#qi-hardware IRC log for Monday, 2010-05-10

tuxbrainmorning channel!02:45
tuxbrainsee metrhil has finnally find a motivating stuff to get into :)02:46
tuxbrainmorning wolfspraul!03:11
wolfspraultuxbrain: morning03:12
tuxbrainI see methril has found something motivating to come into :)03:12
wolfsprauljust uploaded a nice picture showing IDA Systems headquarters in Jaipur03:12
wolfspraulif you have one for the Tuxbrain hq that would be cool too...03:13
wolfspraulyes indeed, very happy methril is excited03:13
wolfspraulthis mail from Fernando was excellent03:13
wolfspraulI will follow up on this, only problem is so much to do! It will all take some time...03:13
wolfspraultuxbrain: did you see, Harald was pretty happy about our MTK research as well... http://laforge.gnumonks.org/weblog/2010/05/09/#20100509-cect_c310003:14
wolfspraulneed to dig deeper, knowledge is power...03:14
wolfspraulBas did an Iris update03:14
wolfspraulAtommann reported that Debian doesn't boot for him, let's see how that goes03:14
wolfspraulroh sent some good links regarding the hoperf modules with some free software we can reuse, which is very very cool03:15
tuxbrainyes :) we have a late night chat about mtk with steve|m who is in charge of mkt stuff in osmocom-bb :)03:16
tuxbrainseem we have god matherial for a community update :)03:16
wolfspraulsteve? which steve?03:18
wolfspraulare you following osmocom? (didn't know :-))03:18
tuxbrainthe chat was in this channel03:18
wolfspraulthis phone I looke at costs 18 USD03:18
wolfspraulunbelievable, eh?03:18
tuxbrainseems that is osmocom who is following us :)03:19
wolfspraulwe could easily donate 20 or 30 to jumpstart development03:19
wolfspraulI love this power of technology, complete phone for 18 USD retail. Unbelievable.03:19
wolfspraulunfortunately many people are missing the real revolution that is going on03:19
wolfsprauleverybody sees the iphone and so, and I agree that is amazing. but they don't see the other revolution going on :-)03:19
wolfsprauloh well...03:19
wolfspraulhey David, this is for you (just kidding)03:20
wolfspraulguess what the retail margin of this phone is?03:20
tuxbrainyeah i we achevive to comercialize such stuff cheap enough with hackable free soft inside I thing we have winner here03:20
wolfspraulthe retail margin per phone is a whopping 5 RMB = 56 EUR cents!!!03:21
wolfsprauland bloodsuckers as we are, when we bought 15 of them we bargained it down to 34 cents!03:21
tuxbrainhahaaha wow, seem like tuxbrain margins hahaah03:22
wolfspraulyeah, I thought this might make you feel better...03:22
tuxbraincommon you have the guts to bargain with this prices????03:22
wolfspraulbut seriously, this 18 USD thing is amazing03:22
wolfsprauloh sure03:22
wolfspraulin China you bargain for every penny03:22
wolfspraulthe carbonized rubber button is 0.5 us cents I think03:22
wolfspraulbut we bargained it down to 0.4 US cents03:23
tuxbrainhey when we sell milions it will means thousands of $ :P03:23
wolfspraulyes the scary thing is that is what is happening03:23
wolfspraulwe need to get reall Linux into this machine...03:24
wolfspraultuxbrain: have you tried Debian on your Nano yet?03:24
tuxbrainsteve|m points to guys trying to do so in similar device mkt arm7 http://peeklinux.com03:25
tuxbrainnot, but soon if not today tomorrow03:25
tuxbrainnow I have another unit to use as openwrt testing I will use the mutant one to play with debian, and try to port avr toolchain there03:26
wolfspraulinteresting don't know this peek device03:26
wolfspraulif this is an anti-vendor effort it's not for me, I don't believe in that03:26
wolfspraulobviously since I try to become a copyleft vendor myself, then there is no more need for anti-vendor ports03:26
tuxbrainyep it seems like a rigid nn with sim card :P03:27
wolfspraulso hope they talk to the peek manufacturers etc.03:27
wolfspraulI hear there are some interesting things coming on sim cards.03:27
tuxbrainno seems it pure reverse ingeniering from the beggining to end03:27
