny> :)
<longsleep> lets try another player
tomboy65 has quit [Ping timeout: 250 seconds]
<longsleep> yes doing that
<longsleep> ok same issue with mpg321
<longsleep> :/
<KotCzarny> and you say wavs play fine?
<longsleep> with paplay yes
<KotCzarny> but stutters with aplay?
<longsleep> yes
<KotCzarny> then either audio codec gets stalled or cpu
<longsleep> mhm CPU seems to be idle
<longsleep> aplay logs underruns
<KotCzarny> set cpufreq to performance?
<longsleep> yeah, no difference there
<longsleep> i currently test everthing with performance
<longsleep> something is broken and i am missing what it is
<longsleep> maybe tkaiser has an idea once his mouse works :P
tomboy64 has quit [Ping timeout: 248 seconds]
tomboy64 has joined #linux-sunxi
<KotCzarny> google: arm alsa underrun
<longsleep> ok
<AA781> should have everything downloaded in the next 3 hours
cajg has quit [Ping timeout: 244 seconds]
<AA781> knock on wood
tomboy65 has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 268 seconds]
<KotCzarny> longsleep: there is gsoc2014 thread about improving allwinner audio soc support
<KotCzarny> and something about reworking dma code
<KotCzarny> maybe it's related somehow
<longsleep> yes, though it seems to work in android with the same kernel
Netlynx has quit [Quit: Leaving]
bonbons has quit [Quit: Leaving]
avph has quit [Ping timeout: 246 seconds]
<AA781> first thing I m gonna do is clean up this bsp ordeal once I got it downloaded
<AA781> and issue a leaner torrent for distrobution
<AA781> does that sound like a smart thing to do longsleep?
<KotCzarny> setup indexed www tree maybe?
avph has joined #linux-sunxi
<AA781> I can that too
<AA781> can do*
avph has quit [Ping timeout: 246 seconds]
<AA781> btw longsleep how well has mate de been working on your board?
<AA781> big fan of mate myself
avph has joined #linux-sunxi
Inode has joined #linux-sunxi
<AA781> I can also have git installed
<AA781> community edition
<AA781> of course
<AA781> I am run a cluster consisted of 2x j1900s with 4gb ram each with 4x j1800 frontends with 2gb ram each
avph has quit [Ping timeout: 246 seconds]
<AA781> currently with a single 250/20 connection
<AA781> lamp stack
<AA781> my single point of failure is currently the internet connection
<AA781> but will be rectified soon enough
<AA781> 25% completed
<AA781> on torrent downloads
<AA781> I cna provide a stand alone system
<AA781> a core 2 duo with 4gb ram
<AA781> if oyu prefer longsleep
nove has quit [Quit: nove]
<apritzel> why is everyone so keen on this shitty BSP?
<AA781> learning the process
<AA781> new to arm
<AA781> ;)
<apritzel> totally the wrong way to learn
<apritzel> seriously
<AA781> what do you suggest?
<apritzel> if you want to learn how to build a house you don't watch children in the sandbox ...
<apritzel> follow the upstream Linux kernel process
<AA781> I am aware of it mate
<apritzel> follow how to rebuild distributions
<AA781> look I am looking into things pine specific at this moment
<AA781> want to be able to help out
<apritzel> but the Allwinner code is total crap, seriously
<AA781> completely aware of it apritzel
<apritzel> by looking into that not only you learn the wrong things, but you also waste time
<apritzel> I'd really appreciate if people would look into a proper U-Boot port, for instance
<AA781> but tto understand the differneces one must have a comparison to go by
<apritzel> no
<AA781> hence why I am looking at the crappy bsp
<apritzel> you should check out the upstream kernel repo with the A64 patches so far
<AA781> will do mate
<apritzel> for instance USB support is still lacking, but shouldn't be too complicated
<AA781> well I am just familiarizing myself with the state of things
<AA781> as I have no boards to test on at this point
<AA781> not until April or May if there is a delay
<apritzel> I understand that, but seriously that BSP is the wrong starting point
<apritzel> you will just learn how to _not_ do things
<apritzel> without knowing it, that is ;-)
<AA781> which to me i important
<AA781> learning how not to make the same mistakes as they have
<apritzel> I don't think that is how it works
<AA781> but comparing to the proper method kind of looking at both and spotting the differnces I understand
