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
Andy-D has quit [Ping timeout: 245 seconds]
nemunaire has joined #linux-sunxi
<atsampson> TheLinuxBug: I power mine like this: http://stuff.offog.org/powerhack.jpg
<atsampson> I don't think you can power them from the SATA power connector, since there's a transistor between that and the 5V rail that would need to be turned on somehow...
<atsampson> and if you find somewhere that sells them for $15 and will ship to the UK, I'd be very happy ;-)
nemunaire has quit [Client Quit]
<TheLinuxBug> hehe
<TheLinuxBug> it was US shipper
<TheLinuxBug> got it 2 days on amazon actually
<TheLinuxBug> had posted the link here I thought
<TheLinuxBug> ill look for it again
<TheLinuxBug> um
<TheLinuxBug> but thanks for that info
<TheLinuxBug> glad I didn't test it and blow that transistor
<TheLinuxBug> I like your hack though, that may be my method
<TheLinuxBug> using gpio
nemunaire has joined #linux-sunxi
<TheLinuxBug> atsampson: do you have that in 5 and 6?
<TheLinuxBug> or 7 and 8
pekka30 has joined #linux-sunxi
vishnup has joined #linux-sunxi
<TheLinuxBug> atsampson: Thanks for the pointers, got that rigged up already and working ;)
<TheLinuxBug> not pretty, but it will work
clonak has quit [Ping timeout: 272 seconds]
interrobangd has quit [Quit: Leaving]
interrobangd has joined #linux-sunxi
Andy-D has joined #linux-sunxi
leviathancn_szu has joined #linux-sunxi
rgwan-phone has joined #linux-sunxi
<rgwan-phone> Hi all
<rgwan-phone> How about port nand-sunxi FTL layer on MTD layer
<rgwan-phone> It seems that the Allwinner FTL is pretty stable
<rgwan-phone> and we can remove sunxi partition table oit
<rgwan-phone> out
<rgwan-phone> use more proper table like you know, MBR or GBT
<rgwan-phone> GPT
cnxsoft has joined #linux-sunxi
interrobangd has quit [Quit: Leaving]
vishnup has quit [Remote host closed the connection]
leviathancn_szu has quit [Read error: Connection reset by peer]
leviathancn_szu has joined #linux-sunxi
<wens> topi`: perhaps debian or fedora has images?
auxym has quit [Ping timeout: 250 seconds]
auxym has joined #linux-sunxi
Nacho has joined #linux-sunxi
<rgwan-phone> I have ripped out FTL layer in sunxi_nand driver
<rgwan-phone> Do some porting it will work on MTD layer
<rgwan-phone> By add “awftl” partition in mtdparts to realize space assignment of this FTL
iamfrankenstein1 has joined #linux-sunxi
iamfrankenstein1 has quit [Ping timeout: 250 seconds]
auxym has quit [Ping timeout: 246 seconds]
auxym has joined #linux-sunxi
orly_owl has quit [Quit: leaving]
p1u3sch1 has quit [Ping timeout: 260 seconds]
orly_owl has joined #linux-sunxi
p1u3sch1 has joined #linux-sunxi
fredy has joined #linux-sunxi
Ueno_Otoko has joined #linux-sunxi
TheLinuxBug has quit [Ping timeout: 264 seconds]
JohnDoe_71Rus has joined #linux-sunxi
IgorPec has joined #linux-sunxi
hipboi has joined #linux-sunxi
hipboi has quit [Remote host closed the connection]
TheLinuxBug has joined #linux-sunxi
auxym has quit [Ping timeout: 260 seconds]
TheLinuxBug has quit [Ping timeout: 250 seconds]
_stephan has joined #linux-sunxi
reinforce has joined #linux-sunxi
TheLinuxBug has joined #linux-sunxi
OpenBlyat has joined #linux-sunxi
TheLinuxBug has quit [Ping timeout: 264 seconds]
IgorPec has quit [Ping timeout: 246 seconds]
TheLinuxBug has joined #linux-sunxi
TheLinuxBug has quit [Ping timeout: 264 seconds]
naobsd has joined #linux-sunxi
domidumont has joined #linux-sunxi
IgorPecovnik has joined #linux-sunxi
_stephan has quit [Ping timeout: 260 seconds]
TheLinuxBug has joined #linux-sunxi
domidumont has quit [Ping timeout: 246 seconds]
<mripard> rgwan-phone: the AW FTL has some issues of its own
<mripard> and it's never going to be merged in mainline
domidumont has joined #linux-sunxi
IgorPecovnik has quit [Ping timeout: 246 seconds]
Gerwin_J has joined #linux-sunxi
IgorPec has joined #linux-sunxi
<rgwan-phone> mripard: My solution is not directly handle NAND flash
<rgwan-phone> I decide to work like UBI
<rgwan-phone> Which is on MTD layer
<rgwan-phone> but under fs layer
<rgwan-phone> Which will be more generic, not sunxo specific
<rgwan-phone> sunxi
<rgwan-phone> But I dont know they attitude on FTL
<rgwan-phone> Or would they accept a FTL based on MTD
<mripard> we have one already, UBI
<rgwan-phone> but ubi isnt ftl
<rgwan-phone> you cannot use FAT or ext* on it
<rgwan-phone> and sunxi ftl has less issue than ubi
cnxsoft has quit [Ping timeout: 260 seconds]
<rgwan-phone> Basically, my hummingboard A20 board use MLC flash on nand_sunxi driver
<mripard> you know, I'm not telling you this to discourage you from doing this
<rgwan-phone> and I havent seen any data loss here(running for nearly 2years
<mripard> I'm telling this because it has *zero* chance on getting merged
<rgwan-phone> What a pity
cnxsoft has joined #linux-sunxi
<mripard> no, it's not a pity
<mripard> UBI has flaws when it comes to MLC NANDs
<mripard> we should fix UBI, instead of coming with yet another framewor.
