mnemoc changed the topic of #arm-netbook to: EOMA: Embedded Open Modular Architecture - Don't ask to ask. Just ask! - http://elinux.org/Embedded_Open_Modular_Architecture/EOMA-68 - ML arm-netbook@lists.phcomp.co.uk - Logs http://ibot.rikers.org/%23arm-netbook or http://irclog.whitequark.org/arm-netbook/ - http://rhombus-tech.net/
<bsdfox> my mele a2000 even with the stock ICS is really slow.. magnitudes slower than my cell phone on the same OS. Are other people happy with A10 performance? even the ESPN scorecenter app (just scores, no video) has 10 second delays fairly often
tuliom has quit [Quit: Konversation terminated!]
<ZaEarl> bsdfox, how big of a display are you using? I haven't found the stock Androids to be very good.
kaspter has joined #arm-netbook
<bsdfox> 1080p
<bsdfox> I think it's running 1280x720 though
<lundman> no, no 10s delays here
<lundman> nothing plugged in, anot using wifi though
<bsdfox> I'll try wired
<bsdfox> seems a little better
<bsdfox> I was hoping to use this as a light desktop for irc/pdf/browsing and occasional video but it's frustratingly slow considering how little I'm doing
<rz2k> anything in dmesg?
<lundman> strange
avernos has joined #arm-netbook
avernos has quit [Changing host]
avernos has joined #arm-netbook
<bsdfox> do you guys have good luck with linaro?
polto2010 has joined #arm-netbook
itamarjp has joined #arm-netbook
eflatun has joined #arm-netbook
eflatun has quit [Ping timeout: 252 seconds]
ZaEarl has quit [Quit: Ex-Chat]
polto2010 has quit [Quit: Computer has gone to sleep.]
ZaEarl has joined #arm-netbook
Turl has quit [Ping timeout: 245 seconds]
proza has joined #arm-netbook
Turl has joined #arm-netbook
polto2010 has joined #arm-netbook
polto2010 has quit [Ping timeout: 240 seconds]
mSquare has joined #arm-netbook
itamarjp has quit [Ping timeout: 252 seconds]
rsalveti has quit [Ping timeout: 240 seconds]
rsalveti has joined #arm-netbook
rsalveti has quit [Changing host]
rsalveti has joined #arm-netbook
gsilvis_ has quit [Ping timeout: 245 seconds]
gsilvis has joined #arm-netbook
projectgus has quit [Ping timeout: 265 seconds]
rellla has joined #arm-netbook
gimli_ has quit [Remote host closed the connection]
projectgus has joined #arm-netbook
<hno> rz2k, you sould remove the unnumbered IOCTL. Only the _BUF1 and _BUF2 needs implementation.
<hno> But I wonder.. is our framebuffer double-buffered?
<hno> not too familiar with framebuffer implementation.
pawel5870 has joined #arm-netbook
Quarx has joined #arm-netbook
ZaEarl has quit [Quit: Ex-Chat]
<mnemoc> techn: ---^
sspiff has joined #arm-netbook
sspiff has joined #arm-netbook
rellla has quit [Ping timeout: 255 seconds]
rellla has joined #arm-netbook
Quarx|2 has joined #arm-netbook
Quarx has quit [Ping timeout: 245 seconds]
popolon has joined #arm-netbook
avernos has quit [Remote host closed the connection]
jinzo has joined #arm-netbook
Quarx|2 has quit [Ping timeout: 240 seconds]
bsdfox has quit [Ping timeout: 255 seconds]
cat_n9 has quit [Ping timeout: 246 seconds]
ceo16 has joined #arm-netbook
<CIA-15> rhombus-tech: Dino master * rebf6ed68753c /allwinner_a10.mdwn:
Quarx has joined #arm-netbook
rsalveti has quit [Ping timeout: 244 seconds]
ceo16 has quit [Ping timeout: 260 seconds]
lkcl has quit [Ping timeout: 268 seconds]
pawel5870 has quit [Ping timeout: 252 seconds]
mysteryname has joined #arm-netbook
rsalveti has joined #arm-netbook
rsalveti has quit [Changing host]
rsalveti has joined #arm-netbook
cat_n9 has joined #arm-netbook
pawel5870 has joined #arm-netbook
eFfeM_work has quit [Ping timeout: 246 seconds]
lerc has quit [Read error: Connection reset by peer]
<mnemoc> [ 0.000000] BROM Magic: \x06 ..... /me cries and return to $work$ :<
<RaYmAn> ?
eFfeM_work has joined #arm-netbook
<mnemoc> i keep failing to read the BROM
<mnemoc> need to get the cpu-id
<RaYmAn> mmu stuff?
