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
torindel has quit [Ping timeout: 246 seconds]
torindel has joined #linux-sunxi
tm512 has quit [Ping timeout: 246 seconds]
torindel has quit [Ping timeout: 272 seconds]
torindel has joined #linux-sunxi
tm512 has joined #linux-sunxi
tm512 has quit [Ping timeout: 246 seconds]
hramrach has quit [Ping timeout: 240 seconds]
tm512 has joined #linux-sunxi
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
hramrach has joined #linux-sunxi
popolon has quit [Quit: Quitte]
torindel has quit [Ping timeout: 248 seconds]
torindel has joined #linux-sunxi
popolon has joined #linux-sunxi
popolon has quit [Read error: Connection reset by peer]
torindel has quit [Ping timeout: 240 seconds]
torindel has joined #linux-sunxi
torindel has quit [Ping timeout: 246 seconds]
TheSeven has quit [Read error: Connection reset by peer]
TheSeven has joined #linux-sunxi
torindel has joined #linux-sunxi
torindel has quit [Ping timeout: 245 seconds]
torindel has joined #linux-sunxi
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
torindel has quit [Ping timeout: 256 seconds]
torindel has joined #linux-sunxi
torindel has quit [Ping timeout: 272 seconds]
torindel has joined #linux-sunxi
benn____ has quit [Ping timeout: 246 seconds]
egbert has quit [Disconnected by services]
egbert has joined #linux-sunxi
benn____ has joined #linux-sunxi
torindel has quit [Ping timeout: 256 seconds]
torindel has joined #linux-sunxi
benn____ has quit [Ping timeout: 245 seconds]
benn____ has joined #linux-sunxi
Gerwin_J has quit [Quit: Gerwin_J]
hramrach has quit [Ping timeout: 240 seconds]
Gerwin_J has joined #linux-sunxi
benn____ has quit [Ping timeout: 240 seconds]
kaspter has joined #linux-sunxi
torindel has quit [Ping timeout: 240 seconds]
torindel has joined #linux-sunxi
techn__ has quit [Read error: Connection reset by peer]
techn_ has joined #linux-sunxi
benn____ has joined #linux-sunxi
torindel has quit [Ping timeout: 272 seconds]
benn____ has quit [Client Quit]
torindel has joined #linux-sunxi
Gerwin_J has quit [Read error: Connection reset by peer]
torindel has quit [Ping timeout: 264 seconds]
torindel has joined #linux-sunxi
\\Mr_C\\ has quit [Quit: .]
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
torindel has quit [Ping timeout: 256 seconds]
torindel has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
tm512 has quit [Quit: Just relax and enjoy this pleasant adventure]
torindel has quit [Ping timeout: 272 seconds]
torindel has joined #linux-sunxi
hipboi has joined #linux-sunxi
torindel has quit [Ping timeout: 248 seconds]
torindel has joined #linux-sunxi
TheSeven has quit [Disconnected by services]
[7] has joined #linux-sunxi
tm512 has joined #linux-sunxi
awatch has quit [Quit: Ex-Chat]
book` has quit [Ping timeout: 248 seconds]
awatch has joined #linux-sunxi
kenny has quit [Ping timeout: 256 seconds]
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
awatch has quit [Read error: Operation timed out]
book` has joined #linux-sunxi
kenny has joined #linux-sunxi
eebrah has joined #linux-sunxi
awatch has joined #linux-sunxi
bamvor has joined #linux-sunxi
torindel has quit [Ping timeout: 246 seconds]
awatch has quit [Ping timeout: 241 seconds]
Black_Horseman has quit [Quit: Zwi se logou mou!!!]
torindel has joined #linux-sunxi
eebrah has quit [Read error: Connection reset by peer]
rz2k has joined #linux-sunxi
eebrah has joined #linux-sunxi
eebrah is now known as Guest82331
Guest82331 has quit [Read error: Connection reset by peer]
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
eebrah_ has joined #linux-sunxi
torindel has quit [Ping timeout: 246 seconds]
torindel has joined #linux-sunxi
rz2k has quit []
torindel has quit [Ping timeout: 245 seconds]
torindel has joined #linux-sunxi
eebrah_ has quit [Ping timeout: 245 seconds]
torindel has quit [Ping timeout: 272 seconds]
torindel has joined #linux-sunxi
Sonic1 has quit [Ping timeout: 272 seconds]
Sonic1 has joined #linux-sunxi
rellla has joined #linux-sunxi
rellla2 has joined #linux-sunxi
JohnDoe_71Rus has quit [Read error: Connection reset by peer]
<gzamboni> hramrach you have to debootst as su (root)
<gzamboni> moin
deasy has joined #linux-sunxi
<arokux2> hno: oliv3r I was talking about a separate branch in u-boot-sunxi, not about merging yet
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has joined #linux-sunxi
HeHoPMaJIeH has quit [Changing host]
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
hramrach has joined #linux-sunxi
JohnDoe_71Rus has joined #linux-sunxi
_massi_ has joined #linux-sunxi
torindel has quit [Ping timeout: 246 seconds]
torindel has joined #linux-sunxi
panda84kde has joined #linux-sunxi
BluesBoy has joined #linux-sunxi
hipboi has left #linux-sunxi ["Leaving"]
FR^2 has joined #linux-sunxi
n01 has joined #linux-sunxi
<hramrach> hello
arokux2 has quit [Remote host closed the connection]
<hramrach> is something like this upstreamably in some way so that you can pass the mac used by u-boot to kernel when there is no ROM http://pastebin.com/4aLtNsyE
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
<mripard> hramrach: no
<mripard> the bootloader should set the mac address in the DT, in the local-mac-address property
<mripard> and the emac driver already reads that field if it's present
notmart has joined #linux-sunxi
notmart has quit [Changing host]
notmart has joined #linux-sunxi
<oliv3r> arokux1: where is your u-boot branch; i don't see it on your github :(
Seppoz_nowork has joined #linux-sunxi
vicenteH has quit [Ping timeout: 264 seconds]
Seppoz has quit [Ping timeout: 260 seconds]
<arokux1> oliv3r: it is not there yet. that is why I've asked if I can create my branch at linux-sunxi/u-boot-sunxi instead of forking into my account.