<mripard> and it's what we are doing
<rgwan-phone> yeah
<mripard> so if you want to help, go ahead
<rgwan-phone> i know
<rgwan-phone> okay
<mripard> but porting the sunxi FTL directly will be a wasted effort
<rgwan-phone> OK, when I finish my test, I would to like hack on UBI
<rgwan-phone> Thanks
<rgwan-phone> But I still want to build a proper FTL on UBI or somewhat....Because some applications require a proper block device...It sucks...
<mripard> which applications?
<mripard> have you heard of ubiblk ?
<rgwan-phone> Mass Storage
<rgwan-phone> Not yet
<rgwan-phone> I will take a look
<mripard> ah, mass storage...
clonak has joined #linux-sunxi
<mripard> is there still some people that uses that broken thing?
<mripard> ubiblk is not going to help then
<mripard> it's read-only
<rgwan-phone> it seems I have to remake a ftl based on UBI
<mripard> or you use MTP ?
<mripard> there's some reason you can't use a block filesystem on a NAND backed device
<mripard> and an FTL is not going to help
<rgwan-phone> But my host doesnt support MTP
<rgwan-phone> If my host supports MTP, I wont wasting my time on doing this
<mripard> even 15-years old windows support MTP
<rgwan-phone> But it is not computer
<rgwan-phone> It is electronic keyboard
<rgwan-phone> I can”t hack on it”s firmware
<mripard> the block filesystem blocks size and NAND block size are usually waay different, so when it comes to writing to it, either you cache your writes hoping that you'll be able to write a page at some point
<mripard> or you erase a full erase block each and every time you write 512 bytes of data
<mripard> both solutions suck
<rgwan-phone> Yes. I have develop a FTL on STM32 devices
<mripard> none are reliable
<rgwan-phone> I use hybrid-mapping/log block to solve this problem
<rgwan-phone> As for reliable......it is really a suck....
<rgwan-phone> with wear-leveling we can partly solve...but ampifie on write will go as 2~5times high
<rgwan-phone> I really like more neat solution like flash fs
<rgwan-phone> but I have to have compatibility with legacy format
<rgwan-phone> and there are billions of FTL based device like SSDs, MemoryCards and UsbDrives
<rgwan-phone> In their life time, I think it can be considered stable
<rgwan-phone> So I think a FTL is necessary in the kernel
<mripard> you've been warned :)
<rgwan-phone> Thank you for your advice
<rgwan-phone> I would try to work a branch out
<rgwan-phone> and keep synch on mainline
domidumont has quit [Quit: Leaving.]
rgwan-phone has quit [Ping timeout: 260 seconds]
Andy-D has quit [Ping timeout: 260 seconds]
cnxsoft has quit [Remote host closed the connection]
cnxsoft has joined #linux-sunxi
Andy-D has joined #linux-sunxi
yann|work has joined #linux-sunxi
viccuad has joined #linux-sunxi
formruga has joined #linux-sunxi
steev has joined #linux-sunxi
_massi has joined #linux-sunxi
naobsd has quit [Quit: naobsd]
Andy-D has quit [Ping timeout: 260 seconds]
ricardocrudo has joined #linux-sunxi
MaDMaLKaV has quit [Ping timeout: 260 seconds]
MaDMaLKaV has joined #linux-sunxi
MaDMaLKaV has quit [Quit: No Ping reply in 180 seconds.]
MaDMaLKaV has joined #linux-sunxi
enrico_ has joined #linux-sunxi
paulk-collins has joined #linux-sunxi
leviathancn_szu has quit [Ping timeout: 260 seconds]
naobsd has joined #linux-sunxi
leviathancn has joined #linux-sunxi
leviathancn_szu has joined #linux-sunxi
naobsd has quit [Quit: naobsd]
viccuad has quit [Quit: WeeChat 1.3]
viccuad has joined #linux-sunxi
naobsd has joined #linux-sunxi
viccuad has quit [Quit: WeeChat 1.3]
naobsd has quit [Quit: naobsd]
enrico_ has quit [Quit: Bye]
_massi has quit [Remote host closed the connection]
bmeneg has quit [Quit: WeeChat 1.3]
Nacho_ has joined #linux-sunxi
Nacho has quit [Read error: Connection reset by peer]
hipboi has joined #linux-sunxi
Nacho_ has quit [Remote host closed the connection]
Nacho has joined #linux-sunxi
simosx has joined #linux-sunxi
auxym has joined #linux-sunxi
Nacho has quit [Remote host closed the connection]
Nacho has joined #linux-sunxi
afaerber_ has joined #linux-sunxi
naobsd has joined #linux-sunxi
afaerber has quit [Ping timeout: 250 seconds]
auxym has quit [Quit: Lost terminal]
<topi`> I just git cloned linus' tree (4.4.0-rc5) and did "make sunxi_defconfig" and compiled the kernel, booted on my Banana pi, but the USB keyboard is not detected
<topi`> has anyone noticed any regressions in 4.4.0 candidates?
<mripard> it works just fine here
hipboi has quit [Read error: Connection reset by peer]
<mripard> (I'm guessing you're talking about -rc3 ?)
