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/
lkcl has joined #arm-netbook
klys has joined #arm-netbook
<klys> ok I am trying to unbrick my tablet. it I think is doing fsck on / and so it doesn't ever start the "loading..." sequence. the question I have is there used to be this file online called debin-wm8650-0.5.img.7z; does anyone here have this file?
<klys> *debian-wm8650-0.5.img.gz
freakazoid0223 has joined #arm-netbook
<froese> klys: I don't know that file. but could you tell me what kernel you are using with your wm8650?
XenGi_ is now known as XenGi
AC`97 has quit [Ping timeout: 252 seconds]
revident has joined #arm-netbook
rsalveti has quit [Ping timeout: 260 seconds]
rsalveti has joined #arm-netbook
<klys> froese: just the kernel that came with it, 2.6.26 for arm
<froese> klys: ah, ok. a vendor kernel then, i suppose, not a vanilla kernel.
<klys> so I found this debian_wheezy_3.0.36.img.gz and am going to try that
AC`97 has joined #arm-netbook
<klys> I did gunzip debian_wheezy_3.0.36.img.gz > /dev/sde; // all I got was an empty FAT32 filesystem.
<buZz> eh yeah that will not work
<buZz> gunzip debian_wheezybla.gz ; dd if=debian_wheezybla.img of=/dev/sde
<klys> ok
<buZz> wm8650? this sounds familiar ..
<buZz> hmm
<buZz> i THINK i have a 7" winCE netbook with this arch at the hackerspace
<buZz> oh
<buZz> no
<buZz> a mate of mine has a 7" tablet with it :P
<klys> ok
<buZz> ok
<klys> oh gees, now I have to download the image again.
<klys> there was one other thing about mounting the tablet's filesystem. It seemed like the filesystem type was affs or something like that, not ext2
<klys> I don't remember the name of it for sure.
<buZz> jffs2?
<klys> thanks
AC`97 has quit [Ping timeout: 260 seconds]
stefanro has joined #arm-netbook
stefanro1 has quit [Ping timeout: 255 seconds]
<froese> jffs2 is only used and raw flash devices. /dev/sde does not sound like one :-)
<klys> well it's in a usb plug
<froese> i have a netbook here with a wm8850.
<froese> usb-drives can run any fs.
<klys> it's a microsd card in a usb plug
<froese> the same.
<klys> so I have to load it with a bootable image with uboot on it.
<froese> eh, no. the u-boot is built-in. it loads the kernel from usb/sd/whatever.
<klys> do you think it could be farking out befor it ever gets to the sd card?
<klys> *before
<klys> it might be
<froese> possible.
<klys> because that's when it's supposed to say "loading..."
<froese> i don't know what it tries to load.
<klys> yeah, I have to figure that out I guess.
<froese> it boots android when the usb-drive is removed?
<klys> well, when the microsd card is removed it's supposed to boot android
<froese> and on the sdcard is a small fat-fs and the debian root-fs (ext3/4/whatever)?
AC`97 has joined #arm-netbook
<froese> but android isn't booting either?
<klys> yes, both are true.
<klys> I don't know whether it has the debian root fs.
<klys> it doesn't yet.
<froese> the wm8xxx u-boots usually look on removable media for a u-boot script (called either wmt_scriptcmd or FirmwareInstall/autorun.1.wmt)
<froese> if i can't find such a script, it tries boot from nand (the internal flash).
<klys> OK this might work
<froese> s/if i/if it/
<ibot> froese meant: if it can't find such a script, it tries boot from nand (the internal flash).
<klys> I might have to get a copy of that script.
<froese> be carefull with those scripts. there are some flying around that modify device parameters and/or install a new u-boot. those are highly device specific.
<froese> i assume you have no serial connection to the device?
<klys> no, this thing didn't exactly come with a serial connector.
<klys> I'm still writing to the sd card.
<froese> noone does ;-)
<froese> usually 3 or 4 pads on the board...
<klys> I can't really concur with your sympathy seeing there is no way to obtain a serial connector.