<mnemoc> since 3.7 multiplatform support will be a requirement, that means we need to detect the chip
<mnemoc> I'm doing it after iotable_init() ... so the MMU should be initialized already... I think
<mnemoc> but not even in that case I can find a way to read BROM_BASE+0x04
<mnemoc> sucks
<mnemoc> according to hno's bootinfo I should read "eGON.BRM", not "\x06\0\0\0....."
<RaYmAn> so you re
<RaYmAn> eh
<RaYmAn> hpw do you read it exactly?
lerc has joined #arm-netbook
<mnemoc> tried different Fs for (const char*)F(SW_VA_BROM_BASE+0x04)) and SW_PA_BROM_BASE
<mnemoc> but I should step back and study how the memory access works here
<RaYmAn> yeah, are the page tables actually setup to allow brom access?
<mnemoc> { SW_VA_BROM_BASE, __phys_to_pfn(SW_PA_BROM_BASE), (SZ_64K), MT_MEMORY_ITCM },
<RaYmAn> on tegra, i dumped it through /dev/mem usibg viewmem tool, so might be interesting to get that workibg first?
<mnemoc> E: Unable to locate package viewmem .... and google insists in showing me a windows app... url? :)
eflatun has joined #arm-netbook
<mnemoc> dd if=/dev/mem | hexdump seems evil :p
<mnemoc> it kinds of bothers me how high this address is. 0xffff0000-0xffff7fff
<RaYmAn> wouldnt do the right thing anyways
<mnemoc> true
<mnemoc> will play it that way
<RaYmAn> skumler.net/viewmem
<RaYmAn> not sure where i have source
<mnemoc> Content-Type: text/plain
<mnemoc> RaYmAn: can you rename it to .bin so it doesn't get corrupted?
<RaYmAn> wh
<RaYmAn> wget :/
<RaYmAn> its only silly browsers that corrupt it
<mnemoc> :)
<RaYmAn> on my phone so was hard enough to find link :p
<mnemoc> no rush, I need to $work$ anyway :p
<mnemoc> $ file viewmem
<mnemoc> viewmem: ELF 32-bit LSB executable, ARM, version 1 (SYSV), dynamically linked (uses shared libs), not stripped
<mnemoc> $ ./viewmem
<mnemoc> -bash: ./viewmem: No such file or directory
<RaYmAn> bah
<RaYmAn> dyn linked
<mnemoc> :)
<projectgus> FWIW there's a quick hack I used a few years ago here http://projectgus.com/files/eken/memdump/
<mnemoc> don't worry. i'll continue this tonight anyway
<projectgus> I think the binary there is static linked, it was for android tablets. armel.
<mnemoc> projectgus: thanks
<mnemoc> projectgus: will try the .c
<projectgus> kk
<mnemoc> projectgus: looks good :)
Turl has quit [Ping timeout: 272 seconds]
Turl has joined #arm-netbook
<RaYmAn> yeah, that should work
pawel58701 has joined #arm-netbook
<RaYmAn> the mmap takes phys offsets which is nice
pawel5870 has quit [Ping timeout: 246 seconds]
<mnemoc> ffff0000: ea000008 ea000006 ea000005 ea000004 ....
<RaYmAn> that looks right - e.g. like jumps :P
<mnemoc> bootinfo says it should be u32 jump + u8 magic[8]
<mnemoc> $ sudo ./memdump 0xffff0000 $(expr 1024 \* 32) | grep 4e4f4765
<mnemoc> ffff4000: ea000006 4e4f4765 4d52422e 00000020
<mnemoc> ffff5230: 4e4f4765 3054422e 00000000 e92d5ff0
<mnemoc> ffff5440: eafffff7 4e4f4765 3054422e 00000000
<mnemoc> ffff5d70: eafffff7 4e4f4765 3054422e 00000000
<mnemoc> but... 4 "eGON"s ....
<mnemoc> nice to have command power :)
<mnemoc> command line*
<RaYmAn> so you were just using the wrong address :)
<mnemoc> yup :)
kaspter has quit [Ping timeout: 244 seconds]
lkcl has joined #arm-netbook
<mysteryname> Hi, I have a Mele with a USB to TTL converter, is there anything I need to do to get output from the serial interface using minicom? I know what the boadrate. but do I need to load a certain image for the Mele to output with serial?
rsalveti has quit [Ping timeout: 252 seconds]
tzafrir_laptop has quit [Ping timeout: 252 seconds]
rsalveti has joined #arm-netbook
rsalveti has quit [Changing host]
rsalveti has joined #arm-netbook
<mnemoc> [ 0.000000] BROM Magic: eGON.BRM
<mnemoc> happy happy joy joy
<mnemoc> pr_info("BROM Magic: %.8s\n", (const char*)(SW_VA_BROM_BASE+0x4000+0x04));
<mnemoc> using the physical as with the tool makes it halt... but the VA works fine
<RaYmAn> makes sense :)
<mnemoc> what is that 0x4000 offset?