<mripard> did you enable the driver?
leviathancn_szu has quit [Quit: No Ping reply in 180 seconds.]
leviathancn_szu has joined #linux-sunxi
hipboi has joined #linux-sunxi
JohnDoe_71Rus has quit [Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org]
Gerwin_J has quit [Quit: Gerwin_J]
hipboi_ has joined #linux-sunxi
hipboi has quit [Read error: Connection reset by peer]
hipboi has joined #linux-sunxi
hipboi_ has quit [Read error: Connection reset by peer]
IgorPec has quit [Ping timeout: 246 seconds]
rgwan has joined #linux-sunxi
leviathancn_szu has quit [Read error: Connection reset by peer]
leviathancn_szu has joined #linux-sunxi
rgwan has left #linux-sunxi [#linux-sunxi]
OpenBlyat has quit [Quit: Leaving]
leviathancn_szu has quit [Remote host closed the connection]
<topi`> let me double check
<topi`> the keyboard is a microsoft k500 wired keyboard, so I guess pretty much a standard USB HID device
leviathancn_szu has joined #linux-sunxi
<topi`> HID support: Generic HID driver is Y
<topi`> USB support: EHCI HCD support is Y
<topi`> I guess that's it?
<topi`> what is "Generic EHCI driver for a platform device"?
cnxsoft has quit [Quit: cnxsoft]
jjardon has quit []
jjardon has joined #linux-sunxi
<topi`> mripard: yeah, this is 4.4.0-rc3 kernel
<mripard> USB_HID
hipboi_ has joined #linux-sunxi
hipboi has quit [Read error: Connection reset by peer]
steev has quit []
simosx has quit [Quit: Leaving]
steev has joined #linux-sunxi
<wens> topi`: that one you need
<topi`> yeah, it's enabled
<topi`> maybe the wrong dtb?
<topi`> I copied the sun7i-bananapi.dtb to the boot partition's dtb dir
hipboi_ has quit [Remote host closed the connection]
mpmc has joined #linux-sunxi
<topi`> if I can't get the usb keyboard to work, then maybe my best bet is to use a serial console
IgorPecovnik has joined #linux-sunxi
ir2ivps8 has joined #linux-sunxi
whitesn has joined #linux-sunxi
IgorPecovnik2 has joined #linux-sunxi
<whitesn> excuse me, did anyone manage to do analog input from A20?
<whitesn> cubieboard2 specifically
IgorPecovnik has quit [Ping timeout: 246 seconds]
IgorPecovnik2 has quit [Quit: Nettalk6 - www.ntalk.de]
<plaes> GPIO analog input?
<topi`> OK, I was able to get in the bpi via ethernet
rgwan has joined #linux-sunxi
<topi`> lsusb says "unable to initialize libusb: -99"
<topi`> so I guess the USB support is somehow botched??
atx has joined #linux-sunxi
<wens> check dmesg?
<topi`> SMP: total of 1 processors activated.
<topi`> isn't the A20 supposed to be dual-core?
<topi`> there's something wrong with this whole affair
<topi`> does this depend on the dtb?
reinforce has quit [Quit: Leaving.]
<mripard> no, it depends on u-boot for the SMP
hipboi has joined #linux-sunxi
<rgwan> whitesn: LRADC?
Ueno_Otoko has quit [Ping timeout: 246 seconds]
<topi`> mripard: ok, then I guess I need a proper u-boot... which would be the best binary to download?
<topi`> I just took the newest image from bananian.org
<topi`> maybe I did not name the dtb file correctly?
leviathancn has quit [Ping timeout: 260 seconds]
viccuad has joined #linux-sunxi
hipboi has quit [Ping timeout: 260 seconds]
<mripard> I don't know what bananian builds
<mripard> but any recent version from u-boot will do
<mripard> (like, any version released this year)
<topi`> does it matter if I have the Lemaker version of Banana Pi or the other one?
<mripard> yes
<mripard> use mainline u-boot
<topi`> I wonder how can I find out which version of BananaPI this board is?
<topi`> ther'es just a "bpi" logo there which is in the "intel inside" style
<topi`> OK, the Banana Pro is a Lemaker product
<topi`> but clearly this board isn't the Pro
<topi`> damnit. I should have gotten a Cubieboard2 instead :/
<topi`> but a german dealer had these bpi's for a very low price
<plaes> please post pics to wiki
leviathancn has joined #linux-sunxi
<plaes> topi`: does it have A20?
<topi`> yes
<libv> yeah, ndh it, and we can merge the new info with another page if it turns out to be the same
<plaes> yeah.. no such thing currently in wiki
<whitesn> yes sorry was away just now
<whitesn> rgwan: i'm not sure how to use the PIN with adc too, the value will always be either 1 or 0 when I read the value from a pots
<whitesn> I tried PIN 41 (PD7)
<wens> a20 is probably the original bananapi?
<whitesn> PWM works with PB2, and digital read / write also works
<whitesn> with all of the gpio pins
<topi`> what's the best way to recognise the hardware? obviously /proc/cpuinfo only tells you it's an A20 chip...
<whitesn> ram , cpu ?
<topi`> hmm, there is a serial number in cpuinfo as well...
<topi`> whitesn: everything is as on a standard Bananapi-A20
<topi`> except that USB doesn't work, and other things are missing as well
<whitesn> i've never used bananapi, what did you flash?
<topi`> dmesg says: Machine model: LeMaker Banana Pi
<topi`> hmm? /cpus/cpu@0 missing clock-frequency property
<plaes> it only says it because you're using image made by lameaker
<topi`> that might be true
<topi`> maybe I just need an upstream u-boot
<topi`> but do I need to compile/create a SPL as well? or how is that uboot loaded?
<plaes> topi`: try something from here: http://www.armbian.com/banana-pi/
<topi`> OK
<topi`> I guess the "legacy" contains the Mali400 binary blobs, and "Vanilla" is the mainline kernel without binary blobs?