<klys> we'll see if this works.
<froese> well, most people hacking around with arm devices soldier an serial cable to their board exactly because it's really hard to fix any boot issues without one.
<klys> when it boots, the green android appears on the screen, and starts checking something. it never completes.
<froese> imo your best bet would be to try to find an original android installer/updater image for your device. that may fix any issues with bad env-parameters, restore a working kernel and filesystem. then you can start again with installing debian on it.
<klys> I'm just trying to get back to the filesystem that was there.
<froese> but if it isn't even booting android ...
<klys> I don't think I installed the sysvinit deb before it crashed, though there is a remote possibility.
<klys> It isn't compatible with arm debs, just armel
<froese> usually debian checks the architecture. you won't even see it if it isn't available for your architecture.
<klys> the kernel already boots into the required architecture.
penguin42 has quit [Quit: Leaving.]
<klys> It's done writing the SD, now to try it.
<froese> these android-strick do have an ethernet, didn't they?
<froese> s/strick/sticks/
<ibot> froese meant: these android-sticks do have an ethernet, didn't they?
<froese> oh no, only hdmi on one side and usb on the other.
<nulluser> Well, if you have an xray machine and a really small drill you might be able to connect to the pads ;)
<klys> should I just rename /script.bin to /wmt_scriptcmd ?
<klys> there is also /boot.scr
<froese> careful! look into the script first! (ignore the first 72 bytes binary junk)
<klys> I will just rename /boot.scr to /wmt_scriptcmd
<froese> afaik, wmt_scriptcmd is used only on older devices.
<klys> how old? 2011?
<froese> boot.scr may be a boot screen.
<froese> how large are the files...
<klys> 475 boot.scr 43148 script.bin 3951200 uImage
hg_5 has quit [Ping timeout: 265 seconds]
<froese> the smallest looks like a script. the 43k one is strange. the uImage is clear.
<froese> just look into boot.scr. it has a 72 bytes binary header, the rest is ascii.
<froese> (but don't change it, the binary header contains a checksum)
<froese> and if you see the command "saveenv" in the script DON'T start it.
XenGi is now known as XenGi_
<klys> I opened it with emacs. the file doesn't look like the format you described at all.
<klys> ok, I'm back out of demand with this thing. I don't know what to do.
<klys> does uboot have a blocklist of the file?
freakazoid0223 has quit [Quit: Leaving]
<froese> no. it has a simple (and buggy) vfat driver built-in.
<klys> because if it doesn't I could move it into /Firmwareinstall
<klys> but it's not a script
<klys> I have LOST.DIR boot.scr evb.bin evb_1.bin script.bin uImage
<klys> correction, boot.scr is a script like you described. (script.bin is not.)
<froese> u-boot files start with the magic number 27 05 19 56. if the file doesn't start with that, u-boot will not run it.
<froese> try "file /mnt/sdcard/boot.scr"
<klys> OK I renamed /boot.scr to /wmt_scriptcmd and the tablet is behaving differently.
<klys> the green man and the kernel versions stay on the screen.
<klys> ...and the green man doesn't start moving around at all.
<klys> Does this uBoot have a blocklist of uImage?
<froese> no. it has a simple (and buggy) vfat driver built-in.
<klys> cool. I might just replace uImage and see if I can get it to find the screen.
popolon has quit [Quit: Quitte]
<klys> I have setenv setargs 'setenv bootargs console=${console} root=${root} loglevel=${logl\
<klys> evel} ${panicarg} ${extra}'
<klys> from the script, is ${console} /dev/ttyS0 ?
<klys> I want it to be /dev/tty1
<froese> where did you find that line?
<klys> boot.scr was renamed and it affected it. the line is from this file.
<froese> ah, it's a script after all. just wondering because you said it isn't.
<froese> look for a "setenv console xxxxx". usually it's ttyS0,115200n8
<klys> okay, how do I update the checksum?
<froese> but does your kernel even has a framebuffer console build in? else you won't see tty1 output.