<RaYmAn> bootrom starts at 0xffff0000, with a few jumps to 0xffff4000 where the actual boot code starts - fel mode is at 0xffff0000 (well, technically, just after all the jumps)
Almamuetya has joined #arm-netbook
<mnemoc> RaYmAn: so the right way would be to "decompile" that jump to find the offset?
<RaYmAn> mnemoc: mm. I severely doubt it changes
<mnemoc> good :)
<RaYmAn> it's too "accurate". I imagine there's lots of zeroes between end of fel code and ffff4000
<RaYmAn> so they probably use a linker script to point it to the right place
merbzt has joined #arm-netbook
<mnemoc> wb merbzt
itamarjp has joined #arm-netbook
merbanan has quit [Ping timeout: 246 seconds]
rsalveti has quit [Ping timeout: 246 seconds]
merbanan has joined #arm-netbook
<cde> mysteryname: as an alternative to minicom, you can try microcom
<mnemoc> the only annoying thing with microcom is that if you don't close before unplugging the dongle it leaves a lock behind and messes up the terminal (reset required)
<mnemoc> i.e. crashes
rsalveti has joined #arm-netbook
rsalveti has quit [Changing host]
rsalveti has joined #arm-netbook
<projectgus> fwiw I'm a fan of plain old gnu screen for serial consoles. 'screen /dev/ttyUSB0 115200'
<mnemoc> scrolling is annoying :p
<andoma> yeah..
<lundman> yeah screen rocks
<andoma> is there a terminal that recovers when suspending/resuming the system?
<projectgus> mnemoc: yeah, I guess I'm pretty used to Ctrl-A Esc by now :)
<lundman> screeeeeen
<andoma> okok .. i'll have to get used to ^A + Esc as well then
tzafrir_laptop has joined #arm-netbook
<mysteryname> cde: thank you
<cde> no pb. it's fairly useful
mysteryname has quit [Ping timeout: 268 seconds]
eflatun has quit [Remote host closed the connection]
eflatun has joined #arm-netbook
ccssnet has quit [Read error: Connection reset by peer]
rm has quit [*.net *.split]
ceo16 has joined #arm-netbook
<mnemoc> RaYmAn: any idea how should I access this before the iomapping?
rm has joined #arm-netbook
<RaYmAn> mnemoc: I don't think you can? I mean, if MMU is enabled, it needs an entry to be able to access it - prior to mmu I'd imagine you can access it with pa?
<mnemoc> halts
<mnemoc> problem is I need to know the cpu-id before doing anything else :<
<RaYmAn> it seems kind of a dirty way to do it :/
<mnemoc> indeed
<RaYmAn> detecting by machineid isn't sufficient? (really - u-boot should set a sane machineid - which should be fine when talking new 3.4+ imo)
<mnemoc> RaYmAn: they want to get rid of the machine id :(
<RaYmAn> oh
<RaYmAn> so kernel is supposed to fully detect stuff+
<RaYmAn> ?
<RaYmAn> No way to do it in u-boot and pass it along? devicetree?
<mnemoc> yes, using DT
<mnemoc> lunch. back in 20
revident has joined #arm-netbook
kaspter has joined #arm-netbook
kaspter has quit [Ping timeout: 252 seconds]
mSquare has quit [Quit: Leaving.]
itamar_ has joined #arm-netbook
hipboi has joined #arm-netbook
<WarheadsSE> DT = "fun"
<WarheadsSE> not sure where they are going to get the price down to $30 (probably subsidized for education) but having access to the hardware.. sweet little device.
<mnemoc> nice
<WarheadsSE> obviously I can't get into a _lot_ of detail, but.. yes, those _are_ USB3 ports.
<mnemoc> at least they are blue :p
ccssnet has joined #arm-netbook
ccssnet has quit [Quit: http://atccss.net]
msb has joined #arm-netbook
bsdfox has joined #arm-netbook
<revident> WarheadsSE, yeah, I saw that yesterday evening. Congrats on product shipping ARCH! Do you know if Globalscale will be opening general purchases like with the DreamPlug?