<plaes> yea, legacy is sunxi-3.4
<topi`> GUIs are for weenies
<topi`> I'm not a friend of binary blobs
<topi`> I hope ARM would get their head out of their asses one day and do something useful for the state of the Mali drivers
<whitesn> so no luck with analog input? :/
<topi`> even Nvidia Tegra GPU has decent support in mainline these days. Nvidia!
<plaes> won't happen in near future..
<topi`> oh, this Pecovnik guy has set up this armbian download site...
<topi`> he's been very industrious :) I remember I saw his images back in 2011 already
<rgwan> whitesn: LRADC is not mux with GPIOs
cajg has quit [Read error: Connection reset by peer]
cajg has joined #linux-sunxi
<whitesn> rgwan: what do you mean?
<topi`> I now booted the 3.4 legacy kernel with the same uboot, and now I see two cpu's
<topi`> so this *has* something to do with the dtb
<rgwan> If I don't have wrong memory.... The LRADC pin 0 and 1 is independent pin
<rgwan> is isn't share pin with PD7
<mripard> topi`: no, this has something to do with u-boot
<mripard> so upgrade u-boot
<mripard> and you'll see you'll have to CPUs.
<whitesn> which mapping you basing pin 0 and pin 1?
<topi`> mripard: why, then, does the 3.4 kernel see 2 cpus?
<rgwan> whilesn: basing?
<rgwan> The LRADC pin 0 and pin 1 is not muliplexed with GPIO pin
<rgwan> you should check it out on your board 's manual
<mripard> topi`: because the 3.4 kernel doesn't rely on u-boot to bringup its secondary CPUs
<mripard> while mainline does
<topi`> oh, interesting change
<mripard> rgwan: btw, with configfs at least, you're able to expose a mass storage device backed by a file, and not only by a partition
<rgwan> I has seen this solution before
<rgwan> but it has low performance than directly use FTL on MTD layer
<rgwan> /s/has/have/g
<whitesn> sec..
<whitesn> sorry i'm kind of new to this, so basically the PIN can't be used for GPIO?
<rgwan> This pin can be used for GPIO
<rgwan> but there are NO ADC in there
<mripard> rgwan: I'd be interested to have numbers once you have an implementation
<rgwan> so you can't meter analog signal
<whitesn> that pin41 is LRADC0
<rgwan> mripard: I will done it soon after my final test
<rgwan> :)
<rgwan> whitesn: Yes
<rgwan> Use this pin to meter analog signal
Deskwizard has joined #linux-sunxi
<whitesn> yeah, it only output either 0 or 1 on value
<whitesn> /sys/class/gpio/gpio41_pd7# cat value
<whitesn> 0
<rgwan> GPIO41 isn't LRADC
<rgwan> The pin number of GPIO isn't the pin number of expansion port!
<rgwan> that is the difference
<rgwan> What kernel you 're using now?
<rgwan> sunxi-3.4 or mainline?
<whitesn> 3.4.67+
<whitesn> just now updated to 3.4.79
<ssvb> whitesn: where did you find it?
<whitesn> it's Cubian
<Deskwizard> hey whitesn
<ssvb> whitesn: the kernel version seems to be rather old
<Deskwizard> *bows down* sir ssvb
<whitesn> oh hi deskwizard haha
<rgwan> Yes, the kernel is really old....
<whitesn> I'm on Cubian now, so yeah probably it's not really updated
<whitesn> oh LRADC0 is supposed to be next to 20, but it's not mapped
<whitesn> hi deskwizard :)
<whitesn> how am I supposed to use this
<Deskwizard> whitesn, hehe I guess we got a lot of interests in common :P whats the issue ur having ?
<whitesn> deskwizard: trying to get analog input to cubieboard
<Deskwizard> whitesn, oh okay
<whitesn> bought it last year, only tried to used to pins now after learning arduinos
<whitesn> i'm just dense with these kind of things :)
<rgwan> whitesn: I has forgot the interface in sunxi-3.4 LRADC driver
<rgwan> maybe you should checkout your fex
<Deskwizard> whitesn, hehehe no worries, just be careful with the thing, its not as foolproof as the arduino ;)
<whitesn> any reco for newer images?
<rgwan> Enable the LRADC section
<whitesn> maybe I'll just flash new OS
<rgwan> and access it use /sys/class/ interface or so...
<rgwan> or use mmap to directly access LRADC's register
<plaes> wasn't it max 2V or something?
<Deskwizard> plaes, something like that
<plaes> then better use some external i2c ADC
<whitesn> the usual one ranges from 0~5v?
<rgwan> input range is not a standard
<rgwan> it depends on ADC's design
<rgwan> you should checkout datasheet first
<rgwan> otherwise, it could cause some damage
<Deskwizard> my vote is for i2c adc also
<whitesn> external?
<Deskwizard> theres probably already support for some in the kernel
<Deskwizard> whitesn, yes
<whitesn> hmm I see, I was just curious and wanted to test the pins and hoping to get sensors there
<Deskwizard> I mean, you can, but theres limitations and limits to adhere to
<plaes> you might have burned it
<Deskwizard> like I said, its not like the arduino, with this, if you mess up...
<Deskwizard> what he said
<whitesn> can you eli5 me how is it when specific components takes too "high" voltages or amperes can messes the device?
<Deskwizard> well, it all goes back to design... its like your car engine, if its designed to top at 6500rpm and you run it at 10000, it'll blow
<Deskwizard> components are design for certain specifications, applications
<Deskwizard> short version: I dont know how to explain it :P
<whitesn> there is no way for the device to limit / protect it?
