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
AreaScout has quit []
<Turl> itdaniher: we got the SDKs, but there's not many people interested in A31 here due to the GPU
bfree__ has quit [Quit: Konversation terminated!]
<brain_> great, now nand-part does not function! one issue after another
<brain_> I have no /dev/nand..... only nanda - nandj
<Turl> brain_: must be a really ancient kernel
<Turl> (or stock allwinner)
<brain_> its 3.0.36...i cant get 3.4 to bootup. the newest sources i have that i can boot are 3.0.76
Nazcafan has quit [Ping timeout: 252 seconds]
<Turl> brain_: 3.0.x from allwinner or linux-sunxi?
<Turl> brain_: if it is from AW, you won't have /dev/nand. It's our addition
<brain_> i dont remember were i got them, had the touchscreen drama so long i forgot. but i notice that there is a modules folder with libnand...on my 3.0.57 sources the modules folder does not have this..ill try that kernel now.
<brain_> 57 is from linux-sunxi
pwhalen has quit [Read error: Operation timed out]
orly_owl has quit [Read error: Connection reset by peer]
orly_owl has joined #linux-sunxi
pwhalen has joined #linux-sunxi
pwhalen has joined #linux-sunxi
pwhalen has quit [Changing host]
orly_owl has quit [Read error: Connection reset by peer]
orly_owl has joined #linux-sunxi
x57522 has quit [Ping timeout: 252 seconds]
nieuwbie has quit [Quit: ERC Version 5.3 (IRC client for Emacs)]
popolon has quit [Quit: Quitte]
bamvor_ has quit [Quit: Leaving]
bamvor has joined #linux-sunxi
egbert has quit [Disconnected by services]
egbert_ has joined #linux-sunxi
jinzo has quit [Quit: Leaving]
pwhalen has quit [Ping timeout: 245 seconds]
pwhalen has joined #linux-sunxi
pwhalen has joined #linux-sunxi
pwhalen has quit [Changing host]
pwhalen has quit [Ping timeout: 245 seconds]
pwhalen has joined #linux-sunxi
pwhalen has quit [Changing host]
pwhalen has joined #linux-sunxi
orly_owl has quit [Read error: Connection reset by peer]
orly_owl has joined #linux-sunxi
orly_owl has quit [Ping timeout: 252 seconds]
orly_owl has joined #linux-sunxi
focus has quit [Ping timeout: 252 seconds]
focus has joined #linux-sunxi
<brain_> my nand is not repartitioned..no turning back now! had to rebuild my drivers for the 3.0.57 kernel but at least it fixed the nand issue i was having. maybe this is why i couldnt boot properly?
<brain_> now*
<brain_> i am just wondering if i screwed up somewere will my oem livesuit image repartition things?
<brain_> i did make backups as stated on the wiki just in case tho
pwhalen has quit [Remote host closed the connection]
VargaD has quit [Ping timeout: 252 seconds]
VargaD has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
orly_owl has quit [Ping timeout: 245 seconds]
orly_owl has joined #linux-sunxi
dlan has quit [Quit: Lost terminal]
<Turl> brain_: livesuit has an option that formats the full nand if you screw things badly
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
dlan has joined #linux-sunxi
<libv> hramrach: i think this page, and the u-boot and script.bin should perhaps be called inet_86vs
<libv> brain_: what hardware is this that isn't booting with 3.4?
<gzamboni> damn https://github.com/ryangithub/sunxi-can-driver is not up anymore, we should have copied that code
<gzamboni> well, we have his email, and he posted to the ML
patapovich has joined #linux-sunxi
<patapovich> Check the last message
diego_r has joined #linux-sunxi
<gzamboni> thanks, just saved here
tomboy64 has quit [Ping timeout: 245 seconds]
ykchavan has joined #linux-sunxi
ykchavan has quit [Remote host closed the connection]
tomboy64 has joined #linux-sunxi
tomboy64 has quit [Ping timeout: 272 seconds]
mrnuke has quit [Ping timeout: 245 seconds]
mrnuke has joined #linux-sunxi
<patapovich> gzamboni: yw
Quarx has joined #linux-sunxi
Black_Horseman has quit [Ping timeout: 240 seconds]
<diego_r> Improv news on KDE website: http://dot.kde.org/2014/01/22/open-hardware-kde
FreezingCold has joined #linux-sunxi
patapovich has quit [Quit: Page closed]
foubarre has joined #linux-sunxi
tomboy64 has joined #linux-sunxi
rellla has joined #linux-sunxi
rellla has quit [Ping timeout: 240 seconds]
rellla has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has quit [*.net *.split]
mrnuke has quit [*.net *.split]
orly_owl has quit [*.net *.split]
bamvor has quit [*.net *.split]
bamvor has joined #linux-sunxi
mrnuke has joined #linux-sunxi
orly_owl has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has quit [*.net *.split]
mrnuke has quit [*.net *.split]
orly_owl has quit [*.net *.split]
bamvor has quit [*.net *.split]
mrnuke has joined #linux-sunxi
bamvor has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has quit [Changing host]
HeHoPMaJIeH has joined #linux-sunxi
orly_owl has joined #linux-sunxi
boycottg00gle has joined #linux-sunxi
notmart has joined #linux-sunxi
<oliv3r> libv: fbdev does the same; so atelast fbturbo is not to blame here
<libv> really? ouch
<libv> but it does fit in with the notrapsignals and the lack of debugging info
<oliv3r> i'll poke jemk abou t it
<oliv3r> if he joins here today
<libv> someone posted a new 10.1 a20 tablet on ebay for 40E, because he buggered up rooting the device
<libv> here's me hoping it is the proper lvds device, but i won't know until it gets here
<HdkR> haha
<oliv3r> that's cheap
<libv> hey, another mans idiocy is linux-sunxi's advantage
<HdkR> Messed up rooting it, forever unusable now
<libv> in this specific case
<libv> for once.