* WarheadsSE shuts up ;)
<mnemoc> WarheadsSE: in reality how does usb3 performe compared to esata?
<mnemoc> in such device obviusly
ikr_ has joined #arm-netbook
traeak has joined #arm-netbook
<WarheadsSE> well, IIRC it's attached as xhci vis PCIe bus, like the PPV4
<WarheadsSE> The PPV4 has some pretty sound numbers to my WD 3TB USB3 drive
<WarheadsSE> this has a much faster processor, and 8 tims the RAM, so I'd expect to see reasonable usage
<mnemoc> my amd e350 laptop ends on it's knees when there is too much usb2 activity :<
<WarheadsSE> sucky
<L84Supper> 2Ghz single-core Armada 370 SoC, is the 370 a typo or a new unadvertised part by Marvell?
<WarheadsSE> it's a 370
<L84Supper> yeah, I don't see it at the marvell site
<L84Supper> so no data sheets
<WarheadsSE> ;)
<WarheadsSE> 2ghz PJ4 dove relative
<L84Supper> any idea how it varies from the 300 and 310 other than core speed?
<WarheadsSE> .. i have a lot of ideas..
<WarheadsSE> and an NDA
<specing> WarheadsSE: would the NDA still be valid if we threatened to take your life?
<L84Supper> everything with Marvell is NDA, and still they are practically useless as a supplier for anything open
<L84Supper> heh, the 370 is not even on the extranet site, you need special permission to get access beyond the NDA
<WarheadsSE> well I can't get into my extranet atm..
pawel58701 has quit [Ping timeout: 245 seconds]
<WarheadsSE> locked it oopsy
<L84Supper> I wonder how many patent violation suits would be filed if marvell opened their specs
<traeak> need ... to ... reform ... patent ... crap ....
hipboi has left #arm-netbook ["Leaving"]
<L84Supper> the current patent system is a mess, it will be interesting to see what happens with Google, Apple, Samsung etc going all out to battle
<WarheadsSE> got back in.
<WarheadsSE> yeah, PJ4
<WarheadsSE> very similar core architecture to the cubox
<mnemoc> they sued github for patents violations :<
rz2k has quit [Ping timeout: 245 seconds]
rz2k has joined #arm-netbook
rellla has quit [Quit: rellla]
rellla has joined #arm-netbook
rellla has quit [Client Quit]
sspiff has quit [Ping timeout: 252 seconds]
<techn> hno: disp has that kind logic already.. but dunno if it works in every situation/config
<techn> hno: also.. I'm trying to take my tables in dev-use.. I'm using your sunxi u-boot.. and I have problems to get them boot every time
<techn> I don't have serial/jtag cable/card so I don't know where is the problem
<techn> but ie. yesterday I tried to boot a10 tabled about 50 times and it booted 3 times or so
<mnemoc> o.o
<cde> o.O
<techn> So I think I'm giving up of idea to dev top of tablet.. until that problem is resolved :(
<mnemoc> techn: where are you located? I'm not using mine
<hno> techn, PMU management is a known problem with SD booting at the moment. Not investigating any boot issues until that is addressed.
<techn> hno: great that it's WIP
<techn> mnemoc: finland.. but I can wait until PMU problem is resolved
<hno> what you can try if you are not afraid of using livesuite to recover is to dump the built u-boot.bin in nanda, letting the allwinner bootloader load it.
<techn> I can dev top of MiniX until then
<mnemoc> techn: sure?
<hno> techn, what was the "disp has that kind of logic already" response to?
<techn> <hno> But I wonder.. is our framebuffer double-buffered?
eflatun has quit [Ping timeout: 252 seconds]
<hno> ok
<techn> sorry for late response.. I'm coding for money 8-9 hours/day + 2hours travelling time ;)
<mnemoc> cat_n9: do you have your 3.6 tree available somewhere?
eflatun has joined #arm-netbook
<techn> hno: btw. battery boot is boot0, and power outlet boot1?
<techn> should they have same boot progress?