<Deskwizard> not if it is not design to do so
<Deskwizard> nothing is implied, if its not designed in, its not there
<Deskwizard> (besides mistakes but thats another subjects :P)
<plaes> whitesn: you can bring the voltage down
<Deskwizard> ^
iamfrankenstein has quit [Ping timeout: 260 seconds]
<plaes> for example, if your sensor is 5V, you can use the voltage divider to make sure that it never goes over 2V
<plaes> or whatever the maximum was
<Deskwizard> Another thing to point out: the LRADC is only 6 bits iirc
<plaes> yea.. LR == Low Resolution
<Deskwizard> ..... youre kidding me
<Deskwizard> LOL
<plaes> good enough for detecting few buttons
<Deskwizard> always thought it meant ladder resistor ...
<whitesn> so one of the application is using resistor
<Deskwizard> since thats how its used hehehehe
<plaes> hm.. LR might mean Ladder Resistor too.. but I always assumed it was about resolution
<plaes> heh.. Low Rate
<whitesn> I wonder if there are any female to male converter that shaped like U
<whitesn> it's quite hard to use the pins on cubie when it's facing down (I also used SATA port)
iamfrankenstein has joined #linux-sunxi
reinforce has joined #linux-sunxi
nove has joined #linux-sunxi
IgorPecovnik has joined #linux-sunxi
robogoat has quit [Ping timeout: 245 seconds]
robogoat has joined #linux-sunxi
IgorPecovnik has quit [Ping timeout: 246 seconds]
yann|work has quit [Ping timeout: 260 seconds]
robogoat has quit [Ping timeout: 246 seconds]
robogoat has joined #linux-sunxi
Andy-D has joined #linux-sunxi
hipboi has joined #linux-sunxi
afaerber_ is now known as afaerber
<Inode> libv: you were right, ordering tablet parts from china turned out to be a real timesink... an LCD I received isn't even what was advertised.
hipboi has quit [Ping timeout: 246 seconds]
<GeneralStupid> Inode: ordering my orangepi too... in the time i tried to use it with kodi i could have earnd a lot of $
<Inode> GeneralStupid: i'm surprised because this is the first time I've had a serious issue with any component or device that i've puchased through aliexpress.
<GeneralStupid> Inode: i decided to not buy any allwinner stuff again
<Inode> GeneralStupid: isn't ditching a chip manufacturer because of a bad integrator, something like shooting the messenger?
<GeneralStupid> i think the hardware guys should also develop a usable driver
<GeneralStupid> and not just some quick and dirty hacked linux kernel
<Inode> for?
<GeneralStupid> me.
<Inode> no, a usable driver for which device?
<GeneralStupid> Inode: for that whole chip. a usable Mali400 driver, usable video accleration, etc...
<GeneralStupid> the ARM chip in that SoC is well supported
<Inode> i think the onus should be on ARM to provide a proper mali-400 driver
<GeneralStupid> yes maybe they are the guys... But that is capitalism. We buy Allwinner crap, allwinner licences ARM Crap. As long as they get their money they wont change
<Inode> that core is their intellectual property after all
<arnd> nobody today licenses out a working GPU implementation
<Inode> i think you'll find the same problem with rockchip and mediatek
<arnd> Intel, Qualcomm, Nvidia and Broadcom have GPUs with open source drivers, but they keep them to themselves
<GeneralStupid> i think i wont buy any non supported chipsets again.
<arnd> vivante might get there in the future, but they are not very common (Freescale and Marvell mostly)
<GeneralStupid> i only buy intel GPUs, because of their linux support.
<ssvb> GeneralStupid: H3 is a very new chip, you should be patient and/or contribute some code yourself if you care about GNU/Linux
<GeneralStupid> ssvb: the problem starts with Linux, and that binary blob crap you need to install...
<GeneralStupid> and, as i said before, i wasted enough time with that device.
<ssvb> GeneralStupid: I hope that your efforts have been fruitful
<GeneralStupid> ssvb: definetly not
<nove> GeneralStupid: do you know, that the hardware that does the video codec decoding was successful reverse engineering?
<GeneralStupid> nove: is there a link to that?
<nove> GeneralStupid: yes in the wiki
huehner has quit [Ping timeout: 245 seconds]
<nove> GeneralStupid: http://linux-sunxi.org/Cedrus
huehner has joined #linux-sunxi
<GeneralStupid> nove: yes, its nice. But useless due to the fact that only mplayer and mpv is patched...
IgorPec has joined #linux-sunxi
<nove> GeneralStupid: the developers of that popular media player, are boycotting allwinner hardware, and at the same time the work that we are doing
<Deskwizard> nove, I think you might be overstating it considering the qty of "No" in there
<GeneralStupid> nove: i really dont need kodi, but there are a lot of players which also not work
<ssvb> GeneralStupid: mplayer is definitely is not useless, you can surely use it to play movies
<GeneralStupid> flash, html5, metv
<ssvb> GeneralStupid: and if you are allergic to the commend line interface, smplayer is a nice gui frontend
<GeneralStupid> no iam not allergic to command lines
<nove> deskwizard: what about supporting the people that are doing this, so that there could be less "NO"
<Deskwizard> nove, how about you calm the hell down, all I said was that theres a lot of no in there.
<Deskwizard> this attitude wont help with support.
<nove> deskwizard: exactly
huehner has quit [Ping timeout: 260 seconds]
<ssvb> GeneralStupid: btw, mplayer and mpv are not "patched", their vdpau backend just works
<GeneralStupid> it just started with dvb-t maybe i could use it with mplayer, ok. But zattoo (web tv) or youtube wont
<Deskwizard> nove, too bad, I had a high opinion of linux-sunxi people.
<GeneralStupid> deskwizard: i think it is really nice work... They do more then allwinner for example
<ssvb> GeneralStupid: you can check https://github.com/linux-sunxi/libvdpau-sunxi but H3 support is still very new
<GeneralStupid> ssvb: i installed that. but as i said it wont help for flash stuff
<ssvb> GeneralStupid: what is needed for flash?