<froese> with mkimage
<froese> but it doesn't "correct" the checksum. it takes a file and adds a correct header to it. so you first have to strip the current header.
<klys> is arch armel?
<froese> just arm is ok for scripts.
<klys> entry point?
<froese> 0 0
<froese> -A arm -O linux -C none -T script -a 0 -e 0 -n Boot-Script -d boot-scr.txt boot.scr
<froese> "mkimage -l boot.scr" shows the settings of the original file (as does "file boot.scr")
freakazoid0223 has joined #arm-netbook
<klys> ok now to build the kernel.
froese has quit [Ping timeout: 264 seconds]
lerc has quit [Read error: Connection reset by peer]
XenGi_ is now known as XenGi
gsilvis has quit [Read error: Connection reset by peer]
XenGi is now known as XenGi_
lerc has joined #arm-netbook
gsilvis has joined #arm-netbook
bsdfox has joined #arm-netbook
XenGi_ is now known as XenGi
bsdfox has quit [Changing host]
bsdfox has joined #arm-netbook
XenGi is now known as XenGi_
revident has quit [Quit: Combustible lemons? Bah, I bring you weaponized asparagus!]
ZaEarl has joined #arm-netbook
mikey_w has quit [Ping timeout: 260 seconds]
mikey_w has joined #arm-netbook
mikey_w has quit [Ping timeout: 260 seconds]
aholler_ has joined #arm-netbook
mikey_w has joined #arm-netbook
aholler has quit [Ping timeout: 244 seconds]
ZaEarl has quit [Ping timeout: 250 seconds]
Jef91 has joined #arm-netbook
bsdfox has quit [Ping timeout: 265 seconds]
Skaag has quit [Quit: All those moments will be lost in time, like tears in the rain]
mikey_w has quit [Ping timeout: 244 seconds]
bsdfox has joined #arm-netbook
mikey_w has joined #arm-netbook
Quarx has joined #arm-netbook
Jef91 has quit [Remote host closed the connection]
Jef91 has joined #arm-netbook
Jef91 has quit [Remote host closed the connection]
Jef91 has joined #arm-netbook
gimli has joined #arm-netbook
servili007_ has joined #arm-netbook
servili007 has quit [Ping timeout: 245 seconds]
Jef91 has quit [Quit: Instantbird 1.3 -- http://www.instantbird.com]
rellla has joined #arm-netbook
ganbold_ has joined #arm-netbook
HSt_ has joined #arm-netbook
servili007_ has quit [Read error: Connection reset by peer]
eebrah has quit [Ping timeout: 265 seconds]
eFfeM has joined #arm-netbook
ganbold_ has quit [Ping timeout: 255 seconds]
ganbold_ has joined #arm-netbook
lerc has quit [Ping timeout: 260 seconds]
eebrah has joined #arm-netbook
stefanro has quit [Quit: Leaving.]
stefanro has joined #arm-netbook
eebrah has quit [Ping timeout: 265 seconds]
von_fritz has joined #arm-netbook
eebrah has joined #arm-netbook
rellla has quit [Ping timeout: 244 seconds]
fredy has quit [Ping timeout: 260 seconds]
Guest81011 has joined #arm-netbook
eebrah has quit [Quit: ChatZilla 0.9.89 [Firefox 17.0.1/20121129152039]]
Quarx has quit []
hg_5 has joined #arm-netbook
eebrah has joined #arm-netbook
eebrah has left #arm-netbook [#arm-netbook]
popolon has joined #arm-netbook
ganbold__ has joined #arm-netbook
ganbold_ has quit [Ping timeout: 265 seconds]
HSt_ has quit [Ping timeout: 245 seconds]
penguin42 has joined #arm-netbook
bsdfox has quit [Ping timeout: 260 seconds]
HSt_ has joined #arm-netbook
Quarx has joined #arm-netbook
ganbold_ has joined #arm-netbook
ganbold__ has quit [Ping timeout: 265 seconds]
barqux has joined #arm-netbook
<barqux> hey, is anyone familiar with the mk802-II? :)
<techn_> barqux: ask ManoftheSea
<barqux> ManoftheSea: on the mk802-II, using the sun4i 3.0.52+ kernel, what's the approach for getting the system to recognize the entire 1GB ram? It currently gives me ~300 MB (500 - mali)
<barqux> techn_: ty
popolon has quit [Quit: Quitte]
popolon has joined #arm-netbook
HSt_ has quit [Quit: Page closed]
<hno> barqux, what method do you use for booting? u-boot from SD, or Allwinner bootloader from NAND?