wolfspraul'interesting' unfortunately from the wrong perspective, from the perspective of the operators who want to stay in control03:27
wolfspraulyeah the reverse stuff cannot work, you will always be chasing03:28
tuxbrainyep, but here we are to propose an alternative,03:28
wolfspraulreverse to gain knowledge is OK, that's what we do. but then need to produce yourself, or at least instruct the manufacturers yourself.03:28
wolfspraulyou have no idea how fast Chinese can change, and they will change over and over and over, for .01 cents savings they will take down compatibility with all the software you have been working on for years.03:28
wolfspraulwithout hesitation03:28
tuxbrainsure peeklinux guys will be pleased to work in a openhardware if we provide one :)03:28
wolfspraulI'll add the peeklinux channel, good link thanks!03:29
tuxbrainsorry copyleft hardware03:29
wolfsprauldid you hear some good (sales) news because of the Linux Magazin artice in Germany?03:30
wolfspraul4 pages in a nice magazine about Ben NanoNote, but don't know the circulation of the magazine03:30
wolfspraulPulster knows already03:30
tuxbrainI have no idea! :) good news :)03:31
wolfsprauljust sent you a link03:31
Action: tuxbrain is dreaming a qi-hardware 50¬ simple but free phone03:35
tuxbrainouch! wlan is very highlighted (too much)03:40
wolfspraultuxbrain: well that's what we are working towards, no?03:40
wolfspraulit's just a huge amount of technology in the little phone, so it needs some time to come together03:41
wolfspraulfortunately some people see the big picture, hopefully it will grow over time and we don't die on the way (but even then I'm sure others continue)03:41
tuxbraindie is not an option03:42
wolfspraultuxbrain: there you go, I know we will hang in for a while :-)03:42
tuxbrainok cu03:43
wolfspraultuxbrain: did you try Debian on your Nano? who did?03:43
Action: tuxbrain will try debian today or tomorrow03:43
emdetewolfspraul: i did ;)04:01
tuxbrainhehehe emdete authors not counting :P04:06
tuxbrainemdete: any progress in sd booting?04:07
emdetetuxbrain: :) yes, you are right. but josch takes over now, he got the dev today04:07
wolfspraulemdete: authors count, but now we need to transfer the experience to the readers...04:07
emdetetuxbrain: just playing around with it. did you have any success?04:08
tuxbrainnot, I will try the "classical" method in short04:08
wolfspraulemdete: the kernel from the latest openwrt-image, can it be used for debian without recompile?04:09
emdetewolfspraul: no, its missing two options, one will be switched on in the future as i understand, the other is in discussion. also we use differen partition sizes for now04:11
emdetebut it is possible to agree on a common set of options i think04:12
wolfspraulah OK04:13
Action: emdete will run to the office - back in a minute...04:13
joschhey guys04:14
joschgot my nn a few minutes ago and am in process of bootstrapping debian for it atm04:15
joschstay tuned04:15
joschtuxbrain: i guess you want to be pinged once i have smth?04:15
tuxbrainjosch: mmmm only if you explainme what is smth, it hurts?04:17
joschi hope it will not04:17
Action: tuxbrain can't wait and is flashing debian right now04:18
tuxbraindebian failed at first try, uboot and kernel ok but rootfs not found (kernel panic) erasing rootfs and reflashing again rootfs06:08
shevekDoes anyone know details about the nand flash?  Which bank is it connected to?  Does it need the NFCE bit set or not?  Which error-correction is better?07:07
qi-commitsCarlos Camargo: Minor changes on SAKC gerbers http://qi-hw.com/p/nn-usb-fpga/07ca55708:42
tuxbrainemdete: No way on booting debian, I guess there is problems with qi u-boot and partitions due it was unnable to open the rootfs, Ben cannot use the avt2 uboot it doesn't work on ben at all, where I can put the kernel panic log?09:09
emdetedepends on the u-boot. qi-default is serial, mine is lcd09:13
tuxbrainI have the kernel panic, I just want to know where to postit09:20