<oliv3r> must be a real tard
<oliv3r> if he messed it up that badly
<libv> rooting went wrong and he didn't know how to google
<libv> isn't this stuff prerouted anyway?
<libv> wtf am i going to do with a stupid 10" tablet once the deivce has been added and the sunxi_kms driver works on it...
<libv> i already have a much more luggable 7"er with the same size ram, same resolution and a20 as well
<libv> if only my gf was normal and played angry birds all day
<HdkR> Is she too busy supporting you? :P
<libv> :)
<libv> either supporting or pestering :)
<HdkR> I like to think that pestering is a form of support
<oliv3r> libv: what disp. does it have? tn or tft?
<libv> oliv3r: it might actually be the full lvds one
<oliv3r> if it's TN, it's useless; if it's non-tn and has 1gb of ram; i'll buy it off you
<oliv3r> with TN i mean Twisted ... Newtons lol id unno
<oliv3r> my tablets both have TN displays, they have viewing angles of 5 degrees
<oliv3r> it's horrible
<oliv3r> useable; but not 'nice'
<oliv3r> libv: how doy ou recongize lvds on the pcb?
<libv> oliv3r: it's probably this one: http://linux-sunxi.org/K1001L1C
<libv> the owner reported that disp didn't come up on the lcd until he put pll3 to 297 in the script.bin
<oliv3r> ok
<oliv3r> i guess there's now ay to find out what screen is actually used; until you get it
<libv> oliv3r: do you really use those tablets for anything else but writing code and testing stuff?
<oliv3r> libv: yeah; in bed :p
<libv> oliv3r: TMI!
<oliv3r> and on the shitter
<libv> ARGH!
<oliv3r> every morning!
<oliv3r> but i notice that 512mb ram on mine is real tight
<libv> tight???
* libv is horrified
<oliv3r> memory is tight
<oliv3r> my poop-flow is great
<oliv3r> so yeah; if it has a decent screen; i'll buy it off you :p
<oliv3r> lets hope it arrives fast
<libv> i am listening to white zombie atm, and rob zombie loves his old horror films
<libv> but that's nothing compared to what is coming out of oliv3r atm :p
<oliv3r> is that LVDS?
<oliv3r> closeu p of the LCD connector
<libv> oliv3r: probably not
<oliv3r> oh
<oliv3r> so it's not the 'extra wide' connector
<oliv3r> can i recognize it from the lcd0_para?
<libv> hdmi/dvi are wholly based on lvds, but unlike hdmi/dvi you do not get a pair of wires per colour channel. you get 6 bits of data per pair (4-5 pairs, of which one is the clock - 4 pairs 18bit, 5 pairs 24 bit) and this up to a given bandwidth, after which you need a second channel
<libv> oliv3r: you should be able to recognize it from there, lcd_if should be 3
rellla has quit [Ping timeout: 252 seconds]
<hramrach> hello
<hramrach> anyone has some idea how to make flat connection to the UART pads?
<hramrach> the pasd are on the side of the PCB that rests against the display so the thinner the better I guess
<hramrach> still I need to find where they actually connect
merbanan has joined #linux-sunxi
<hramrach> btw how do I tell what kind of display is on the tablet?
<hramrach> I tried stfw for the random stuff printed on the cable and only thing that came up is that it's also used on a rk tablet
rZr has quit [Read error: Connection reset by peer]
rZr has joined #linux-sunxi
<hramrach> and yes, the display does not come up on this one either
<libv> hramrach: lcd_if is?
<hramrach> it says 0 in the fex but might just be bogus
<libv> hramrach: try the pll3 trick anyway, you never know
<libv> or was it pll7
<juanfont> is there any decent 3g tablet with AW?