rz2k has joined #arm-netbook
<barqux> hno: u-boot from SD
<barqux> hno: Also, the kernel cmdline contains mem=448M@0x40000000 atm. Which I realize will limit my memory, but if I remove it, it won't boot
<barqux> on top of it all I have no UART or any way to get error messages unless the kernel boots, so debugging is a bitch
rellla has joined #arm-netbook
eebrah has joined #arm-netbook
rellla1 has joined #arm-netbook
rellla has quit [Ping timeout: 265 seconds]
<hno> barqux, First of all you need an u-boot that is built for mk802-II 1GB.
<barqux> hno: I assume that the version provided with sunxi is suitable. There are settings for the mk802-II in there which seems suitable, but please correct me if this assumption is wrong.
ZaEarl has joined #arm-netbook
slapin_nb has joined #arm-netbook
popolon has quit [Quit: Quitte]
XenGi_ is now known as XenGi
datagutt has joined #arm-netbook
popolon has joined #arm-netbook
andi2342346 has joined #arm-netbook
<andi2342346> hello, does anybody know why CEDAR module does not compile for allwinner-a10? i get this error: No rule to make target `drivers/media/video/sun4i/cache-v7.c', needed by `drivers/media/video/sun4i/cache-v7.o
XenGi is now known as XenGi_
rellla1 has quit [Quit: rellla1]
freakazoid0223 has quit [Quit: Leaving]
provel_ has joined #arm-netbook
<provel_> Hi all
<provel_> If there are some serious developer here, I have a nice question
<provel_> that is: we nee KMS driver, and in kernels from 3.5 and above... there IS a Mali-400 KMS driver, I think
<provel_> as samsung exynos are mostly Mali-400 MP4
<provel_> aren't them?
bsdfox has joined #arm-netbook
<provel_> everybody's sleeping?
provel_ has quit [Quit: leaving]
rellla has joined #arm-netbook
<L84Supper> http://www.nytimes.com/2012/12/30/technology/indias-aakash-venture-produces-optimism-but-few-computers.html?hp&_r=0 "$50 tablet computer, to be built in India" and now a13 tablets are under $50 in bulk from China :)
<rm> $53 shipped in 1pcs quality
<rm> quantity*
<mnemoc> cheaper than dirty cheap
<rm> these "$50 tablet" news are written by (and for) clueless people who think there are no tablets but iPad and samsung
<L84Supper> the NYT article is actually about how a tablet like that was promised in India over a year ago and it still hasn't arrived
<mnemoc> :)
<L84Supper> it was a large government project that has failed again
<L84Supper> culture, corruption, both or ??
<L84Supper> IMHO opinoin it's too much involvement with tradition is the design vs something made to be assembled as close to 100% as possible
<L84Supper> machines are cheap and don't get sick as often, they don't get distracted by family problems or social issues
<L84Supper> India doesn't have the manufacturing experience that they have in Shenzhen but automation helps level the playing field
itamarjp has joined #arm-netbook
itamarjp has joined #arm-netbook
itamarjp has quit [Changing host]
<rz2k> [20:41:22] <provel_> that is: we nee KMS driver, and in kernels from 3.5 and above... there IS a Mali-400 KMS driver, I think - ARM were asked for mainlining, they said "no".