<hramrach> mripard: so to conform to this u-boot would have to patch the DT with the fabricated mac address
<mripard> to have a consistent mac address between u-boot and linux, yes
<oliv3r> arokux1: your dev work should be in your own repo; we don't really want to have dev branches in the official repo
<arokux1> oliv3r: that is not a problem for me at all, but it seems to contradict to linux-sunxi/linux-sunxi where mnemoc encourages to host WiP branches there, so his scripts can do hourly builds.
<oliv3r> mnemoc: huh?
<oliv3r> arokux1: i don't see any 'wip/dev' branches, only ones that are 'done and need testing'
<oliv3r> not random dev kernels
<arokux1> oliv3r: sunxi-next-usb
<oliv3r> that is done though isn't it
<arokux1> oliv3r: it is not marked as WiP
<oliv3r> or is it your wip/ stuff?
<arokux1> oliv3r: no
<arokux1> oliv3r: yes, EHCI done, working on OHCI
<arokux1> oliv3r: but I try to commit working stuff
<oliv3r> anyway, so far, we haven't done that with u-boot, hno said 'dev happenes in developers own repo's'
<oliv3r> arokux1: did you solve your EHCI bug?
<arokux1> oliv3r: yep
<oliv3r> awesome
<oliv3r> what was it?
<arokux1> oliv3r: nothing, there was no bug :)
<arokux1> oliv3r: some errors are there, but they seem to be harmless
<oliv3r> so wifi still doesn't work?
<arokux1> oliv3r: it does and did. I just didn't tried because of the errors, but as said, they seem to be harmless
<arokux1> mnemoc: it seems like you need to extend you scripts so that they can cope with developers trees not only ones under linux-sunxi
<arokux1> oliv3r: all USB devices that I have work. there were some problems with USB sound cards. do you have any? :p
<oliv3r> 2 or 3 yeah
<arokux1> oliv3r: so maybe you can give my branch a try
<oliv3r> yeah once i have some sunxi time i will
<oliv3r> i wanted to review your tree first :)
<arokux1> oliv3r: thanks.
<oliv3r> so if you can push it to your github; i'll fetch it and git diff it
rellla3 has joined #linux-sunxi
<arokux1> oliv3r: you mean u-boot tree?
<oliv3r> aye :)
<oliv3r> i have a few minutes i can spend on looking over it
<arokux1> oliv3r: sorry, I have it at home and i'll be there on sun evening :(
<arokux1> oliv3r: wait with reviewing of the kernel tree till I've integrated OHCI stuff
<oliv3r> ah ok, no prob; i'll review your u-boot stuff on sunday then :)
<arokux1> ok :)
<oliv3r> if it's as good as I expect itt o be; we can merge it to the head
<arokux1> oliv3r: alright.
<arokux1> it seems like my sun4i board (Mele A1000) has died, so now I'm very lucky to have sun5i (A13-olinuxino) and very soon sun7i (A20-cubietruck) thanks to Olimex and Cubietech
<oliv3r> i'm supprised that it died; probably something wrong with the ram though
popolon has joined #linux-sunxi
<arokux1> oliv3r: nothing changed, same software - hardware is highly unstable.
<oliv3r> yeah i think your memory is having power supply issues or something
<popolon> I noticed recent kernel are far less stable than previous one
n01 has quit [Ping timeout: 240 seconds]
<popolon> no good results at all with recent kernel
<popolon> after mnemoc , that's init that crash
<popolon> or userspace programs
<popolon> but everything depend on kernel version
<popolon> no more success with 3.4 or stage 3.4 :(
<popolon> don't know at all, how to find what is the source of the problem
<popolon> I can still use old one
rellla3 has quit [Quit: Nettalk6 - www.ntalk.de]
vicenteH has joined #linux-sunxi
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
enrico_ has joined #linux-sunxi
royger has joined #linux-sunxi
torindel has quit [Ping timeout: 240 seconds]
torindel has joined #linux-sunxi
Soru has joined #linux-sunxi
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
\\Mr_C\\ has joined #linux-sunxi
jinzo has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
<mnemoc> arokux1: ok
<oliv3r> mnemoc: i think these guys aren't quite sure how far the software stack is for the cb
kaspter has quit [Remote host closed the connection]
<mnemoc> popolon: try the previous sunxi-v3.4 tag
<mnemoc> oliv3r: meaning?