<libv> juanfont: is there any decent tablet with allwinner?
rZr has quit [Excess Flood]
<juanfont> :P
<hramrach> hmmm, I will try to search for the pll thingy
<libv> hramrach: LVDS LCD no clock
<libv> is the name of the email thread
<hramrach> thanks
<libv> "[PATCH] Fix K1001L1C LCD bug"
rZr has joined #linux-sunxi
<libv> but you probably need to set pll3_freq = 297 under [target]
<libv> imho this really should have no effect, but then, i will find out the real cause if this tablet that i just bought is the one i think it is
<oliv3r> libv: lcd_if = 0 :(
<oliv3r> hramrach: you can solder very thin pads; no need for big gulps
<oliv3r> libv: that patch got merged allready I think
<libv> oliv3r: yeah, but it doesn't make sense to have this fix there
<oliv3r> the driver should do the right thing? (tm)
<libv> yup, it should, but it apparently doesn't
<libv> why, i will hopefully be able to find out
<oliv3r> i guess you are fixing this in the KMS driver though? or does the kms driver still need the disp driver to talk to the hardware?
<libv> the fact that this is needed doesn't make sense to me atm
<libv> but my kms driver uses both pll7 (for the engine clocks) and pll3 (for the display clock)
<libv> so wait and see what the issue is there
rz2k has joined #linux-sunxi
<oliv3r> kms is a replacement for disp + fb?
<oliv3r> or just disp?
<oliv3r> or disp + fb + g2d?
<oliv3r> hopefully Post FOSDEM i'll understand the whole allwinner display architecture better :)
<libv> disp & fb
<oliv3r> and lima obviously is mali/gpu/3d, so what driver would g2d go under? xorg-server-video-sunxig2d?
<libv> no drm driver does 2d
<libv> yet the idiots happily tossed together display and 3d
geecko has joined #linux-sunxi
<libv> so perhaps g2d will remain as is, it will need to get dma_buf and the afterthought (wtf!) that is dma_fence
<oliv3r> libv: i'm looking at the K1001L1C pcb; and i noticed that the LCD connector has an extra chip there, a 'lvds' controller of some sort? what's the function of that chip?
bsdfox has quit [Ping timeout: 248 seconds]
<oliv3r> hmm, that 50 pin flatcable seems labled 'tp' so that may be a touchpanel controller chip
<oliv3r> the LCD seems to be top right; with quite some bodge wires
foubarre_ has joined #linux-sunxi
bsdfox has joined #linux-sunxi
foubarre has quit [Read error: No route to host]
foubarre_ has quit [Read error: Connection reset by peer]
foubarre has joined #linux-sunxi
<oliv3r> that map looks really cool :)
<oliv3r> heh; look how bright NL is :D
<oliv3r> NL/Germany
<oliv3r> heh, there's so little data available right now, you can clearly see the route i take to work every day :)
<libv> oliv3r: it's like a map of civilization :p
<brain_> can someone help with with my horrid math skills? http://linux-sunxi.org/Installing_to_NAND
<brain_> look were it says EXAMPLE TABLE "Using NAND /dev/nand, with 15958016 sectors (7792.0MB)."
<brain_> I simply cannot figure out were they got that number 15958016
<libv> oliv3r: yeah, that probably takes in the rgb data and encodes it over to lvds
<oliv3r> libv: you sure it's not the touchscreen driver?
<oliv3r> brain_: libv wrote that page I think
<oliv3r> I want to control my UPS over USB, but apcupsd won't start
<oliv3r> now that question is really very frequently asked
<ccaione> brain_: 7792*1024*2
<libv> brain_: first state what tablet you are working on
<brain_> that worked fine, i was multiplying by 512x2...been a while and i suck at math. its a Zeki TB782B.
<brain_> that you
<brain_> thank you
<brain_> ugh just woke up
<libv> brain_: which of the tablets on linux-sunxi is that also known as?
<brain_> I really do not know but the generic sun4i-crane kernel configs bootup without any issue.
<libv> brain_: new_device_howto
<brain_> it really is a cheap tab but I love the way the hardware is setup internally.
<ccaione> lsl, new_device_howto is the libv's personal crusade
<ccaione> :)
<libv> ccaione: yes, for your benefit as well.
bsdfox has quit [Ping timeout: 272 seconds]
<ccaione> yeah, if one day I'll buy a sunxi tablet
<libv> ...
<brain_> you want me to write something up I don't mind. I have plenty of notes. I am upgrading to the 10" A20 version once I get through with this...trying to breath some new life into the thing. i already have a decent tab for daily usage but I am so sick of android
<libv> brain_: i don't say this for no reason.
<libv> brain_: you are depending a lot on wiki pages that others have written and information that others have gathered for you.
<libv> brain_: it is only natural that you do that little courtesy of creating a useful wiki page (where you can also document your own findings with your tablet) and send in a patch for u-boot and sunxi-boards
<libv> brain_: if you don't do any of that, you are outright stating to guys like me that they are wasting their time, that any effort put in has gone to waste on just a single person
<brain_> So do I just need to create an account on the wiki page and get started or does it need to be approved? I really do not mind I would love to have my notes and hardware mods, info and data put someplace to help others.