<RaYmAn> no, it's boot0 -> boot1 -> u-boot
<mnemoc> techn: two stages of the same boot
<mnemoc> techn: you may want to read http://rhombus-tech.net/allwinner_a10/a10_boot_process/ .... and then dump the new knowledge in our wiki :)
<techn> I'm just wondering how those boot's differ in eatch other.. battery/outlet
<mnemoc> techn: that is hidden by the PMU
eflatun_ has joined #arm-netbook
TomNL has joined #arm-netbook
<TomNL> hey guys, did anyone try to compile xbmc?
eflatun has quit [Ping timeout: 244 seconds]
eflatun_ has quit [Ping timeout: 244 seconds]
eflatun_ has joined #arm-netbook
<mnemoc> TomNL: empat0's?
pawel5870 has joined #arm-netbook
<TomNL> yes
<ceo16> TomNL: I'm tring but is difficult to find informations
<TomNL> hehe i also wanted to start trying...seems he is using ubuntu 12.04 rootfs
<ceo16> I'm compiling in lubuntu 12.04 but I need to undestand if cross-compiling or compiling directly from mele
eflatun__ has joined #arm-netbook
<TomNL> are you running lubuntu on mele or mk802?
popolon has quit [Quit: Quitte]
<ceo16> mele
<TomNL> which image did you use?
eflatun_ has quit [Ping timeout: 244 seconds]
<ceo16> this
<ceo16> with kernel 3.0.39+
* mnemoc wonders what kernel tree do they use
<TomNL> so i just dd this image to an SD?
<TomNL> dont need to change anything for mele?
<mnemoc> TomNL: the script.bin for sure
popolon has joined #arm-netbook
<ceo16> yes in sd
<mnemoc> but making your own SD is trivial....
eflatun__ has quit [Ping timeout: 244 seconds]
eflatun__ has joined #arm-netbook
popolon has quit [Client Quit]
ikr_ has quit [Ping timeout: 244 seconds]
pawel5870 has quit [Ping timeout: 256 seconds]
<TomNL> well i guess it's good to have own rootfs which you can copy to SD after compiling xbmc
<TomNL> mnemoc: is there some howto for setting up the cross compiler?
<mnemoc> TomNL: i use linaro's ppa on ubuntu.... works out of the box
gimli_ has joined #arm-netbook
<mnemoc> TomNL: options depends on what you run
<TomNL> ok i will check
<mnemoc> for ubuntu hosts
Quarx has quit []
<TomNL> thnx
<mnemoc> the other very popular toolchain is http://www.codesourcery.com/
<TomNL> +TOOLCHAIN=/usr/arm-linux-gnueabi
gimli_ has quit [Read error: Operation timed out]
<TomNL> this whats empat0 was using, this is in the ppa right?
ZaEarl has joined #arm-netbook
<mnemoc> no, that's probably codesourcery's
<mnemoc> instead of setting a path where you dumped the toolchain people usually sets the CROSS_COMPILE prefix
<mnemoc> and Makefiles do e.g. CC=$(CROSS_COMPILE)gcc
rsalveti has quit [Ping timeout: 252 seconds]
<mnemoc> and if one has the toolchain in a separated dir, one adjusts $PATH accordingly
<techn> Is there any real improvements when using those custom crosscompilers? other than bit faster binary?
<techn> I'm using ubuntus own and it seems to work ok
<mnemoc> no... the benefit is to be able to compile for arm in your x86 host
rsalveti has joined #arm-netbook
rsalveti has quit [Changing host]
rsalveti has joined #arm-netbook
eflatun has joined #arm-netbook
<mnemoc> techn: ubuntu's toolchain is linaro's toolchain
sspiff has joined #arm-netbook
sspiff has quit [Changing host]
sspiff has joined #arm-netbook
gimli_ has joined #arm-netbook
<techn> Source: gcc-4.4-armhf-cross (1.52)
<techn> Version: 4.4.7-1ubuntu2cross1.52
<mnemoc> 4.6??
<mnemoc> err
pawel5870 has joined #arm-netbook
<techn> Maintainer: Ubuntu Core developers <ubuntu-devel-discuss@lists.ubuntu.com>
<TomNL> mmm thats also seems to be arm hard float..but it seems cedar libs are not available hard float
<WarheadsSE> classic.
<techn> same for 4.6
<mnemoc> TomNL: right. at the moment we don't have cedarx libs for armhf
eflatun__ has quit [Ping timeout: 244 seconds]
<mnemoc> gcc --version
<mnemoc> gcc (Ubuntu/Linaro 4.6.3-1ubuntu5) 4.6.3
<mnemoc> arm-linux-gnueabi-gcc --version
<mnemoc> arm-linux-gnueabi-gcc (Ubuntu/Linaro 4.6.3-1ubuntu5) 4.6.3
<WarheadsSE> well considering that ubuntu/linaro followed arch linux arm in going full hard float ....
<WarheadsSE> you arent going to find much in the way of major distros using soft armv7
<techn> mnemoc: seems to be same for me
<mnemoc> WarheadsSE: that's where the r-pi becomes harmful :<
<WarheadsSE> our rpi has hf now, but we're continuing soft for it
<mnemoc> TomNL: you have to install an older version to have armel
<WarheadsSE> we had a v7s for about ~ 1 month, then went pure hf
<WarheadsSE> blech, the "armel" is armv4.