Action: tuxbrain will take a picture :P09:20
tuxbrainto upload on my own website09:21
emdetetuxbrain: just tell us the url ;)09:23
emdetetuxbrain: there is a wrong root= in the kernel args... can you do a `strings` on the u-boot binary you flashed to check the kernel params?09:27
tuxbrainops closed the bad window09:28
tuxbrainemdete:see the pic?09:29
emdetetuxbrain: yes09:29
tuxbrainany clue?09:29
shevek<emdete> tuxbrain: there is a wrong root= in the kernel args... can you do a `strings` on the u-boot binary you flashed to check the kernel params?09:29
emdeteshevek: thnx! :D09:29
shevektuxbrain: You just missed that. ;-)09:29
emdetetuxbrain: but i have overseen the line on top of the screenshot - ubi0... makes sense. but a strings is always a good thing to do ;)09:30
emdetetuxbrain: block(0,0) is probably a fallback09:31
shevekemdete: I expect it to be the detected major and minor number, which both remain 0 when detection fails.  But that makes it behave as a fallback, though a useless one. ;-)09:32
emdete0,0 is probably first ide and often right ;)09:33
tuxbrainnow the cuestion is... how I do that "strings"09:33
tuxbrainI'm a damn shopkeeper!09:33
emdetetuxbrain: :D in the shell type string u-boot*.bin09:33
tuxbrainin what shell?09:33
emdetehm... how do u flash?09:33
shevekemdete: No, 0 is not actually registered as a block device.  8 is ide.09:33
tuxbrainok using uboot09:34
emdeteshevek: oh, then it's always crap ;)09:34
tuxbrainok in the host shell!!!!09:34
tuxbrainmmmm I start feeling like a windowser in fosdem09:35
emdete:D we like you for selling nanos, not for using shells! :D09:35
tuxbrainouch!, my geek-ego is lying crying on the floor...09:36
tuxbrainresult of #strings openwrt-xburst-u-boot.bin|grep bootargs09:40
tuxbrainbootargs=mem=32M console=tty0 console=ttyS0,57600n8 ubi.mtd=2 rootfstype=ubifs root=ubi0:rootfs rw rootwait09:40
tuxbrainbootargsfromsd=mem=32M console=tty0 console=ttyS0,57600n8 rootfstype=ext2 root=/dev/mmcblk0p2 rw rootwait09:40
tuxbrain? I dont see any difference (appart of the mem and the mac between qi- uboot and yours)?....09:45
emdetetuxbrain: looks fine09:46
emdetetuxbrain: HE! we are even more excited if you do both!09:47
emdetetuxbrain: the kernel is the one from pyneo.org?09:47
emdetei'm a bit lost. larsc, do u have any idea what could go wrong?09:50
joschtuxbrain: ping11:03
tuxbrainjosch: pong11:04
joschtuxbrain: i just built a 75MB ubifs debian rootfs11:04
joschtuxbrain: i will now upload the ubi image and a tarball11:04
joschit boots fine and one can login with user: root and no pw11:04
tuxbrainok I will try once uploaded, I hope I have more luck than with the big one11:05
joschtuxbrain: the big one from pyneo.org?11:05
joschtuxbrain: what happened there?11:06
emdetetuxbrain: i think we have to fix in the beginning, probably u-boot or kernel is to blame, not the rootfs...11:06
tuxbrainemdete: I agree too11:07
emdetetuxbrain: vegyraupe succeeded with it, maybe we try to track down the diffs between your doiings and his11:07
vegyraupetuxbrain: dif you use the newest uboot ?11:09
vegyraupenewest == latest11:09
tuxbrainyep only black screen11:09
vegyraupewhich one r u using now ?11:10
tuxbrainthe one xianfu compliled with uart on tp4 tp5 ,11:11
tuxbrainit also has lcd kernel log11:12
vegyraupek tell you waht11:12
vegyraupeI am gonna upload the uboot I used11:13
tuxbrainok, I will try11:13
vegyraupeemdete: I used ur kernel, right?11:13
tuxbrainflashing uboot better that reflashing rootfs :P11:13
emdetevegyraupe: yes. you need to, fpu-emu, partition & inotify are needed11:14
emdetevegyraupe: we have to agree on that with the openwrt guys...11:14
emdetefpu is in discussion...11:14
vegyraupeso the only difference between tuxbrain's and my stories are the uboots11:15
tuxbrainvegyraupe: yep11:15
vegyraupeI am trying to find it11:15
vegyraupenot sure where i put it11:15
vegyraupegive me a few11:15
tuxbrain<dramatical mode> the future of copyleft hardware is on that uboot </dramatical mode>11:17