<libv> it doesn't need to be approved
bsdfox has joined #linux-sunxi
<brain_> Ok great! I will sort my notes and start posting what I have onto the wiki page. I ran my own forum for 4 years so i know what you mean about helping one person in private versus posting in the public forums.
<libv> cool :)
popolon has joined #linux-sunxi
bsdfox has quit [Ping timeout: 248 seconds]
bsdfox has joined #linux-sunxi
boycottg00gle has quit [Remote host closed the connection]
diego_r has quit [Ping timeout: 245 seconds]
diego_r has joined #linux-sunxi
<oliv3r> some a3x SDK info for v4.5; it mentions h265
diego_r has quit [Client Quit]
diego__r has joined #linux-sunxi
diego__r is now known as diego_r
<oliv3r> brain_: we have a really nice 'new_device_howto'
<oliv3r> backreaders: that sdk info: > 3 supports H.265 video playback , 720p @ 24fps @ 1Mbps;
<oliv3r> so a31 can appearantly do it
<oliv3r> 4.4 calls fiber
<oliv3r> is called*
<brain_> I just got registered. as soon as I get everything sorted out so I know the information is correct I will create a new page. I've already read through the new_device_howto page. I want everything to be correct. like I said I spent 6 months in limbo tackling the touchscreen driver alone and kind of got a bit unorganized since. I want to get the 3.4 kernel to boot for the wiki page. do not worry once I take my medicine the Wiki page wi
<brain_> ll get going.
<brain_> i NEVER wrote a wiki page so its kind of new to me, most of my information is sittiing in Tomboy notes :/ gonna be a new experience for me in that regard since other will be reading it at some point.
paulk-collins has joined #linux-sunxi
<libv> brain_: it's almost plain text with just a few simple formatting characters
<libv> brain_: the example page should give you enough hints to get this done succesfully
<oliv3r> brain_: there's a nice template you can copy/paste :)
<oliv3r> i'm soo tired today :(
<brain_> I will definitely need a template of some sort to starte with, does the system allow file hosting or do i need to provide that? i've got a massive collection working under the microscope to trace out things, debug. methods used for find the uart, how I mapped out the IRQ's used when i was fighting the driver issue. i just need some time its still very early here.
<oliv3r> brain_: nah, the wiki is a big database kind of
<oliv3r> brain_: as for pictures, you can upload them and link them into your document
<binaryferret> out of interest brain_ what was the touchscreen that you wrote the touchscreen driver for?
<brain_> ft5306, yes there is one out there but it did not function on this tablet, I had to severely modify it before it would work. I trashed the eeprom and ended up having to use a bin file from another tablet with the wrong demensions which i fixed once the driver worked.
<binaryferret> Nice :D
<binaryferret> Well, sad that you had issues, but nice that you solved it.
<libv> hramrach: "also known as" should list manta ...
<libv> http://linux-sunxi.org/New_Device_howto now with extra libv ranting goodness at the start!
FreezingCold has quit [Ping timeout: 248 seconds]
JohnDoe_71Rus has quit [Read error: Connection reset by peer]
<gzamboni> we can see better the optimus in the video from the url i posted
brain_zeki has joined #linux-sunxi
<brain_> gonna reboot to livesuit and fix my partition mistake. back in a few.
brain_ has quit [Quit: Leaving]
<gzamboni> humm, mele will have a 4K device with A31
<gzamboni> he said he was runing it on the linux envirement, with the blobs i supose
<oliv3r> brain_zeki: i thought you had a sd22something
<oliv3r> solomon tech
<brain_zeki> Not sure were you got that, chip is ft5306eeb i believe. Driver is totally hacked together by the oem.
<brain_zeki> It completely ignores the fex file..all hard coded
<brain_zeki> Good livesuit fixed the partitions.
<libv> oliv3r: ah, you must be right, that must be the touchscreen chip in that picture
<libv> ft53 is probably what one can blurrily read off the board
<libv> so we probably do support lvds directly
brain__ has joined #linux-sunxi
brain_zeki has quit [Quit: AndroIRC - Android IRC Client ( http://www.androirc.com )]
<foubarre> gzamboni: interesting news about the A31. Do you think 4K will only be attainable to display videos or also for games/GUIs?
<oliv3r> duck! puneetb incoming!
<gzamboni> foubarre, dont know, i supose they can/should do gaming also
<gzamboni> 4k tvs are too expensive for the moment
<HdkR> hm?
<HdkR> There's a 4K 37" television for $500
<libv> HdkR: at 60HZ?
<libv> Hz even
<HdkR> no, you know it isn't
<HdkR> Nothing cheap will be driving 60hz
<libv> last time anyone posted about it (was that you?), it was 30Hz only
<HdkR> yea, was me
<libv> :)
<libv> foubarre: no way is our memory bus capable of decently handling 4k video :)
<HdkR> Just because it is only 30hz doesn't mean it is bad, just don't play video games on it
<oliv3r> crap; looks like broadcom driver won't be so easily ported
<oliv3r> from m ainline backported that is
<oliv3r> guess we just have to port the cubieboard driver
ganbold_ has joined #linux-sunxi
linkmauve1 has joined #linux-sunxi
<HdkR> oh, new information?