<mnemoc> popolon: currently 3.4 and stage-3.4 are identical
<mnemoc> popolon: but we make tags to allow bisecting
<popolon> Is it a git tag ?
<mnemoc> yes
<mnemoc> if the latest tag fails, try the previous. and so one
<mnemoc> on
<mnemoc> until you find two good vs bad tags
<popolon> is there a mean to debug more deeply at boot, and give you more return on errors ?
<oliv3r> mnemoc: that they are asking a little to early for what they want
<oliv3r> i guess they expected the cb etc stacks to be as advanced as those of the rasppi etc
<oliv3r> where everything just works (for the most part)
<mnemoc> maybe in the process we can fix those weeknesses....
<popolon> I understand the cb2/A20 I use are really new.
<mnemoc> but yes. it's understandable
<mnemoc> popolon: not you. some other fellows who contacted oliv3r privately
<popolon> ok
<oliv3r> mnemoc: heh, 3D is somewhat workable, but not amazing, and video decoding is not really done or workable either way; and taht's what they'd want, a fully fledged stable working mediacenter
* mnemoc slapped by reality
<popolon> there is already lot of progress in few month
Black_Horseman has joined #linux-sunxi
<mnemoc> popolon: totally agree, but it's still a not-out-of-the-box platform
fredy has quit [Excess Flood]
fredy has joined #linux-sunxi
<popolon> yes
<oliv3r> you can't buy a CB, pop in an SD card and have a fully working mediacetner (ala raspbmc)
<popolon> should probably be in BIG characters on linux-sunxi mainpage ?
<oliv3r> cedarX crashes randomly, doesn't support files etc
torindel has quit [Ping timeout: 264 seconds]
<oliv3r> and we're getting there, devs are hard at work as time perimits
<juanfont> oliv3r, is allwinner helping?
torindel has joined #linux-sunxi
book` has quit [Remote host closed the connection]
<oliv3r> juanfont: LOL
vicenteH has quit [Read error: Connection reset by peer]
vicenteH has joined #linux-sunxi
<juanfont> oliv3r :(
book` has joined #linux-sunxi
rings_IIV has quit [Read error: Connection reset by peer]
rings_IIV has joined #linux-sunxi
paulette has joined #linux-sunxi
<paulette> I'm fall in a strange USB issue with 3.0.101 and android openbox 4.0.4 when I disable or enable "Developper options". Seem driver enter in a big loop (20-50 secs). I was able to capture kernel messages with TTL and cat /proc/kmsg: http://pastebin.com/pEQB91je . Any idea about this strange behavior ?
mouchon has quit [Ping timeout: 245 seconds]
mouchon has joined #linux-sunxi
<oliv3r> other then saying that 3.0 is 'EoL'; looks like the driver is choking, not sure on what though
rings_IIV has quit [Read error: Connection reset by peer]
rings_IIV has joined #linux-sunxi
<oliv3r> mnemoc: btw; k9 DOES do threading, it simply groups mssages so it looks like one, and when you click on it, you see the messages related to that group
popolon has quit [Quit: Quitte]
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
<steev> what what!
<steev> er, i mean, good morning
<steev> common typo
<steev> eesh
<steev> webkit-gtk is at 10 hours compiling on the cb2
<oliv3r> lol
<oliv3r> distcc!
<steev> i can't!
<steev> i could if i used crappy generic cflags like i was some other shitty distro
<steev> also, i need to make sure it actually builds before even introducing a possible failure because distcc
<paulette> What mean 'EoL' ? I'm not so fine with Achronims
<steev> end of life
<steev> the linux-sunxi developers have only so much time to devote to various kernel versions, and they've decided on 3.4 for support, and 3.12(?) for mainline/future works
<paulette> However I've tried to use 3.4 instead of 3.0.101 and it fall in so much problems ... 3.0 works for all than usb
<oliv3r> paulette: then you have something special that needs to be fixed, as 3.4 runs quite fine on most device
<paulette> or better USB works but when a reset is asked for change mode (for example for Accessory Mode) it become crazy
<paulette> I can back to 3.4 if you say that work fine with android
<steev> shouldn't be any reason it doesn't
<steev> worst case, you need to add some android specific stuffs?