tuxbrainvegyraupe:I see a drop of sweet in you front?11:17
emdetetuxbrain: which probably just fall into his notebook triggering a short-circuit...11:20
emdetevegyraupe: still there?11:20
vegyraupeemdete: yepp11:21
vegyraupesaved the drop from destroying the universe11:21
joschtuxbrain: http://mister-muffin.de/nanonote/11:21
tuxbrainjosch: BTW what you have put in the small rootfs  and what differs with actual minimal?11:21
joschtuxbrain: the following packages have been added:11:22
emdetejosch: it's completly new and not based on mine, right?11:22
vegyraupetuxbrain: try http://downloads.qi-hardware.com/people/vegyraupe/openwrt-xburst-u-boot.bin11:23
joschtuxbrain: udev,module-init-tools,sysklogd,klogd,psmisc,mtd-utils,ntpdate,debconf-english,locales,screen,less,vim-tiny,console-tools,conspy,console-setup-mini,man-db,fbset,input-utils,libts-bin,ifupdown,netbase,iputils-ping,dhcp3-client,curl,wget,openssh-server,vpnc,rsync11:23
joschemdete: yes i built it anew11:23
joschthis is also the reason libts-bin is in there11:23
joschbecause i wrote the script for the openmoko platform11:23
tuxbrainjosch: tar.lzma content=ubi content?11:24
emdeteconspy is cool... :D wasnt ifupdown already in?11:24
emdetetuxbrain: still wanting to boot from sd? :D11:24
joschtuxbrain: no it is a tarball of the rootfs11:24
joschtuxbrain: i remember you wanted to put it on sd11:25
joschemdete: what do you mean by "already in"?11:25
emdetein my mini-rootfs11:25
tuxbrainyep :) , wolf and vegy knows how I supply my lack of knowledge, with pure obstination11:25
joschemdete: might be but only because you installed it - i can check that later11:26
emdetejosch: don't need to. i just point my wiki to you page ;)11:27
emdete(and forget about my rootfs)11:27
joschemdete: dont do that yet11:29
joschemdete: i'm still not finished with it11:29
emdetewe settle down you stuff and then we switch11:30
tuxbrainno luck just flashing the uboot, same result as uboot of lastest release11:31
vegyraupetuxbrain: I suggest you empty the whole NAND11:32
tuxbrainI will wipe out whole nand and start again, I will flash the smaller rootfs this time11:32
vegyraupeand then go to usbboot manually11:32
tuxbrainlet fanfares sound loud and proud!!! I boot in debian :) :) :)11:45
tuxbrainvegyraupe: does that uboot has the lastests xianfu patches on uart stuff?11:50
emdetetuxbrain: what did you do different this time?11:50
tuxbrainvegyraupe uboot + josch rootfs11:51
tuxbrainpyneo kernel11:51
joschoh n1ce11:51
emdetevon allem nur das beste11:52
emdeteoh, sry11:52
emdetevegyraupe: could you translate? :D11:52
vegyraupethe best of both worlds :)11:53
vegyraupeI guess11:53
vegyraupeok so the prob is uboot ?11:53
emdetevegyraupe: both? three! :D11:54
tuxbrainyep most probably, now that we found a sure combination I will try to reflash the big one again to eliminate that variable also11:54
tuxbrainemdete: you forguet lamer obstination to the equation11:55
tuxbrainflashing rootfs directly if it doesn't work I will wipe out nand and do vegy-uboot+pineokernel+pineoenourmousedebianrootfs again, if it stillnot working I will return to save trio11:59
tuxbrainvegyraupe: you know if that uboot has the lastets xiang uart patches?12:00
vegyraupetuxbrain: I am sure it doesn't12:00
vegyraupeit is from the april image12:01
vegyraupesomeone needs to figure out what changed in those 2 uboots12:01
tuxbrainsh?&t , well we can't have it all isn't it?12:01
vegyraupebut we can damn well try :D12:01
tuxbrainyes I hope xiang or larsc wake up from the sofa and put at least this item in their empty todo list :P12:04
tuxbrainmeanwhile I will work in the compiling part of the arduino plan, due flashing has few chances if we can isolate keyboard from uart12:07
tuxbraincan -> can't12:08
tuxbrainvegyraupe: now the sweed drop is in my front, does the uboot support the U+Power to flash or was the buggy one that doesn't work?....12:30
vegyraupetuxbrain: pwr+U *should* work12:33