<HdkR> not new, it speculates about ChromeOS support though
<HdkR> So...yay for PVR on "linux" again?
<oliv3r> that isn't the optimus though; and some yank; i don't belive he works at aw
<gzamboni> oliv3r, i saw that video, i think you will be one of the "first customers" to receive it. Did they tell you went they will send your board ?
<HdkR> oh, I didn't watch the video
<oliv3r> maybe it is yeah
<oliv3r> gzamboni: dunno; but not happy about powervr; so dunno what i'll do with it :)
<oliv3r> maybe i'll deny it :) and say 'i don't want powervr'
<gzamboni> lol
<oliv3r> shows an a23 dual core, dual sim, dual standby tablet
<foubarre> gzamboni: i suppose too, but there are always surprises...
<foubarre> libv: .. and 640 KB ought to be enough for everyone. ;)
deasy has joined #linux-sunxi
tomboy64 has quit [Quit: And remember, aal is well.]
<brain__> lol
geecko has quit [Quit: bye bye]
nove has joined #linux-sunxi
pwhalen has joined #linux-sunxi
AreaScout has joined #linux-sunxi
techn_ has joined #linux-sunxi
npcomp has quit [Ping timeout: 264 seconds]
<brain__> I had to read a wiki....to learn how to use the wiki...
npcomp has joined #linux-sunxi
<oliv3r> brain__: you dawg, you read the wiki so you can wiki the wiki!
<brain__> thats the phrase i was searching for!
<brain__> once I upload a few pictures I need to somehow learn to link to my page in the tablet category, for some reason I couldnt figure it out.
<libv> brain__: in the example page, it is almost done for you
<libv> brain__: CATEGORY, at the bottom of the page
<nove> oliv3r: for sure that is by software, only 720p 1Mbps
<brain__> right, but how does it appear like lets say were tablets > Z > Zatab... how would I list it there?
<oliv3r> nove: have you read the backlog?
<oliv3r> nove: you have :)
<oliv3r> nove: h265 though; i would exect these cpu's to be way to weak for any h265
<oliv3r> nove: anyway, i should debug if its' mplayer vs mpv; but mplayer2 seems to not work too great with cedar atm
<brain__> ahh there we are. Well its a small start but im getting the hang on it. gonna add some more info and see what pictures I had locally before I really get involved.
<libv> brain__: great
MasterChief10 has joined #linux-sunxi
MasterChief10 has quit [Client Quit]
<nove> oliv3r: i cant help you there, maybe you should add some printf with timestamps to vdpau, to see were is the slow part
<oliv3r> probably :)
wingrime has joined #linux-sunxi
<oliv3r> hey wingrime
<wingrime> oliv3r: yay
orly_owl has quit [Ping timeout: 240 seconds]
orly_owl has joined #linux-sunxi
orly_owl has quit [Client Quit]
<nove> is this the future for the kernel side of VE? http://linuxtv.org/downloads/v4l-dvb-apis/codec.html
<nove> it even has VPX
<nove> and for anything missing we can add *_CEDRUS_* controls
<nove> wingrime: jemk ^
jemk has joined #linux-sunxi
<libv> brain__: aha, yes, those two pads really looked suspiciously like a uart
shineworld has joined #linux-sunxi
shineworld has left #linux-sunxi [#linux-sunxi]
<oliv3r> nove: sounds new; when was that published?
<oliv3r> nove: ah, it usees M2M :)
<brain__> yea they were easy to find. I have a LOT of pictures to add to show how a clean installation can be done but right now I am gonna try to get this nand-boot working. i'll come back to it in between kernel builds.
<nove> oliv3r: i just when to the site i clicked the first link to looked related
<libv> brain__: could you do a picture of the complete back of the device? this is a good way to identify tablets on sight
<jemk> nove: yes, I think this is the way to go, but it isn't available in 3.4
<brain__> its a start at least
<libv> brain__: yeah
<brain__> Yes I will take more pictures
<oliv3r> jemk: oh your here too; mplayer starts to stall when resizing/looping longer with mplayer2/debian
<nove> jemk: then time for mainline, we don't need disp to start
<libv> brain__: the block size calculation is quite iffy with the current code
<oliv3r> well i'd port it to 3.10 first; disp + most mainline bits
<libv> nand-part isn't the most refined of tools
<jemk> oliv3r: also when not resizing?
<libv> brain__: but here's a top tip: give a partition a size that's known too large...
<libv> brain__: and the kernel will then, upon resync, spew out the real size of the partition
<libv> you can look for it in dmesg
rellla has joined #linux-sunxi
<oliv3r> jemk: when not resizing, after about 20 minutes I guess; i use -loop 0
<jemk> nove: yes, i tried to do my h264enc demo on mainline, but there are clocks and such things missing
<brain__> Ok i'll try now and see what happens...what I dont understand is how does the u-boot know to load uimage if nandb is gone which held the bootargs? since nandb will now be the rootfs?