<GeneralStupid> its unusable without hw accleration
<ssvb> GeneralStupid: what kind of video player do you normally use for it?
<plaes> wat?
<GeneralStupid> ssvb: flash stuff (zattoo, there is a kodi plugin) and dvb-t
<nove> deskwizard: if we would have more support, instead of only "it isn't 100% perfect", we would be much more motived to continue
<nove> motivated*
<GeneralStupid> nove: i think a patched kodi version would really gain the usability
mnr has joined #linux-sunxi
<plaes> how about start fixing up the http://linux-sunxi.org/Kodi page first...
<nove> GeneralStupid: go ask the developers of that software, they are the ones that are in the best position to do this work
huehner has joined #linux-sunxi
<ssvb> plaes: that's a good advice, this Kodi wiki page should just suggest the users to use mplayer/mpv with libvdpau-sunxi
<GeneralStupid> there is an already patched version from zodi
<plaes> then why haven't you added this info to the wiki?
<ssvb> plaes, nove: the way it is now, the Kodi page just seems to be sending a wrong message to the users :)
<plaes> I recently renamed it from xbmc :P
<GeneralStupid> zidoo...
<ssvb> well, anyone googling for "kodi xbmc allwinner" and finding it would just think that hardware accelerated video decoding is not possible on Allwinner devices yet
<nove> GeneralStupid: did you check this forum thread http://forum.kodi.tv/showthread.php?tid=246617&pid=2153555
<GeneralStupid> nove: when i fet that correctly they have problems because of glp violations
MaDMaLKaV has quit [Ping timeout: 260 seconds]
MaDMaLKaV has joined #linux-sunxi
<nove> GeneralStupid: then they could if they wished, use the reversing engendering work
<GeneralStupid> looks for me if vdpau is an interface?
<GeneralStupid> i dont see how an interface could violate a license...
<nove> GeneralStupid: the gpl violations is in the binary blob that allwinner gives
<GeneralStupid> nove: okay, yes, maybe... But that binary blob is just an interface? So it is not shipped with kodi?!
<nove> GeneralStupid: the binary blob is the software that makes the hardware work
<GeneralStupid> nove: but, linux-sunxi reengeneered that?
<nove> GeneralStupid: but as the hardware was already sucessfull reverse engendered, we don't need the blob
<nove> GeneralStupid: yes
<GeneralStupid> i still dont get it, for me its like saying we dont support TCP because there is an existing TCP Stack which is closed source...
<jemk> kodi would "just" have to add support for vdpau with open gles. they already support vdpau, but only with opengl
<jemk> there is a binary driver that doesnt work well and violates gpl, and there is libvdpau-sunxi which is open source
rds_ has joined #linux-sunxi
<rds_> it makes no sense what AW is doing with themselves with this CedarX, it is like they want to hide something that has been RE...
<rds_> they shooting themselves in the foot
<rds_> a good hardware acceleration with the work of Sunxi on mainline
<rds_> their Socs will be so far away ahead of the competition!
<rds_> see the OPI-PC, because it is cheap $15, it is many supports, but SW for it sucks badly and newcomers get upset because their toy does not work
<nove> we can't do everything
<mnr> rds_: Either they don't see that or there might be NDA issues - does anybody know whether the video decoding engine is an allwinner in-house development or a licensed IP block?
<ssvb> rds_: Allwinner is providing SDK for Android, and this Android happens to work on real devices that are on market
<mnr> If it is a licensed IP block, they might actually be legally barred from releasing the docs for it due to an NDA.
<rds_> ssvb: why not for Linux and double their market ?
<ssvb> rds_: what make you think that it would double?
<plaes> :D
<rds_> it seems the marketing at AW is not smart! If they had payed, like $2M - $4M for this group to get the SW mainline, they had been in very good spot
<rds_> to be bought out by Intel :-)
<rds_> also, their roadmap of processors has been weird.... so many families in a short period of time! it is like there are different teams doing different projects....
<ssvb> rds_: the mainline kernel development is expensive and can't compete with a quick in-house hack in terms of "time to market"
<rds_> which are the purpose of H3, H8, A80, A83T, A64 ?
avph has quit [Read error: Connection reset by peer]
<ssvb> rds_: for example, have a look at how the mainline kernel struggles with NAND support and still can't get it done properly
<rds_> I know it is expensive! that is why I keep saying here for ages that should've hired this team ( of this community ) to help them, not spend millions a year with Linaro!
MaDMaLKaV has quit [Ping timeout: 246 seconds]
rds_ has quit [Quit: Page closed]
MaDMaLKaV has joined #linux-sunxi
<GeneralStupid> but on the android devices allwinner has a good market share
<GeneralStupid> because they are cheap... Linux is not their market i guess
MaDMaLKaV has quit [Quit: No Ping reply in 180 seconds.]
<mripard> GeneralStupid: and what kernel android uses ?
MaDMaLKaV has joined #linux-sunxi
<GeneralStupid> 3.4
<GeneralStupid> its the same one
<GeneralStupid> and orangepis SDK is a bad joke
MaDMaLKaV has quit [Read error: Connection reset by peer]
<mripard> my point was it's linux
yann|work has joined #linux-sunxi
<GeneralStupid> Android?
<mripard> if they would invest in the linux development, they would have a return on investment
<mripard> but they don't
<GeneralStupid> Android is really not linux...
<mripard> Android uses the linux kernel
<mripard> it's not GNU/Linux
<Inode> it is, android is a linux kernel with a userland for masochists
<maz_> GeneralStupid: last time I looked, I could definitely spot my code in it.