<rz2k> ah, he left. anyway.
andi2342346 has quit [Ping timeout: 260 seconds]
hg_5 has quit [Ping timeout: 252 seconds]
ZaEarl has quit [Ping timeout: 250 seconds]
servili007 has joined #arm-netbook
ZaEarl has joined #arm-netbook
rellla has quit [Quit: rellla]
HSt_ has joined #arm-netbook
Quarx has quit []
revident has joined #arm-netbook
freakazoid0223 has joined #arm-netbook
XenGi_ is now known as XenGi
andi2342346 has joined #arm-netbook
<andi2342346> does anybody know why sun4i_cedar.ko module does not compile? i get this error: No rule to make target `drivers/media/video/sun4i/cache-v7.c', needed by `drivers/media/video/sun4i/cache-v7.o
XenGi is now known as XenGi_
eebrah has quit [Ping timeout: 244 seconds]
XenGi_ is now known as XenGi
XenGi is now known as XenGi_
lkcl has quit [Ping timeout: 255 seconds]
servili007 has quit [Read error: Connection reset by peer]
servili007 has joined #arm-netbook
freakazoid0223 has quit [Quit: Leaving]
hg_5 has joined #arm-netbook
lkcl has joined #arm-netbook
ZaEarl has quit [*.net *.split]
barqux has quit [*.net *.split]
ManoftheSea has quit [*.net *.split]
niftylettuce has quit [*.net *.split]
destinal has quit [*.net *.split]
mnemoc has quit [*.net *.split]
barqux has joined #arm-netbook
niftylettuce__ has joined #arm-netbook
destinal has joined #arm-netbook
ManoftheSea has joined #arm-netbook
mnemoc has joined #arm-netbook
Jef91 has joined #arm-netbook
<rz2k> andi2342346: what kernel do you use
<rz2k> with our git one everything should be ok
eebrah has joined #arm-netbook
tinti has joined #arm-netbook
tinti has quit [Read error: Connection reset by peer]
<andi2342346> @rz2k i did git pull today... kernel is 3.4.19+
<andi2342346> i am running linaro 12.11 (armhf) everything else i need compiles fine, but the cedar module cannot be build
<rz2k> what
<rz2k> oh, sorry.
<rz2k> read that as 3.0.19 :p
<rz2k> try using our sunxi-3.0
<rz2k> 3.4 might be unstable
<andi2342346> i am not so familar with git, but i looked at the last commit-dates and it seemed that the 3.4 tree is much more up to date
<andi2342346> i will give 3.0 a try -thanks for the hint... btw. do you it is possible to use ffmpeg or other software to use hardware-decoding of mpeg2 with the cedar module?
<rz2k> no
<rz2k> cedar can be used by two things, XBMC with cedar support and VLC with cedar support
<rz2k> there are two libs for communicating with cedar, libcedarx (pure C one) and libcedar (C++ lib), you can write the binding to ffmpeg if you want.
<rz2k> (both libs depend on blobs from allwinner)
<andi2342346> hmm thats the reason why i need it - maybe the commandlineversion of vlc (cvlc?) may work for me... i will take a look at it... actually i really would like todo transcoding but hardware encoding is not supported by the mali gpu - right?
<andi2342346> if its not too complicated i will try to use the libs - thanks for the info
<andi2342346> and i will give xbmc a try... at the moment i run vdr headless on my mele
<andi2342346> vdr requires full feature card (hardware mpeg2 decoding), if you want to use OSD etc. so maybe its possible to add cedar support to vdr, this would be really nice...
<rz2k> cvlc is actually the only way to use the cedar
<rz2k> GUI is not supported and the guy who did the initial port seems gone
<rz2k> for xbmc and vdr talk with rella
<rz2k> he comes around often
<andi2342346> ok, thanks for the info... i will check it out if i have more time...
von_fritz has quit [Quit: vonfritz leaves, don't panic]
lkcl has quit [Ping timeout: 252 seconds]
lerc has joined #arm-netbook
hg_5 has quit [Read error: Connection reset by peer]
eFfeM has quit [Quit: Leaving.]