<paulette> for EG: touch doesn't work (also with module running), same for mouse click
* steev doesn't really touch android much these days
<steev> at least, not without getting paid
<paulette> I mean resistive touch sun4i_ts.ko
<steev> working on android for free is stupid
<oliv3r> restive touch panel should 'work' as that's part of the SoC, rather then the various i2c capacatives
<paulette> with 3.0 I've used sun4i_crane_defconfig, same with 3.4 but something goes wrong with touch and mouse
<paulette> ok I will try to back to 3.4 and I will check what I've mistake
<paulette> however the question is: 3.0.x is ended (no more support from any) and 3.4 is the right horse
<oliv3r> 3.0 isn't supported by us much anymore, since there's so much to d o, and only so little time
<paulette> ok ... I will spend some hours to try to understand the bug and if I found something I will write here what is
Sonic1_ has joined #linux-sunxi
Sonic1 has quit [Ping timeout: 240 seconds]
rellla has quit [Ping timeout: 240 seconds]
rellla2 has quit [Ping timeout: 260 seconds]
<oliv3r> paulette: awesome :)
Black_Horseman has quit [Remote host closed the connection]
<paulette> unforunately I NEED AOA working for a class project using CB1
rellla has joined #linux-sunxi
rellla2 has joined #linux-sunxi
<oliv3r> AoA? Android on Android?
<paulette> Android Open Accessory
<oliv3r> you and your abbreviations! :p
<oliv3r> ah ok
<oliv3r> shouldn't be a big prob
<paulette> Actually system get crazy when I send ACCESSORY_START usb command from PC to android device (CB1)
<Sonic1_> AoA action over here
<paulette> just after that: libusb_control_transfer(handle, 0x40, ACCESSORY_START, 0, 0, NULL, 0, 0); usb driver enter in infinite loop
<libv> cubietruck claims to need 5.2V?
<Turl> libv: can't you power it via usb?
<libv> i have a load of universal power supplies
<libv> but those never go to .2 of a V :)
<libv> and 2A from a usb socket is not allowed
<libv> you would need a proper pure usb power supply for that
<oliv3r> yeah A20 is too power hungry for pure USB power
<libv> usb is 100mA, with negotiation up to max 500mA
<oliv3r> it can be done with light loads, but my olimexino even fails to boot over USB (otg)
<oliv3r> ironically, with the supplied usb -> powerconnector I can boot using 500 mA
<oliv3r> this though is from a hub in my monitor
<Turl> I can boot cb2 with a motorola charger, probably <1A (completely headless though)
<Turl> I kind of burned something on my monitor when plugging it there :|
<Turl> CB1 used to work fine however
<oliv3r> yeah A20 power draw can be a little to high for the power plug
<oliv3r> i think you can charge it via otg just fine
<oliv3r> since you normally have a battery that catches these big draws
<oliv3r> and in charges in standby mode
<Turl> I don't have a batt on cb2 :p
<oliv3r> i'm saying on a tablet :p
<Turl> ah
<oliv3r> and a set-top-box has a big adapter anyway
<oliv3r> well i didn't say it, i ment it! :)
<Turl> :p
<Turl> well, gotta go :) cya later
<oliv3r> :)
deasy has quit [Quit: Nom d'un quark, c'est Edmonton !]
popolon has joined #linux-sunxi
<libv> did we ever figure out the livesuite protocol?
<libv> iirc, a year and a half ago, hno was working on that one, and then a former dutch telechips poker called ithamar also did some things like half a year later
<libv> well, he's still dutch, just former telechips poker
<libv> looks like i will be uploading mesa for saucy today
<oliv3r> libv: awesome! :)
<oliv3r> and yeah, i think the protocol is mostly known
<oliv3r> but it might not be needed with fel-mode tricks
<libv> but no-one bothered for long enough to provide a proper tool and a proper writeup?
vicenteH has quit [Ping timeout: 272 seconds]
<oliv3r> i think we need a tooling to actually do everything, but the technical possibility is known
<panda84kde> libv: hi. with "mesa for saucy" do you mean a mesa package with "some form of lima driver"?
<oliv3r> i think nobody has engaged in that because the mtd stuff isn't merged yet
<libv> panda84kde: yes and no
<libv> panda84kde: it's a slightly hacked mesa which allows you to build and run the upcoming lima driver against
<libv> +it
<libv> the saucy mesa needs just one patch to allow using the binary shader compiler
<libv> (just exporting highlevel hooks for compiling and linking shaders)
<panda84kde> libv: I see. Thanks
<libv> and then 3 patches to provide a libdricommon, some pkgconfig files, and then some header files which install into the system
<libv> but more importantly, the lima mesa driver now works from mesa-8.0.4 until mesa-9.2.1
<libv> there will be no rebuilding kernels or no reinstalling large parts of your installation for lima
<oliv3r> big win!