<mripard> but it definitely runs linux
<GeneralStupid> it is a linux kernel, thats where it ends...
kaspter has quit [Read error: Connection reset by peer]
<maz_> GeneralStupid: that's all I care about.
<GeneralStupid> I compiled it, exactly how i need it... And then?
<GeneralStupid> The way to install a boot image in android is pure pain... Tell me whats the problem with a boot partition?
MaDMaLKaV has joined #linux-sunxi
<TheLinuxBug> "<Inode> it is, android is a linux kernel with a userland for masochists" <- LOL
<GeneralStupid> TheLinuxBug: for me thats no joke
<TheLinuxBug> GeneralStupid: are you looking for instructions on how to build android on sdcard?
<Inode> for me neither, i've had numerous unfortunate tasks of making android installations do things i took for granted with many modern *nix operating systems
<TheLinuxBug> ohh
<TheLinuxBug> sorry misread your comment
<GeneralStupid> TheLinuxBug: i had a working android on my sd card, i just needed WLAN tools
<TheLinuxBug> er
<GeneralStupid> normally its absolutely no problem to cross compile kernel modules...
<TheLinuxBug> you never got the modules compiled?
<TheLinuxBug> just grab kernel source and compile modules
<GeneralStupid> yes i did
ricardocrudo has quit [Remote host closed the connection]
<GeneralStupid> i did
<GeneralStupid> and then, whats then?
<TheLinuxBug> insmod drive.ko
<GeneralStupid> what to do with that modules.
staplr has joined #linux-sunxi
<GeneralStupid> TheLinuxBug: does not work
<TheLinuxBug> ?
<GeneralStupid> i dont have the device running, because it didnt work, i spent around a week after work in trying...
<TheLinuxBug> Well, I know compiling from kernel source on my BPi I did it for changing the port multiplier code and building several wlan modules
<GeneralStupid> He misses modules
<TheLinuxBug> maybe your not cross compiling?
staplr has quit [Remote host closed the connection]
<GeneralStupid> insmod was missing some functions
<TheLinuxBug> thats a first
<TheLinuxBug> should find that article on compiling modules I had to get when I was working on my android sdcard
<Inode> GeneralStupid: toolbox insmod or busybox insmod?
<GeneralStupid> Inode: tried both
<Inode> do you recall the error messages?
MaDMaLKaway has joined #linux-sunxi
MaDMaLKaV has quit [Ping timeout: 246 seconds]
<GeneralStupid> Inode: sorry
<GeneralStupid> i asked it in here a few weeks ago
<GeneralStupid> is there a backlog of this channel?
<Inode> ya
<GeneralStupid> sorry
<GeneralStupid> https://dpaste.de/Cb3f
<Inode> "Snippet you have searched is not available (anymore)."
<GeneralStupid> thats why i said sorry :)
<Inode> heh
<GeneralStupid> and then i built a complete boot image but cant install it
<GeneralStupid> Inode: yes, its awful...
<GeneralStupid> Inode: i think the problem is, that it is a drive which needs mac80211
staplr has joined #linux-sunxi
paulk-collins has quit [Ping timeout: 246 seconds]
paulk-collins has joined #linux-sunxi
MaDMaLKaway has quit [Ping timeout: 260 seconds]
MaDMaLKaV has joined #linux-sunxi
IgorPec2 has joined #linux-sunxi
IgorPec has quit [Ping timeout: 246 seconds]
IgorPec2 has quit [Ping timeout: 246 seconds]
flok420 has joined #linux-sunxi
the193rd has joined #linux-sunxi
premoboss has joined #linux-sunxi
ricardocrudo has joined #linux-sunxi
staplr has quit [Remote host closed the connection]
Andy-D has quit [Ping timeout: 260 seconds]
staplr has joined #linux-sunxi
destroyedlolo has joined #linux-sunxi
<destroyedlolo> hello,
<destroyedlolo> I'm still playing with my tablet ( http://linux-sunxi.org/index.php?title=MID-KLIPAD_HC-913)
<destroyedlolo> But I'm facing some issue with the touch screen :(
Andy-D has joined #linux-sunxi
<destroyedlolo> As per android configuration, the touchscreen uses 'ft5x_ts' : it is loaded and working
<destroyedlolo> I can see it is connected to mouse0 and event1
<destroyedlolo> I installed tslib and configured for /dev/fb0 and /dev/input/event1
<destroyedlolo> but the pointer never move when I try ts_calibrate or ts_test.
<destroyedlolo> What I'm missing ?
<destroyedlolo> no output from cat /dev/input/event1 as well.
soderstrom has joined #linux-sunxi
mnr has quit [Quit: leaving]
yann|work has quit [Ping timeout: 260 seconds]
vaid has joined #linux-sunxi
<vaid> Hello
<vaid> is this live?
<destroyedlolo> I here at least :)
<vaid> nice haha, wait a sec, I'm writing my question, just a sec :)
<vaid> First of all let me present myself: Simply put; I'm a product developer from Sweden and I am originally just a software dev. iOS, Windows, Mac OSX etc.
<vaid> I bought the Olimex A10 4GB about a week ago
<vaid> and I found out pretty quickly that booting Debian from NAND is not possible
staplr has quit [Remote host closed the connection]
<destroyedlolo> I duno if I'll be able to help : I'm here also to ask some help about my touchscreen :) But I will try.
<vaid> however I did manage to get it to work actually, using a No-Kernel image from www.armhf.com
<vaid> now I try to install the wifi drivers but the installation script tells me that it cant find header files of kernel version 3.4.90+
<vaid> I don't know how to install the kernel header files, and whatever I Google keeps indicating that I need to compile the kernel along with the modules and drivers
yann|work has joined #linux-sunxi
<vaid> so my question is: can I compile the kernel found in the sunxi wiki to work with the board I am using? and how do I install it and all of that?