xDEADCA7 has joined #arm-netbook
servili007_ has joined #arm-netbook
servili007 has quit [Ping timeout: 245 seconds]
xDEADCA7 has quit [Quit: Sto andando via]
provel_ has joined #arm-netbook
<provel_> Hi all
<provel_> any programmer here? I need to know how to get current fb display size with mali drivers
<libv> provel_: s/with mali drivers// and you have your answer
<provel_> any programmer here? I need to know how to get current fb display size with mali drivers
<provel_> ok libv
<provel_> thank
<provel_> so I just need to know about fbcon
eebrah has quit [Quit: ChatZilla 0.9.89 [Firefox 17.0.1/20121129152039]]
<provel_> hum, I suppose a little ioctl is needed
Jef91 has quit [Quit: Instantbird 1.3 -- http://www.instantbird.com]
AC`97 has quit [Quit: or perhaps not. brb. shutting down to install raid card.]
lkcl has joined #arm-netbook
gimli has quit [Ping timeout: 260 seconds]
<rz2k> provel_:
<rz2k> [20:41:22] <provel_> that is: we nee KMS driver, and in kernels from 3.5 and above... there IS a Mali-400 KMS driver, I think - ARM were asked for mainlining, they said "no".
<rz2k> they started to roll out the skeleton for KMS driver couple of the releases now, but it still is not functional
jlj has quit [Ping timeout: 256 seconds]
<rz2k> and you dont need the fbcon, you need standard fb ioctls, like FBIO ones
andi2342346 has quit [Quit: bye]
jlj has joined #arm-netbook
<provel_> ok thanks
<provel_> yep , now I got that resolution ioctl.... but now I'm stopped again
<provel_> do OpenVG, even mesa OpenVG, works with mali fb drivers ?
<rz2k> mesa should work
<provel_> damm
<rz2k> mali has the openvg acceleration, but you need a Mali DDK
<provel_> I was porting retroarch from rpi to mele a200
<provel_> but with mesa openvg1.0 doesn't works
<provel_> and with xorg I have lot of cpu wasted for xorg... like 60%
<rz2k> welcome to ARM
<provel_> damn the stupid supplier
<rz2k> it will be always like that until ssvb will write his fbdev wrapper. http://cgit.freedesktop.org/~siamashka/xf86-video-fbdev/log/?h=wip-sunxi-mali
<rz2k> we've decied to toss out the xf86-video-mali because it is done horribly wrong
<provel_> eh.... but I would like a working openVG
<libv> provel_: why?
<provel_> retroarch
<rz2k> maybe his project uses Qt
<rz2k> I have this as only oVG application in mind right now.
<provel_> naaa I'm just porting my fav emulator
<provel_> and the vgImageSubData is needed
<barqux> what emulator? :D
<provel_> retroarch : after porting is nes,snes, genesis, neogeo, cps1, cps2
<rz2k> cool
<provel_> mame is on the work upstream
<provel_> but as of now I can barely run snes
<rz2k> I believe, we have the SDL emulators out in the wild
<provel_> sdl? there's some sdl backend working without xorg?
<provel_> (and with gles accel...)
<rz2k> I mean linux already has many of them, something should work on decent speed
<rz2k> :p
<ssvb> provel_: cpu wasted when doing 2D or 3D graphics?
<provel_> rz2k : tried.... snes9x and psnes under retroarch are about 100% whitout shaders
<provel_> and sound gets 500ms latency
<rz2k> damn
<provel_> sigh.... I'll keep on hold until there isn't any openVG improvement
<provel_> afterward I expect 50% cpu for snes
<provel_> ok, late here, bye everybody
<libv> heh
jlj has quit [Ping timeout: 265 seconds]
<rz2k> provel_: ssvb has done some measurements on xf86-video-mali. https://groups.google.com/d/msg/linux-sunxi/6lLMjfvZj44/lH5kFRB0bN8J
<ssvb> provel_: what exactly is your use case? I can try to test it, or you you can do some profiling yourself to identify the bottleneck(s)
<rz2k> and we dont have any support from ARM on Allwinner, no working OpenVG hardware acceleration is provided right now. we had the libraries with OpenVG built in, but it turned out that they had many missing symbols.