<libv> you should be able to pull the latest lima, build it, install it, and be done
<libv> oliv3r: it's scary enough getting anything to work on any arm board
<libv> so once you have something working, you never ever want to change it
<libv> so if i, like others, require an updated kernel, an updated libdrm, mesa, xserver, ..., then noone is going to end up using lima
<oliv3r> yeah
<libv> and if the latest lima does not work... you check out a slightly older version of lima, and restore functionality
<oliv3r> for us it's no biggie to swap parts out, but for general users, the less they have tos wap the better
<libv> and then i know exactly where to search for fixing the problem
<libv> yup
<libv> i have been whining about this for the best part of the decade
<libv> but none of the xorg community accept this as a valid goal
paulette has quit [Quit: Leaving]
<libv> but then, they are known for nastily killing proper open source projects, so it's not as if their opinions should be valued too much
shineworld has joined #linux-sunxi
<panda84kde> I'm surely late but: http://liliputing.com/2013/10/arm-unveils-mali-t760-mali-t720-graphics.html Is it bad that I see no "Allwinner" listed as licensee
<libv> panda84kde: it will be about a year until those ship
<libv> we do not even have t62x or t65x shipping yet
<libv> only t604 is out and about
torindel has quit [Ping timeout: 256 seconds]
<panda84kde> yeah, so Allwinner can be in the it for the future
torindel has joined #linux-sunxi
<libv> if they want to... it seems that they are mostly on a pvr (death)trip atm
pfdm has joined #linux-sunxi
<panda84kde> oh. I won't but them then.
<panda84kde> *buy
<popolon> libv, I believe t658 is in Exynos included in Note III and Note 10.1 2014 (international version, not us one)
<popolon> exynos 5420
<popolon> as far I know, only exynos 5250 use mali T604
<libv> isn't that a 5410?
<oliv3r> libv: from what i heard, is that sinc ethey payed money for the PowerVR IP, they will bring out a few socs with it; RoI and all that
<popolon> 5410 use powervr
<libv> in the note3?
<popolon> in note IV that's a 5410
<pfdm> rellla: I figure it out ! So for those who are playing with libcedarv et libvecore, they are doing some kind of nasty initialisation (probably a #define) in libvecore which depends on something in libcedarv. Therefor libcedarv has to be loaded before libvecore to avoid a segfault !
<libv> ah, right, it is shipping already
<popolon> yes
<popolon> :)
<rellla> pfdm: what is your base repo of xbmc you are working on?
<libv> oliv3r: afaict, olimex and cubie are giving allwinner a miss because of that
<libv> oliv3r: so i am not sure that this was the most sensible business choice
<oliv3r> oh yeah, i told both to avoid a31 like the plague :)
<libv> drachensun of course has no issues with pvr
rellla2 has quit [Quit: Nettalk6 - www.ntalk.de]
<oliv3r> well powerVR was bought by AW before they even knew olimex/cubietech would avoid those chips
<mnemoc> gentoo boys :/
<mnemoc> err.. wrong scroll
<libv> mnemoc: ?
<oliv3r> mnemoc: ubuntunoob!
<Sonic1_> popolon: T628MP6 is in the Exynos 5420, not a T658
<mnemoc> libv: replying a comment from steev
<mnemoc> but three pages behind :|
<steev> mnemoc: fail!
<libv> anyway, we will see scores of t624 and perhaps a t658 shipping before t7xx is out
<mnemoc> steev: yup
<popolon> Sonic1_, have you a ref ?
<steev> it happens
<libv> and even then, i do not expect t7xx to be such a huge difference from t6xx
<libv> definitely not of the "it feels the same, but is mostly different" change between the mali m400 and mali t604
<pfdm> rellla: I was working on vidonme, but then i switch to yours for test and i get the same behavior, now i get yours to display a picture (not good now), but im gonna rebuild the videonme one for test ! I've seen there are some difference in the way of managing linuxrenderA10, specially the OSD.
<Sonic1_> popolon: I don't need a fact check when I know what it is. The Exynos 5420 is T628MP6. The Galaxy Note uses the Exynos 5420 and Snapdragon 800 depending on region. The Exynos 5410 uses a PVR544MP3. The Samsung Galaxy S4 uses The Exynos 5410 and a Snapdragon 600 depending on region. The Exynos 5250 uses a Mali-T604. The Samsung ARM Chromebook and Nexus 10 use the Exynos 5250
<libv> :)
<popolon> that's to update wikipedia if that's the case
<libv> popolon: no, it is not
<popolon> 628 MP6
<popolon> means the second generation or T6xx
<libv> popolon: yes
<popolon> (but need to update the french version then :)
<popolon> anyway source given seem wrong (need to double check)
<libv> popolon: this is the trouble with translating from english.
<libv> we had this discussion before
<libv> it might work for an encyclopedia
* popolon is wrong in fact
<Sonic1_> Numbers are hard when translating from English to French
<libv> but it definitely should not be done for technical information
<popolon> libv, i didn't translated from english
<libv> as it is hard enough to keep one page correct and up to date
<popolon> but added fact when I found information in newspaper (all was wrong, perhaps samsung announce was wrong at this time)
<popolon> ok
<popolon> the ref say that
<popolon> :)
<popolon> but even samsung site i wrong sometimes
<popolon> on the galaxy note 10.1 (2013) the wrote 1 GB of ram instead of 2, on UK and FR sites ....