<AA781> visually
<apritzel> if you don't know about ARM, than you have to learn a lot of stuff anyway
<AA781> reading up on u-boot
<apritzel> and learning the old and misguided approach is not of no help
<AA781> not learning it
<AA781> learnign tocompare an dnotice the bad habits
<AA781> and what to avoid ;)
<AA781> is all I am doing mate
<AA781> btw
<apritzel> I still consider this a waste of time
<apritzel> it would make some sense if you'd do it the other way round
<AA781> aside form the bsp provided (with patchwork) what else is working on those boards (with usb/ethernet etc)
<AA781> just a question as I am coming from x86 amd64 to all of this
<AA781> and seeing as there is so little reference areas on where to start
<AA781> and sources to look at
<apritzel> there are patches for the upstream kernel
<apritzel> the core support should go into the current merge window
<apritzel> which will give you the UART, basically
<apritzel> with some more patches you get MMC, I2C, GPIO
keh has quit [Ping timeout: 244 seconds]
<apritzel> an Ethernet driver is in the testing phase
<apritzel> USB support seems to lack some tiny things like proper reset
<AA781> I saw the work you did in terms o 4.x kernel
<AA781> 4X
<AA781> I think it was you
<apritzel> yes
<AA781> so you need usb/eth/mmc/12c/gpio/
<AA781> u-boot needs work
<apritzel> U-Boot isn't so bad so far
<apritzel> it's missing network support (new driver needed, same problem as Linux)
<apritzel> but I can load 32-bit and 64-bit kernels from the SD card
<apritzel> AA781: the point is: if you are really new to ARM, you will need to learn an awful lot of things
<AA781> isn't it booting in a 32bit mode
<AA781> from what I read
<apritzel> yes
<AA781> I'm a quick study
<apritzel> but x86 is booting in 16-bit mode ;-)
<AA781> I know
<AA781> ;0
<AA781> ;)
<AA781> brb
<AA781> wife is calling
<apritzel> problem is: you look at how things are done in the BSP and get the wrong idea
<apritzel> and believe me: coming from x86 to ARM takes a while (I did this three years ago)
hulu1522 has quit [Ping timeout: 244 seconds]
<apritzel> regardless of how much you thought you know about (x86-) computers and/or Linux ;-)
<apritzel> actually I'd prefer you'd look at it from the x86 point of view and totally ignore the BSP
<apritzel> the only use of the BSP is to find information about undocumented things
<AA781> ok
<AA781> back
<AA781> go on apritzel
<AA781> value your input on the subject
<AA781> as I have a month or two to get caught up
<apritzel> the 32-bit boot is not a real problem
<apritzel> you can switch to 64-bit at any time
<AA781> got it
<apritzel> and the chip is more or less 32-bit anyways (no memory or I/O beyond 4GB)
<AA781> 32 bit limitation is no concern understood
<apritzel> so in two months you can surely learn a lot, but not both the bad and the good way
iamfrankenstein has quit [Quit: iamfrankenstein]
<apritzel> you can look at the wrong later for curiosity or amusement
<AA781> ok
<AA781> I understand that there were no no's regarding how they entered el3 mode
<AA781> with the bsp
<apritzel> while you mention it: the BSP contains quite some dead code
<apritzel> for instance this "enter the kernel in EL3"
<AA781> longsleep mentioned that
<apritzel> this was apparently a misguided early experiment
<AA781> henc ethe 24gb torrent
<apritzel> but they forgot to remove this from the code
<apritzel> well, the bad code is not responsible for the 24 GB ;-)
<AA781> nope that's the android inclusion and various other things
<apritzel> indeed, all the build files, for instance
<apritzel> longsleep has the diff in his repo, you can look at this
<AA781> forgot to ask wher ehis repo is
<apritzel> github.com/longsleep ;-)
<AA781> at it
dev1990 has quit [Quit: Konversation terminated!]
<apritzel> but again this is the BSP diff, so gives you the wrong idea
<AA781> crap no sources for boot0
keh has joined #linux-sunxi
<apritzel> that is the boot0 source for some older SoCs
ricardocrudo has quit [Ping timeout: 248 seconds]
<apritzel> we are still waiting for the A64 version
<apritzel> (keep your fingers crossed)
<AA781> I wait with beited breath
<AA781> it is a requirement for further development n'est pas?