tuxbrainok I will confirm in a *couple* of minutes12:35
qi-commitsJuan64Bits: Adding expansion pcb for sakc http://qi-hw.com/p/nn-usb-fpga/5938d6513:21
qi-commitsLars-Peter Clausen: pwm fixes. http://qi-hw.com/p/qi-kernel/f1df89313:59
qi-commitsLars-Peter Clausen: jz4740-battery: Add MODULE_ALIAS entry http://qi-hw.com/p/qi-kernel/78fb31513:59
mirkohrmpf, stupid opkg... :(14:24
mirkoit does not even allow underscores in package names14:24
mirkowell, it does... - but it results in very misleading faults...14:24
mirkolarsc, i guess we have to rename all the libX*-packages as well14:25
mirkoand dependencies to them14:25
larscwe could patch opkg14:29
mirkolarsc, yeah, however that way our packages are incompatible with other systems which use opkg14:29
mirkoit's quite common that people are using openwrt-ipk's in their environment14:30
larscthen we should push that patch upstream14:30
mirkolarsc, you don't think they did that for reason?14:32
larscdon't know14:32
qi-commitsCarlos Camargo: Adding post route simulation to FPGA  examples http://qi-hw.com/p/nn-usb-fpga/717c35e15:57
qi-commitsCarlos Camargo: Adding plasma test_bench files http://qi-hw.com/p/nn-usb-fpga/fa1b48e16:08
qi-commitsCarlos Camargo: Adding modelsim simulation files http://qi-hw.com/p/nn-usb-fpga/23184f316:09
tuxbrainemdete , josch, vegyraupe thanks to the serial output I think I finally arrive at some conclusions see the list post here16:43
tuxbrainhere :P http://en.qi-hardware.com/pipermail/developer/2010-May/002932.html16:43
tuxbrainI want to thank you all for the support in this.16:43
maden!stats test17:41
madenwrong chan17:57
tuxbrain:) ok17:58
Action: tuxbrain is fascinating seing apt-get installing packages18:04
wolfspraultuxbrain: great news! you did it again18:06
tuxbrainobstination and friends are powerful weapons :P18:07
wolfspraulkristianpaul: u there? does Debian work on your Nano now?18:07
zeartuxbrain, any benefits from running debian on nano?18:07
kristianpaulwolfspraul: not18:07
kristianpauli need time to test18:07
Action: kristianpaul now at work18:07
tuxbrainzear: a looooooooooooooooooooooot of pakages ready to install and play18:07
zearany graphical ones? :)18:08
kristianpaulseems after try some times flashing it works as i was told here before18:08
tuxbrainzear:not tested yet, I'm focused on compile/flashing arduino stuff18:09
zearah, ok18:09
wolfspraulkristianpaul: you shouldn't need to flash several times. either it's flashed or not.18:10
wolfspraulif it doesn't work after the first flashing, let me know18:10
tuxbrainkristianpaul: I agree with wolfgang, I have made a lot of trys I think the problem is on the big image , well not really on the rootfs itself is more like uboot/kernel , can't deal with such a monster. take a look at http://en.qi-hardware.com/pipermail/developer/2010-May/002932.html and you will find a win combination :)18:14
tuxbrainit's late and I'm barely falling sleep in front of the keyboard but.... I successfully compiled and Arduino sketch in a Ben NanoNote! :), tomorrow will start the flashing phase and see if every thing was really ok , good night20:00
tuxbraingood news is that bitrate of upload to Arduino 328 is same as default in ben serial, so let's cross fingers20:02
tuxbrainisn't free software beatifull?20:02
qwebirc59542hi, everybody,20:47
qwebirc59542i'm trying to compile with MAKE the openwrt and always says me this20:47
qwebirc59542 make[3] -C toolchain/binutils install  make[3] -C toolchain/gcc prepare  make[3] -C toolchain/kernel-headers prepare make -r world: build failed. Please re-run make with V=99 to see what's going on make: *** [world] Error 120:47
qwebirc59542just next of TOOLCHAIN/kernel-headers20:47
methril__tuxbrain_away: it's amazing!! i love it ;)23:01
qi-commitsJuan64Bits: Moving files. http://qi-hw.com/p/nn-usb-fpga/7766f6323:06
qi-commitsJuan64Bits: Fixing expansion board. http://qi-hw.com/p/nn-usb-fpga/ac9d3c823:09
--- Tue May 11 201000:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!