<popolon> the device has 2 GB of ram, I've one
<popolon> that's strange error for a marketing site
<Sonic1_> Did they mess up and say the Note 10.1 2014 edition doesn't have 3GB of RAM as well? :P
<popolon> LPDDR3e RAM for exynos 5420
<popolon> didn't know the 'e' version
<rellla> pfdm: iirc vidonme misses the whole /dev/fb0 thing as it is made for android
<rellla> also you need the magic libbdv.so which seems to be the wrapper around allwinner's blob!?
<rellla> if i may ask - what exactly do you mean by libcedarv and libvecore? i'm mixing up things... can you redirect me to the actual sources on github you are working on and testing?
<pfdm> rellla: yes they miss fb0 ! well for libbdv.so it just call libcedarv, so its just a rename for me. I mean libvecore.so the one we don't have the source, and libcedarv.a or .so, the wrapper with vdecoder.
<pfdm> there are the blob source i'm using, with some modification for debugging. https://github.com/patricius972/cedarx-libs, and the vidonme fork https://github.com/patricius972/xbmc
<rellla> and how do you get libbdv.so? or better, how does your vidonme fork use cedarx-libs without it?
<pfdm> in the fork they load libbdv.so , i renamed cedarx-libs libbdv.so by convenience and it works. Because they use only 4 functions in libbdv.so which are libcedarv_init,exit, mem_palloc,mem_pfree , mem_get phyadrr
<pfdm> all these function are in cedarx-libs
<pfdm> with the same prototype so it's just a name thing
<rellla> pfdm: ok. got it.
<pfdm> rellla: Is /dev/fb0 absolutely necessary to linuxrenderA10 ? or just calling disp is enough?
<rellla> don't know. question for the disp guys here ;)
n01 has joined #linux-sunxi
shineworld has quit [Quit: Leaving]
<rellla> pfdm: seems that you don't need it at all ...
<drachensun> libv: A lot of people did gripe about the powervr, so I might have been better off with something else, but alternatives were dual core with low res screen, high performance device with a brand name chip that I would have had to sell for more than retail, maybe the Rockchip but it was more expensive with incomplete tools and source from what I could tell
<drachensun> I dont think thats what killed it though
<drachensun> none of the press that wrote about it the first time would write about it this time "too many crowd funders out there"
<rellla> <pfdm>rellla: I figure it out ...
<rellla> can you re-explain it for me? how did you solve it?
<pfdm> rellla : i was playing with my small example, and then i thought maybe something in my makefile was wrong, so i compile it with the same flags as xbmc, and then i got the segfault. The problem is the order of loading the library
<pfdm> libcedarv must be loaded before livecore to make it work
<pfdm> I've read some stuff on this common c++ mistake. When something is initialized in libvecore which depends on the initializatiion in libcedarv, and because it is at loading, i guess its a constant like a define
<pfdm> once this "symbol" is present after loading libcedarv , when loading libvecore there is no more segfault and libve_open succeed
HeHoPMaJIeH has quit [Remote host closed the connection]
n01 has quit [Ping timeout: 252 seconds]
<oliv3r> rockchip doesn't even have an open bootloader; so that's a hard sell to the linux crowd imo
<hramrach> PC does not have open bootloader either, and there it is, running Linux
<hno> libv, I focused on the FEL protocol. The protocol used by livesuit after it has brought the board up seems to be very similar, but have no interest in what protocol those two proprietary pieces of software is talking to each other. (livesuit and the program livesuit uploads to the board using FEL protocol)
<oliv3r> hramrach: no pc runs coreboot?
<oliv3r> hramrach: but let me rephrase it to suit you better, rockchip is GPL violating with their usage of u-boot
<hramrach> PCs there were reverse-engineered run coreboot. As reverse-engeneered ARM SoCs have open bootloader
<oliv3r> with a regular PC bios, atleast you know it's propriatary form the get-go and know there is no violation and with no alternatives, you just have to swallow that pill; or get coreboot supported
<oliv3r> hramrach: what soc runs a RE bootloader at this time?
<hramrach> I don't really know the details but given the typical approach I would expect you get same junk everywhere
<hramrach> on AW you get partial sources
<hramrach> after much complaininig
<oliv3r> hramrach: we got full u-boot sources
<oliv3r> and how can you say that's a bad thing? they where violating the GPL, they where obligated to share it
<oliv3r> as is rockchip
<oliv3r> but rockchip doesn't care
<hramrach> next to nobody cares. If they feel like it they can use the GPL issue as excuse but they know you will not really sue them
<WarheadsSE> ZyXEL cracks me up
<WarheadsSE> you bitch, they open a 24h only ftp account
<WarheadsSE> and ONLY for the EXACT sources you asked for
<oliv3r> lol
<oliv3r> mirror it all
<WarheadsSE> i did
<WarheadsSE> well, correction: I ripped out what I needed, brought it to mainline capability, and then release the source
vicenteH has joined #linux-sunxi
eebrah has joined #linux-sunxi
<rellla> pfdm: libcedarv.a is linked in xbmc.bin while building, so it everything libvecore.so is based on, should be available at runtime.