<WarheadsSE> easier to just built out ct-ng for soft.
<TomNL> allwinner should provide armhf libs
<mnemoc> the problem is the cedarx libs are "native armel"
<mnemoc> TomNL: hipboi is trying
<WarheadsSE> well what is "armel" to them
<mnemoc> whatever the "ubuntu armel" does by default :p
TomNL has quit [Quit: Bersirc 2.2: Looks, feels and sounds (?!) different! [ http://www.bersirc.org/ - Open Source IRC ]]
<WarheadsSE> that comes from debian, which is armv4
<WarheadsSE> which is massively off from what the chip capabilities are.
<mnemoc> yup
bsdfox has quit [Ping timeout: 240 seconds]
* WarheadsSE shakes head ..
* mnemoc wonders if they good accept to let one of us get the code just to compile it correctly...
<mnemoc> s/good/would/
<WarheadsSE> probably not, even with NDAs
<mnemoc> ibot: ?
eflatun has quit [Remote host closed the connection]
eflatun has joined #arm-netbook
<mnemoc> techn: I told you I can mail you mine
<RaYmAn> mnemoc: bubble wrap envelope! :P
<mnemoc> RaYmAn: Indeed :'(
<techn> yeah.. after thinking I need to get some input to a13 tablet.. since touch and usb is not working :(
<techn> but I rather buy one
<mnemoc> techn: mail hipboi
<RaYmAn> the metal holder is just glued on and seperates easily, given you easy access to all the pins
eflatun has quit [Read error: Connection reset by peer]
projectgus has quit [Ping timeout: 265 seconds]
<mnemoc> i don't find those worthy
eflatun has joined #arm-netbook
<RaYmAn> the dx one is nice because it doesn't require soldering if you just have some female jumper cables
<mnemoc> overly expensive, absurd shipping cost and you still have to solder the pins
<RaYmAn> :)
<RaYmAn> allwinner/tom's is of course much nicer, but it's a bit harder to get hold of
<mnemoc> techn: mail hipboi. he said he can get them on demand
<techn> mnemoc: thanks for the ti
<techn> p
* RaYmAn still ponders if trying to convince olimex to make them would be worthwhile
<mnemoc> tsvetan is not very open to suggestions...
zowtar has joined #arm-netbook
<RaYmAn> still worth a try
<mnemoc> sure
projectgus has joined #arm-netbook
<specing> < mnemoc> they sued github for patents violations :<
<specing> Who? What? Why?
<WarheadsSE> someone hasnt been reading the news
<mnemoc> specing: a troll obviusly. https://lwn.net/Articles/516736/
gimli__ has joined #arm-netbook
<WarheadsSE> mnemoc: ever glanced through that?
eflatun has quit [Remote host closed the connection]
eflatun has joined #arm-netbook
<WarheadsSE> dedupe, distributed file systems with unique ids based on content, hosted files, processing distributed files
<WarheadsSE> the fuck
dstyle has joined #arm-netbook
<WarheadsSE> need to have that shit shot down
gimli_ has quit [Ping timeout: 244 seconds]
<mnemoc> WarheadsSE: i try to avoid reading the patents... it really mads me the crap people gets patented
<ZaEarl> which is exactly why the patent system is busted
<ZaEarl> what good is "public disclosure" if no one reads them?
RITRedbeard_ has joined #arm-netbook
<ZaEarl> all they are now are weapons to beat down competitors
<dstyle> Good day, with so much wonderful things happening (empat0's XBMC and mail3D) its time for me to set up a build enviroment. Could someone tell me which gcc crosscompiler is known to work most flawless for Allwinner A10? I saw some different cross compilers and would like to run safe from the start.
<mnemoc> at best. many trolls do it as their business model
<ZaEarl> dstyle, I use whatever is built in to ubuntu 10.04
pawel5870 has quit [Ping timeout: 264 seconds]
<ZaEarl> Here's how to setup for Android building: http://linux-sunxi.org/Building_CyanogenMod_for_ZaTab
eflatun has quit [Remote host closed the connection]
eflatun has joined #arm-netbook
<dstyle> ZaEarl, thank you i give it a try
<dstyle> I have installed 12.04 so i hope the 12.04's gcc will produce also stable code
<mnemoc> be sure to set CROSS_COMPILE to arm-linux-gnueabi- in that case
tzafrir_laptop has quit [Ping timeout: 244 seconds]
RITRedbeard_ has quit [Ping timeout: 245 seconds]
<ZaEarl> Has anyone asked Allwinner to release sources to CedarX?