<oliv3r> brain__: you probably can best just start the kernel from nanda/boot.axf
<brain__> last time I tried I got invalid image magic error when u-boot searched
<oliv3r> brain__: but u-boot has these settings hardcoded by default afaik
<oliv3r> brain__: nuking nandb is possible; but shouldn't be the first step :)
<brain__> how would I start the kernel using boot.axf?
<oliv3r> brain__: via the .ini :p
<brain__> I attempted to change the line in the ini file....right now it lists uboot.bin...
<brain__> it then jumped to fel mode
<brain__> complained about sizes
techn_ has quit [Ping timeout: 240 seconds]
<nove> oliv3r: yes, and what is missing we can ask to add
<brain__> so realistically uboot is not needed to boot a kernel directly? maybe if I edit the ini first but update the side segment, i did not try that
<libv> brain__: the nand howto actually isn't complete
<brain__> size*
<libv> brain__: it requires patches of mine that are sitting on the ml
<libv> "u-boot/lichee-dev nand boot patches"
<brain__> yea i realized that, i am able to kind of get the general idea but i noticed the ini files were a little different on mine.
<brain__> ok for example..
<libv> this then allows you to put the necessary files in nanda/linux/ and be done
<brain__> i have no recovery.ini, could I add one? would it attempt to boot the kernel using that ini or do i need more changes?
<libv> yes, it could be more dynamic, but as said in the first email, lichee-dev is not a long term solution
<libv> brain__: check out this email
<brain__> ok
<nove> jemk: then, we could start with a wiki page, a VE kernel TODO to do the planing, oliv3r can you help us (i am not very aware of kernel things)
<jemk> oliv3r: arch doesn't have mplayer2 in the repo and i don't want to compile it, you are on your own ;) maybe you can look at "perf top" to see where it is slow
<jemk> or drop mplayer2, its end of life
<atsampson> mpv seems to work remarkably well -- it's replaced mplayer/mplayer2 for me
<wingrime> nove: don't be so quick
<wingrime> maz: we need display support for mainline
<wingrime> err
<brain__> I see whats going on in that patch, I could apply this manually to the u-boot sources to test it...think that would be a good idea?
<wingrime> nove: we need DISP firstly
<brain__> Also, it seems possible I could do a testrun using the commands directly no?
<oliv3r> wingrime: with mem2mem; you actually decode to ram; so you can decode to png if you have to
<oliv3r> brain__: use the recovery.ini to get an idea how a linux.img would look like
<oliv3r> jemk: debian doesn't have mplayer 1 or mpv; only mplayer2; so i guess i'll try the fedora binary from the other partition tomorrow :)
<maz> wingrime: display support? I'm afraid you're asking the wrong guy. I'm very happy with a serial console! ;-)
<libv> wingrime: yes, yes
<nove> wingrime: but i think is at lest time to start planing
<libv> first i need to get the buffer swapping of fbturbo fixed, so i can demo something useful at FOSDEM
penghb1 has quit [Ping timeout: 246 seconds]
<nove> wingrime: then not starting, only planing for now, do we create a wiki page for this?
<wingrime> nove: wa always can use libdrm scheme and use existing code
orly_owl has joined #linux-sunxi
<nove> wingrime: can libdrm protect the use of the buffer registers? so that the userland doesn't be allowed to write to any memory address
<nove> wingrime: i think v4l is better in this case, it even says, "Memory-to-memory devices can often be used as a shared resource: you can open the video node multiple times ..."
<libv> wingrime/nove: please use dma_buf/dma_fence
<libv> i will base the upstream kms driver on that, as it isn't available in 3.4 yet
<libv> this will also be true for any future mali kernel work
<jemk> libv: you mean the buffer sharing mechanism? I think this is available for v4l: http://lwn.net/Articles/486300/
<libv> yes, it should be available, but only in more recent kernels than 3.4
<libv> brain__: you should be able to apply something similar to the sun4i file, yes
<libv> brain__: this part is pretty chip independent, it just tells u-boot where to look for script.bin and the uImage
<nove> libv: my none existent knowledge of the kernel doesn't allows me to know if that is good or not, i just want to make things going
<libv> i thought you guys were talking about memory, and giving/restricting access to it
merbanan has quit [Ping timeout: 252 seconds]
<nove> libv: i was referring to the security problem, of having the cedar hardware that when activated will write to the address that is in a output buffer register
<nove> so stoping userland to put there bad values for address
<libv> nove: the kernel driver should check for that
printallthething has joined #linux-sunxi
derethor_ has joined #linux-sunxi
derethor_ has left #linux-sunxi [#linux-sunxi]
HeHoPMaJIeH has quit [Quit: Leaving]
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
rellla has quit [Quit: Nettalk6 - www.ntalk.de]
rz2k has quit [Remote host closed the connection]
foubarre has quit []
<brain__> libv, PROGRESS!