<apritzel> for proper U-Boot support: yes
<AA781> so we are dead in the watr on u-boot front till comes along
<apritzel> I have a work-around in place which loads upstream U-Boot with the help of Dom0
<apritzel> s/Dom0/boot0/
<AA781> ok
<apritzel> (shouldn't look at Xen source simultaneously ;-)
<AA781> hehe
<apritzel> plan B is to reverse engineer boot0 to get the required information
<AA781> quite an endeavour
<AA781> from the sounds of it
<AA781> the bin file provided by allwinner
<AA781> excuse me if I sound liek a total newb
<AA781> anyway extract/decompile it
<apritzel> apparently we have permission to redistribute boot0
<apritzel> boot0 is from 8KB till 40KB on any bootable SD card
<AA781> read the liner notes about that
jernej has quit [Ping timeout: 252 seconds]
<apritzel> disassembling is not the real issue, but doing it in a legal way is more involved
<AA781> but doing so (redistributing) is just proliferating the bad
<AA781> rock and a hard place
<apritzel> well, so far you have two ways to initialize the DRAM:
<apritzel> either using boot0
<apritzel> or linking the closed-source binary blob called "libdram" into U-Boot
<AA781> or fell
<AA781> sorry wrong window
<apritzel> the other alternative is to just get along with the 160 or so KB of SRAM ;-)
jernej has joined #linux-sunxi
<AA781> very interesting
<AA781> scarfed from lichee
<apritzel> that's why we are keen on getting our hands on information about the DRAM initialisation
<AA781> surprised no 0one at pine64 is working on it
<AA781> btu then again not surprised as I have noticed very input from them regarding linux development
<AA781> they apparently talk to allwinner all the time
<apritzel> Allwinner promised them Linux support
<AA781> and I understand questions were given to ask allwinner
<AA781> wish I could see the agreement between the 2
<AA781> to really see the details as to what they considered linux support
<apritzel> which for Allwinner means that you can boot some Linux userland with their shitty Android kernel
<AA781> yay
<apritzel> as longsleep showed that is indeed possible
<apritzel> but not what I understand as Linux support
<AA781> the sam egoes for getting mali to work too
<apritzel> I expect to download a random distribution and boot it
<apritzel> yeah, Mali is still a huge pile of work
<AA781> lima hasn't done anything in awhile from what I read
<AA781> (see I have been reading ;)
<AA781> that'sa question I would like to ask
<AA781> if the binary is only provided
<AA781> can that binary be reused for future kernels or do we rely on them to update for said kernel version
<apritzel> apparently Free Electrons have found a way to use the Mali driver with upstream kernels
<AA781> wow
<AA781> nice last paragraph
<AA781> are they saying they've done it
<AA781> or plan on
<apritzel> I think they did it
<AA781> sound sliek done
<AA781> way to go CHIP
<AA781> aren't they running the a10?
<apritzel> something derived from the A13, I think
<AA781> if truly revolutionary
<AA781> use the single binary
<apritzel> but the point is: this work should translate easily into the A64
<AA781> and it's good for future proofing
<AA781> and cedarx has porven to be funcitonal from what I read
AA781 is now known as lennyraposo
<lennyraposo> that's better
<lennyraposo> AA78 is my alias elsewhere
<lennyraposo> I see why you have your site set on the usb/eth/
<lennyraposo> i2c
<lennyraposo> issue
<apritzel> we should first learn how to walk before we try running ;-)
<lennyraposo> if we can ween the crappy dependency on the silly android blobs this project may go somewhere
<lennyraposo> mate you are very informative
<apritzel> frankly I personally don't care so much about graphics, but other people do
<lennyraposo> neither do I
<apritzel> I am happy with serial and Ethernet ;-)
<lennyraposo> just using terminal
<apritzel> the mouse is there to change focus between the xterms ;-)
<lennyraposo> lol
<lennyraposo> realistically for me
<lennyraposo> ssh in
<lennyraposo> done
<lennyraposo> a gui is not needed for what I plan to do
<lennyraposo> brb
<apritzel> but we need either Ethernet support or USB (to use USB-Ethernet dongles)
<lennyraposo> true
<lennyraposo> the ethernet used is a realtek based am I not correct?
vishnup has joined #linux-sunxi
<apritzel> no
<lennyraposo> oh
<apritzel> it's a new Allwinner-made MAC
<lennyraposo> that's just the wifi
<apritzel> right
<lennyraposo> i have about 20 tabs opened
<lennyraposo> on things related to all this
bwarff has joined #linux-sunxi
<apritzel> I tried to find something similar in drivers/, but failed to do so
<apritzel> so we assume it's a genuine new IP block Allwinner came up with
<apritzel> (for whatever reason)
<apritzel> (I'd pick some RTL for which drivers are already upstream and save me a bunch of work)
<lennyraposo> didn't the h3's allready sport gbit
<apritzel> well, the A20 already has Gbit
<apritzel> which is some Designware IP