<RaYmAn> why would they?
<ZaEarl> Well, it's the opinion of some GPL experts that the shimming they're doing is a GPL violation.
itamarjp has quit [Ping timeout: 240 seconds]
<RaYmAn> practically everything they do is a GPL violation though.
itamar_ has quit [Ping timeout: 244 seconds]
<ZaEarl> sure, but they have released GPL sources for a lot of stuff
<RaYmAn> after it was leaked by a customer, wasn't it? :P
<WarheadsSE> dstyle: another option http://archlinuxarm.org/developers/distcc-cross-compiling (see ARMv7l hard)
<RaYmAn> also, the kernel parts are open aren't they? So what is the violation? Userspace access through ioctl's are pretty obviously tolerated
<RaYmAn> (I genuinely would like to know the answer ! :P)
gimli__ is now known as gimli
<ZaEarl> well, the kernel won't boot without the userspace, so it's definitely a borderline case
<mnemoc> RaYmAn: in the A13-SDK cedarx driver comes as .ko :|
<mnemoc> RaYmAn: and mach-sun5i as built-in.o
pawel5870 has joined #arm-netbook
<ZaEarl> since the open kernel sources are nothing more than a shim, they'll never be accepted into mainline
<RaYmAn> mnemoc: sure, that's definitely violating
<dstyle> WarheadsSE, i give it also a try. 10x!
<mnemoc> ZaEarl: we have complete GPL sources, out of sync with their tree, but complete and GPL
<RaYmAn> ZaEarl: isn't that mostly the same as e.g. graphics drivers?
<RaYmAn> or am I missing some subtle point? :)
<ZaEarl> RaYmAn, probably the same
<ZaEarl> mnemoc, define "complete"
eflatun has quit [Ping timeout: 244 seconds]
<mnemoc> ZaEarl: beside some i2c ts drivers we have functional code for everything in A10 and A13
<ZaEarl> but what about CedarX?
<mnemoc> kernel side, we have sources for both
<mnemoc> and code was properly GPLed (after leak)
<ZaEarl> but the kernel side is simply a shim
<mnemoc> just like every other GPU/VPU in the ARM world
<ZaEarl> agreed
<mnemoc> using mmap/ioctl kernel interfaces from closed source userspace is perfectly legal
<ZaEarl> but "because everyone else is doing it" isn't a very good legal defense
<RaYmAn> it's kind of a grey area I think - no one has ever pursued it in court and it's just accepted
<mnemoc> it is 100% legal
<mnemoc> or it would be imposible to have ANY closed app in linux
<RaYmAn> only ones communicating with kernel :P
<mnemoc> :)
<RaYmAn> glibc is LGPL, isn't it?
<RaYmAn> well, obviously
<ZaEarl> apps are different than kernel drivers
<RaYmAn> it's quite hard to draw a line
<mnemoc> ZaEarl: sure, and kernel drivers are GPL
<mnemoc> and open source
<ZaEarl> will mainline accept them?
<RaYmAn> how is that relevant to whether they break GPL?
<mnemoc> legally, yes. by form, no
<mnemoc> the code is still too ugly to try mainlining
<RaYmAn> I severely doubt you can GPL something as simple as headers required for doing IOCTL's
<ZaEarl> because the GPL isn't clear, other than saying "derivitive works" must be open souced
<RaYmAn> ZaEarl: the problem is, "everyone else is doing it" is perfectly valid excuse until someone takes it to court.
<ZaEarl> and a userspace driver that has no use outside of shimming to the kernel is quite easily argued to be a derivitive
* WarheadsSE walks away
<mnemoc> userspace libs aren't "derivative work" of kernel drivers
<mnemoc> unless they include code from them
<ZaEarl> mnemoc, i'd like to believe you
<RaYmAn> mnemoc: it hasn't really been determined whether that's the case tbh - not as far as I know anyways
<RaYmAn> I'd say it's common belief that you are right given no one has pursued it.
<mnemoc> RaYmAn: you can't have userspace code that doesn't talk in one way or the other with the kernel
<RaYmAn> I agree that it would completely undermine the GPL and make it unuseable, if it spread like that, but that doesn't mean it can't be the case.
<ZaEarl> that's where "derivative" comes into play. apps clearly aren't derivatives.
<ZaEarl> drivers are questionable
<RaYmAn> stuff communication through /dev nodes aren't derivative either
<mnemoc> RaYmAn: what's the difference between sending an ioctrl to a GPL kernel or opening a file created by a GPL app or talking over tcp/ip with a GPL server?