<pfdm> rellla: i use a shared library version. easier to test the same version without relinking
torindel has quit [Ping timeout: 240 seconds]
<rellla> yeah, then you have to ensure, that your shared lib is available when accessing libvecore.so, because libvecore.so depends on it
torindel has joined #linux-sunxi
<pfdm> It's more than just available, because in the old version the order of loading them doesnt crash it. because basically it stops when you have undefined symbol. But agree if it's static lib, this problem should not show up.
Soru has quit [Remote host closed the connection]
torindel has quit [Ping timeout: 246 seconds]
torindel has joined #linux-sunxi
soul has joined #linux-sunxi
soul is now known as Guest16442
<pfdm> rellla: have u tested the new blob ? what's your status on this ?
<drachensun> oliv3r: yeah, lack of u-boot source would have made my dual boot control impossible, not to mention offended my customers.
<drachensun> I've just noticed MTK is in some acer tablets
<rellla> pfdm: i wanted to do all at once. use the new lib, on a20, try the new allocator ... everything using the old kernel, handy patched....
<drachensun> and the kernel source is available from them
<drachensun> but the notes in the kernel release say u-boot has been replaced by something else in MTK's Android 4.2
<drachensun> so no source on whatever that is
<drachensun> or the Android build
<drachensun> or access to the image creation tools
<rellla> pfdm: have stopped that, because of no success. maybe i should think a bit more next time instead of doing some try and error...
<drachensun> I did managed to find the image flashing program before but without being able to remake images, it wasn't really an option either
<pfdm> rellla: you mean the sunxi_mem ? ok, I think if we can have at least the new blob working as the old one, it'd be ok. Hoping they fixed the h264 artifact.
<rellla> pfdm: yes. i agree. my next step will be to set up a a20-system with cedar-sun7i in kernel. then first give the old blob/xbmc a try.
n01 has joined #linux-sunxi
<rellla> once succeeded, try xbmc with new blob and old allocator a try as you did. then maybe take a look at sunxi_mem
<rellla> pfdm: once working, it would be good to point out the differences between empat0 and vidonme implementation - and improve xbmc code.
<pfdm> I am on archlinux, i have some PKGBUILD and packages if you want . For me , which the latest kernel cedar patch xbmc/old blob, is working as before.
<pfdm> rellla: yes true. I've started doing that, I need to understand more what disp is doing, and they add some bitstream modification apparently taken from ffmpeg .
<rellla> pfdm: debian here.
<rellla> pfdm: maybe, huceke's xbmc for allwinner is worth a try, too.
<pfdm> rellla: do you have a link for this repo ?
<rellla> and if we have at least new blob running, i suggest to use the same wrapper for libvecore in xbmc and vlc like willswang does it. we probably can decide in this wrapper to use our RE'ed code or the closed source binary deoending on the codec requested.
<rellla> huceke aka gimli in irc
<libv> hno: what does that mean?
<libv> hno: that we have the knowledge to flash nand directly?
<pfdm> rellla: thanks. Do you know , how does this version compare to the empat0 one ?
alabaster has quit [Ping timeout: 245 seconds]
<rellla> pfdm: huceke uses a overlay renderer, but i have no clue, what is the difference. for decoding, he uses his own dll https://github.com/huceke/libcedar/
alabaster has joined #linux-sunxi
BluesBoy has quit [Quit: HydraIRC -> http://www.hydrairc.com <- \o/]
<rellla> they talked to each other about a year ago about some problems in using libvecore, but then both stopped developing, once found out that the blob is buggy... i never tried huceke's version and i'm not familiar with it.
n01 has quit [Read error: Connection reset by peer]
<rellla> pfdm: maybe this is of interest for you, too: http://linux-sunxi.org/XBMC#Sources_implementing_A10_on_XBMC . cross-checking a10 vlc code could help, too ... i don't know of more cedar implementations ;)
<pfdm> rellla: Ok , the code looks different , eventhough i think vidonme version is mostly based on empat0 patch, they might have taken some idea there (bitstream converter stuff) . I guess it gives the same problem with the old blob. Might worth trying in on the new ones, ill see if i have time for a quick compile.
<rellla> good luck, i'll do my next tries probably this weekend...
awatch has joined #linux-sunxi
rellla has quit [Quit: Nettalk6 - www.ntalk.de]
<pfdm> rellla: thanks ill let you know
torindel has quit [Ping timeout: 246 seconds]
torindel has joined #linux-sunxi
_massi_ has quit [Quit: Leaving]
awatch has quit [Quit: Ex-Chat]
torindel has quit [Ping timeout: 272 seconds]
torindel has joined #linux-sunxi
WarheadsSE has quit [Ping timeout: 272 seconds]
Sonic1_ is now known as Sonic1
WarheadsSE has joined #linux-sunxi
tm512 has quit [Quit: Just relax and enjoy this pleasant adventure]
jinzo has quit [Ping timeout: 256 seconds]
eebrah has quit [Ping timeout: 272 seconds]
enrico_ has quit [Quit: Bye]
eebrah has joined #linux-sunxi
eebrah is now known as Guest54156
royger has quit [Quit: leaving]
deasy has joined #linux-sunxi
FR^2 has quit [Quit: Connection reset by peer]
paulk-aldrin has joined #linux-sunxi
torindel has quit [Ping timeout: 240 seconds]
paulk-aldrin is now known as paulk-leonov
paulk-leonov has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
paulk-aldrin has joined #linux-sunxi
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
panda84kde has quit [Quit: Konversation terminated!]