<brain__> i did not do anything to the partition layout yet, I simply built u-boot with that patch, updated it on my tablet, put my uImage into nanda/linux and bam it launched the kernel.
<brain__> it fails to boot however due to no root= option, cant mount a filesystem...i think im on the right track here tho thanks. any pointers?
<brain__> good news is that patch worked on the sun4i tree and worked were the last attempt failed completely.
<brain__> were should I set the root=option? should I compile it into the kernel itself or use u-boot args?
Quarx has quit []
netlynx has joined #linux-sunxi
netlynx has joined #linux-sunxi
Faisal has quit [Read error: Connection reset by peer]
pwhalen has quit [Ping timeout: 252 seconds]
pwhalen has joined #linux-sunxi
pwhalen has quit [Changing host]
pwhalen has joined #linux-sunxi
<Turl> brain__: u boot args would be prefered
<jemk> Turl: hi, are you sure pll4 is compatible with pll1 on a20? To me it looked like it wouldn't have p and m when playing around, only n and k and the manual says this too.
<Turl> jemk: let me recheck docs & code
<Turl> jemk: on A13 it is compatible
<Turl> jemk: on A20 it doesn't seem so hm
<Turl> jemk: good catch
<Turl> jemk: looks compatible with pll5 before the divisors though
<jemk> Turl: they added pll8 and it seems like pll4 is changed and is now same as pll8
<Turl> jemk: can you try something like http://sprunge.us/SPgK + changing the compatible?
<jemk> Turl: that seems to work
techn_ has joined #linux-sunxi
VargaD has quit [Ping timeout: 259 seconds]
<Turl> jemk: I'll make a proper patch and submit it to mike then
<Turl> jemk: what are you working on with ve on mainline btw? :)
VargaD has joined #linux-sunxi
Faisal has joined #linux-sunxi
rz2k has joined #linux-sunxi
<jemk> Turl: I tried to do the h264 encoding poc on mainline, but gave up pretty fast and did it in userspace again.
<jemk> Turl: now I want to try again and do at least some tests, but there are more clocks missing and i don't really understand how this clock things work.
<Turl> jemk: if you give me a list of the clocks you're missing I'll implement them
<wingrime> Turl: egg and chicken problem
<jemk> Turl: :) dram gates (0x100) and VE clock (0x13c)
<jemk> ve clock ideally with the reset function, but i think this isn't the job of the clock framework
popolon has quit [Quit: Quitte]
<wingrime> jemk: seems, many clocks have reset in register but it not documented
<wingrime> reset bit
AreaScout2 has joined #linux-sunxi
<atsampson> is there anything else I should add to http://linux-sunxi.org/Sanei_N90 ?
AreaScout has quit [Disconnected by services]
AreaScout2 is now known as AreaScout
bvn13 has joined #linux-sunxi
<Turl> jemk: ok, I'll ping you when I have something to test
<Turl> libv: ^^
FreezingCold has joined #linux-sunxi
t3st3r has quit [Ping timeout: 240 seconds]
t3st3r has joined #linux-sunxi
wingrime has quit [Ping timeout: 265 seconds]
<brain__> WOOT! got it!
<brain__> Thank you Turl, libv and everyone else. the missing link for me was the bootloader patch. that got me on the right track. I merged all my partitions so all I have left is nanda for the bootloader files, small nandb for env, and one large ext4 partition which is holding my rootfs. I made zero changes to my rootfs and it boots up a WHOLE lot faster than my sd image.
<brain__> now im gonna see about upgrading my kernel to 3.4 if possible. dont worry i have EVERYTHING documented along the way for the wiki.
<brain__> no more android and plenty of storage for native building. gonna have some fun with this.
<oliv3r> lo
<libv> brain__: use those u-boot args that you put in that patched config file
<libv> ah, ok :)
tomboy64 has joined #linux-sunxi
<brain__> just got it going, working on updating the kernel.
<brain__> thank you libv for your help and patience!
<libv> atsampson: looks good. I would break up the tips/tricks/caveats section with seperate headings, but that's a taste thing
<libv> brain__: well, keep up the good work, we still need some patches for your board :)
<oliv3r> mnemoc: i think we need to back-patch 3.4 stuff again; there's stuff we want in later 3.4
<libv> atsampson: looks real nice though, well done
<oliv3r> brain__: if your not using u-boot, via the ini directly booting a bImage; then yes, you have ot compile it in; if you use u-boot to boot a kernel; you can use the 'env' partition, which holds al u-boot settings, in nandbi think (kernel is nandc, unliky what i told you earlier)
techn_ has quit [Read error: Connection reset by peer]
<oliv3r> wingrime: if we can confirm reset bits on certain registers; we should make use of it with the register controller
jemk has quit [Ping timeout: 240 seconds]
netlynx has quit [Quit: Leaving]
techn_ has joined #linux-sunxi
Faisal has quit [Read error: Connection reset by peer]
Faisal has joined #linux-sunxi
<Faisal> hi there, anyone knows how i can enable hdmi sound by any chance on the cubietruck ?
techn__ has joined #linux-sunxi
<Faisal> hi there, anyone knows how i can enable hdmi sound by any chance on the cubietruck ?