<ZaEarl> maybe if we can get the kernel to boot without the userspace, that might be clear.
<RaYmAn> mnemoc: I'm not a lawyer :P
<mnemoc> :)
<RaYmAn> wait - why wouldn't kernel boot without cedarx userspace?
<ZaEarl> because the kernel driver requires it
<mnemoc> eh?
<mnemoc> nothing in the kernel requires anything in userspace
<RaYmAn> ^^
<RaYmAn> I think you might be talking about android requiring it?
<mnemoc> android video stuff isn't GPL
<ZaEarl> Try removing libCedarX.so and see if it boots
<RaYmAn> kernel boots just fine.
<mnemoc> neither is their libc
<RaYmAn> Android doesn't.
<ZaEarl> yes, I'm only familiar with android
<RaYmAn> that's entirely different from kernel requiring it
<mnemoc> ZaEarl: you are confusing what "boot" means
<RaYmAn> if you build android without using cedarx, I'm pretty damn sure you will be able to boot without libcedarx
<mnemoc> and you'll have `adb shell` to show you the OS is actually running
<RaYmAn> mnemoc: that's not always the case though
<mnemoc> the ld failure to find cedarx.so will break adb?
<RaYmAn> tegra devices can get stuck at bootloader screen, while in fact the system is booting sufficiently to mount /system and just hang on doing the boot animation - with no adb available
<RaYmAn> it might - it's hard to say for sure without tring :P
<RaYmAn> trying*
<mnemoc> yuck
<RaYmAn> android boot process is...special :P
<techn> mnemoc: I think adb activation could need some userspace magic to enable debug-mode?
<RaYmAn> it's possible to work around - point was that you can't use it directly as "proof" it boots :)
<RaYmAn> (that it's not there is not proof that it's not booting I mean)
proza has quit [Ping timeout: 244 seconds]
eflatun has joined #arm-netbook
eflatun has quit [Read error: Connection reset by peer]
<ZaEarl> There's a nice clarification in COPYING that says if you use "normal system calls" that's ok for user programs.
<ZaEarl> does cedarx use "normal system calls"?
<mnemoc> yes
eflatun has joined #arm-netbook
<mnemoc> closed source userspace drivers use ioctl() and mmap() to access hardware, but they are normal system calls used by most normal apps and normal libs too
eflatun_ has joined #arm-netbook
pawel5870 has quit [Quit: Leaving.]
eflatun has quit [Ping timeout: 244 seconds]
rellla has joined #arm-netbook
<L84Supper> anyone know a of good USB A/V capture dongle?
eflatun_ has quit [Remote host closed the connection]
eflatun_ has joined #arm-netbook
tzafrir_laptop has joined #arm-netbook
<rm> don't know any good ones, I only tried "EasyCAP" (which you will see offered a lot), and the quality was quite crappy
<rm> much worse than on a PCI TV tuner
eflatun_ has quit [Remote host closed the connection]
eflatun_ has joined #arm-netbook
eflatun_ has quit [Ping timeout: 244 seconds]
eflatun_ has joined #arm-netbook
rellla has quit [Quit: rellla]
gimli has quit [Quit: Verlassend]
eflatun__ has joined #arm-netbook
eflatun_ has quit [Ping timeout: 245 seconds]
eflatun has joined #arm-netbook
eflatun__ has quit [Ping timeout: 244 seconds]
Almamuetya has quit [Ping timeout: 252 seconds]
xxiao1 has quit [Quit: WeeChat 0.3.7]
xxiao has joined #arm-netbook
eflatun has quit [Remote host closed the connection]
eflatun has joined #arm-netbook
eflatun_ has joined #arm-netbook
eflatun has quit [Ping timeout: 244 seconds]
eflatun_ has quit [Remote host closed the connection]
eflatun_ has joined #arm-netbook
revident has quit [Quit: Combustible lemons? Bah, I bring you weaponized asparagus!]
ceo16 has quit [Ping timeout: 240 seconds]
eflatun_ has quit [Ping timeout: 244 seconds]
eflatun_ has joined #arm-netbook
eflatun_ has quit [Ping timeout: 244 seconds]
dstyle has quit [Ping timeout: 245 seconds]
vgrade2 has quit [Ping timeout: 245 seconds]
popolon has joined #arm-netbook
marcan has quit [Read error: Operation timed out]
tuliom has joined #arm-netbook
Turl has quit [Ping timeout: 245 seconds]
bsdfox has joined #arm-netbook
bsdfox has quit [Changing host]
bsdfox has joined #arm-netbook
popolon has quit [Quit: Quitte]