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
<pfdm> I have a question regarding poweroff, why it does not turn off the green led ? and/or the red one ?is that a flex issue ?
pfdm has quit [Ping timeout: 250 seconds]
Guest54156 is now known as eebrah_
eebrah_ is now known as eebrah
alabaster is now known as hint
eebrah has quit [Read error: Connection reset by peer]
hint is now known as alabaster
torindel has quit [Ping timeout: 246 seconds]
gzamboni_ has joined #linux-sunxi
torindel has joined #linux-sunxi
paulk-collins_ has joined #linux-sunxi
eebrah has joined #linux-sunxi
eebrah is now known as Guest49348
popolon has quit [*.net *.split]
paulk-collins has quit [*.net *.split]
gzamboni has quit [*.net *.split]
ynezz_ has quit [*.net *.split]
Guest49348 is now known as eebrah_
torindel has quit [Ping timeout: 246 seconds]
torindel has joined #linux-sunxi
pfdm has joined #linux-sunxi
torindel has quit [Remote host closed the connection]
torindel has joined #linux-sunxi
rings_IIV has quit []
jemk has joined #linux-sunxi
rings_IIV has joined #linux-sunxi
eebrah_ has quit [Read error: Connection reset by peer]
eebrah has joined #linux-sunxi
ynezz_ has joined #linux-sunxi
popolon has joined #linux-sunxi
\\Mr_C\\ has quit [Read error: Connection reset by peer]
\\Mr_C\\ has joined #linux-sunxi
eebrah has quit [Read error: Connection reset by peer]
eebrah has joined #linux-sunxi
ynezz_ is now known as ynezz
iamfrankenstein has quit [Ping timeout: 272 seconds]
paulk-collins_ has quit [Read error: Connection reset by peer]
iamfrankenstein has joined #linux-sunxi
eebrah has quit [Ping timeout: 272 seconds]
eebrah has joined #linux-sunxi
eebrah is now known as Guest4878
FR^2 has joined #linux-sunxi
<Nyuutwo> is there any tool to dump iomux settings?
<oliv3r> Tsvetan: how many capacitors are missing on the bottom of the olimexino-micro-a20? I just, and don't even know how that's possible, but dropped a gop of solder on exactly the bottom of the CPU. i carefully removed it, but there's 2 missing capacitors, and I don't remember if those where missing before. I think they where, but i dont' know for sure
<mnemoc> ouch
<mnemoc> Nyuutwo: look at pio in sunxi-tools
<Nyuutwo> thank you
<Nyuutwo> mnemoc: it looks that this tool just does gpio part not mux
<hno> libv, the livesuit protocol have nothing to do with flashing. It's just a transfer protocol.
<hno> And yes we have the knowledge to flash to nand directly. Even code for doing it (yuq on github), just not polished.
<hno> livesuit first transfers a bootstrap that configures the board, then a better USB protocol handler and finally the actual flasher application and image to be flashed, and runs the flasher application on the board.
iamfrankenstein has quit [Ping timeout: 260 seconds]
Guest4878 has quit [Quit: leaving]
<Nyuutwo> i'm trying to use sunxi-reg to dump pin muxer, but on android I don't have adress file in sysfs, only dump, write and compare
<Nyuutwo> I probably found way of using it
notmart has quit [Quit: notmart terminated!]
Black_Horseman has joined #linux-sunxi
Black_Horseman has quit [Changing host]
Black_Horseman has joined #linux-sunxi
pfdm has quit [Ping timeout: 250 seconds]
techn__ has joined #linux-sunxi
techn_ has quit [Ping timeout: 240 seconds]
Black_Horseman has quit [Quit: Αποχώρησε]
Black_Horseman has joined #linux-sunxi
jemk has quit [Ping timeout: 272 seconds]
pfdm has joined #linux-sunxi
paulk-aldrin has quit [Remote host closed the connection]
<Turl> Nyuutwo: pio can handle all pins
<Turl> Nyuutwo: you can use devmem/devmem2 too to read the config
hramrach has quit [Ping timeout: 240 seconds]
<Nyuutwo> Turl: I have read already
<Turl> ok :)
hramrach has joined #linux-sunxi
FDCX has quit [Read error: Connection reset by peer]
FDCX has joined #linux-sunxi
FDCX has quit [Read error: Operation timed out]
tm512 has joined #linux-sunxi
FDCX has joined #linux-sunxi