<destroyedlolo> About the header, don't you have a package named linux_header or something like that ?
<vaid> I tried following the documentation found on sunxi-linux on GitHub and I did the Make menuconfig
<vaid> but now I'm stuck. I feel like a chicken in the desert, I don't know what I'm doing haha
<vaid> destroyedlolo: where should I be able to find that package?
<destroyedlolo> If you're able to boot with armhf's one, you need to retreave the config file they used to build the kernel.
<vaid> I did apt-cache search linux-headers or whatever the command is, and it told me that version 2.something.something was available for download
<vaid> I see. but where should I retreive it? on their website it says Wheezy (no kernel)
<destroyedlolo> Or, most easily, first check if your board is known by configure script.
<vaid> configure script of armhf?
<destroyedlolo> Of Sunix.
fl_0 has joined #linux-sunxi
<vaid> ok, where can I see that?
<vaid> I believe it is
<vaid> yep it is
<vaid> however I have the one with 4GB NAND flash on board
<vaid> I have read somewhere that sunxi-linux does not support A10 NAND Booting
reinforce has quit [Quit: Leaving.]
<destroyedlolo> I don't think the size of the NAND maters.
<vaid> no, me neither, but the booting process of sunxi linux is not compatible with the A10 CPU booting process thingidy thing
<destroyedlolo> As long as it has the same CPU, memory and network device, it's likly the same board (conditionnal :) )
<vaid> I did something quite different to boot to NAND actually
<destroyedlolo> Can't you boot from an SD card ?
<vaid> yes, but I need to boot from NAND
<vaid> I will be using the SD slot for something else
<vaid> what I did might inspire to get a proper sunxi to work on the A10 via NAND
<destroyedlolo> for that, I can't help : I'm using only SD card on my BananaPIs and Tablet.
<vaid> oh ok.
<vaid> but, could I do something like this:
<vaid> to just copy the partition of the SD card to the NANDA partition?
<vaid> oops, NANDB, sorry
<vaid> NANDA is used for pre-booting
<vaid> is there any tool for debian that allows to copy one partition to another partition?
<destroyedlolo> NANDA contains kernel and configuration file isn't it ?
<vaid> yes, that's true, I mean NANDB sorry
fredy has quit [Ping timeout: 245 seconds]
<destroyedlolo> So, kernel and bin file in NANDA and rootfs content in NANDB
<vaid> yes, rootfs is in NANDB, can I copy the rootfs content from the SD card to NANDB? If yes, what is the best way to do so?
<destroyedlolo> I'm doing :
<destroyedlolo> 1/ format the partition
<destroyedlolo> 2/ recurcive cp
fredy has joined #linux-sunxi
<destroyedlolo> It's working for my SD cards at least.
<vaid> I tried that yesterday, but it gave me errors on /proc/ and /sys/
<destroyedlolo> Your source was from a running system ?
<vaid> hmmm... I'm not quite sure what you mean, but in terminal I was SU
<vaid> when trying it
<destroyedlolo> I mean you can't copy the '/' of a system in use.
<destroyedlolo> If you do so, you need to exclude /dev /proc /run /sys and all kernel generated FS
<vaid> oh ok
<vaid> so what are my options?
<destroyedlolo> Your source is an SD card ?
<vaid> yes it was, I guess
<vaid> should I boot to NAND first and then connect SD card and do recursive copy?
<destroyedlolo> Put it on a PC, tar and zip your rootfs partition and put the .tgz on the SD
<destroyedlolo> then boot using this SD and extract the .tgz in NANDB partition.
<vaid> aaaah, I see I see
premoboss has quit [Quit: Sto andando via]
<destroyedlolo> Normaly, your kernel will build missing FS on it own.
<vaid> Could you elaborate that last part a little bit?
<vaid> I don't quite understand
<destroyedlolo> I'm running gentoo, /dev /proc /run /sys are empty directory on my source tarball.
<destroyedlolo> When booting, my kernel is mounting tmpfs or devtmpfs or whatever itself and populate them.
<vaid> can I read more about this behaviour?
<destroyedlolo> In Gentoo handbooks (looks for Gentoo's web site).
<destroyedlolo> I guess all modern distribs are doing the same :)
<destroyedlolo> Anyway do you have any experiences with touchscreen ?
<vaid> Alright, I will read.
<vaid> Not really, but ask the question and I'll see what I can do
<destroyedlolo> I resurect a bricked tablet by installing Gentoo (on an SD card :) ).
<destroyedlolo> I loaded ft5x module which was used by Android side ... but I can't get any feedback from the touchscreen :(
<destroyedlolo> And I duno what I can do now :(
<vaid> hav eyou sene this? http://linux-sunxi.org/Touchscreen
<destroyedlolo> Yep
<destroyedlolo> And the FEX file came from Android side, so should work.
<vaid> I have no idea to be honest
<vaid> I'm still new to these things
paulk-collins has quit [Quit: Quitte]
<destroyedlolo> I'm easy with Linux stuffs or basic SUNXI hacking but this touchscreen stuff is new to me :(
<destroyedlolo> as well :(
<vaid> can you maybe restart or reload the module?
rellla has quit [Ping timeout: 260 seconds]
rellla has joined #linux-sunxi
vaid has quit [Quit: Page closed]
<destroyedlolo> It doesn't help unfortunately.
atx has quit [Quit: Leaving.]
soderstrom has quit [Ping timeout: 246 seconds]
andoma_ has joined #linux-sunxi
andoma has quit [Ping timeout: 260 seconds]
naobsd has quit [Remote host closed the connection]
destroyedlolo has quit [Quit: Leaving]