<oliv3r> Faisal: works by default :p
<oliv3r> or should; most likly your script.bin is isconfigured
techn_ has quit [Ping timeout: 260 seconds]
<Faisal> oliv3r: i downloaded the cubian image on the cubieboard website and its not working :(
<Faisal> oliv3r: sript.bin ?
<Faisal> oliv3r: googled it, thx
<oliv3r> Faisal: probably wanna check #cubieboard first
<oliv3r> or cubietech; or cubietruck; dunno what they fly under these days
<Faisal> oliv3r: thx will do, i figure that file is already configured properly since its on their website, I'll ask them in any case, thx a lot
<oliv3r> exactly;t he entire image should work
bvn13 has quit [Quit: Ухожу я от вас]
jemk has joined #linux-sunxi
bvn13 has joined #linux-sunxi
FR^2 has joined #linux-sunxi
rellla has joined #linux-sunxi
pacopad has joined #linux-sunxi
rellla has quit [Remote host closed the connection]
<libv> hramrach: if the uart is disabled, then there is a good chance that it is multiplexed with the sdcard
<libv> hramrach: a multimeter should quickly reveal that
nove has quit [Quit: nove]
notmart has quit [Quit: notmart terminated!]
tomboy64 has quit [Ping timeout: 252 seconds]
Night-Shade has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
Black_Horseman has joined #linux-sunxi
TomiK has joined #linux-sunxi
yessikattal has joined #linux-sunxi
yessikattal has quit [Read error: Connection reset by peer]
yessikattal has joined #linux-sunxi
<hramrach> libv: it is connected to the SD slot :/
<hramrach> quite clear with a SD breakout board
<yessikattal> los hombres escribanme en privado
<hramrach> わかりません〜
yessikattal has quit [Client Quit]
<hramrach> libv: and they have uart3 free which is not connected to anything. Chinese :/
<libv> hramrach: :)
<hramrach> actually thy have yet another pair of testpoints marked rx and tx :o
<oliv3r> hramrach: that's probably uart0
<hramrach> and still next to the sd slot
<hramrach> and also connected to the slot
<hramrach> and covered by the TP cable most of the time, awesome
<oliv3r> why did they add 2 identical test points?!
<oliv3r> you can short them with the multimeter?
<hramrach> I can short them to the SD slot both
<oliv3r> that's just beyond stupid
<hramrach> each pair on one side of the PCB
<oliv3r> can you make a closeup of the A13 chip near pin 150?
<oliv3r> e.g. the 'left' side
<hramrach> it goes to a via if anywhere
<hramrach> almost no connections on that side
<oliv3r> yeah backside if your photoskills allow it too
jemk has quit [Remote host closed the connection]
<hramrach> I would need a real light and something that can focus on close objects
<hramrach> and the whole abckside ahs a sticker to insulate from the metal LCD casing
<oliv3r> well you can try look closely yourself too of course
<oliv3r> try to find pin 151 and 152
popolon has joined #linux-sunxi
<hramrach> probably no connection or going under the chip
<hramrach> there is some other part right next to A13 there so quite hard to look at that spot
<hramrach> but hey, those are taken by CTP, anyway. At least the fex says so
pacopad has quit [Quit: pacopad]
<oliv3r> pin 150 should be marked with a white dot
<hramrach> yes, it is
<oliv3r> on my tablet 151 and 152 are about 1mm long, then go to a via
<oliv3r> on the other side; there's 2 pads
<hramrach> on my tablet 1mm from the A13 there is a row of random SMD parts
<hramrach> What AXP is commonly found with A13?
<hramrach> the AXP I have seems to be ?09
<hramrach> can't read the middle
<oliv3r> 209
<oliv3r> either you have axp 152; or 209
<hramrach> ok, 209 is likely
paulk-collins has quit [Quit: Ex-Chat]
penghb has joined #linux-sunxi
rz2k has quit [Read error: Connection reset by peer]
brain_ has joined #linux-sunxi
brain__ has quit [Quit: Leaving]
<hramrach> yes, it's 209. found a readable angle :)
yessikattal has joined #linux-sunxi
<yessikattal> los hombres escribanme en privado
<yessikattal> quien es hombre diga yoo
FR^2 has quit [Quit: und weg...]
<oliv3r> hramrach: the only other 09 i know, is 809
<oliv3r> which is a80 stuff
yessikattal has quit [Remote host closed the connection]
yessikattal has joined #linux-sunxi
yessikattal has quit [Remote host closed the connection]
mkutsevol has joined #linux-sunxi
ganbold_ has quit [Remote host closed the connection]
sil_ has joined #linux-sunxi
tomboy64 has joined #linux-sunxi
<Seppoz> hramrach : i agree
techn__ has quit [Ping timeout: 245 seconds]