#qi-hardware IRC log for Tuesday, 2010-05-11

qi-commitsXiangfu Liu: update READE, add INSTALL file http://qi-hw.com/p/xburst-tools/cd7bc7f01:47
fred-ohi guys, new nanonote owner here. i have an embarrasing question: how do i exit gmu? :)03:38
tuxbrain_awayfred-o: alt+enter03:40
fred-osweet! thanks :)03:41
tuxbrainIt flash!!!!! it flash!!!! Nanonote has flashed and arduino board!!!!! yippy kay hey!!!!03:57
Action: tuxbrain is dancing the happy geek dance03:57
vegyraupetuxbrain: congrats!03:57
vegyraupecan we get a video?03:57
vegyraupeof the dance I mean ;)03:57
tuxbrainhehehe sorry human being is not ready for this yet03:58
vegyraupeI understand ;)03:58
tuxbrainbut take for sure a new heavy metal video clip will born from this :)03:59
emdetetuxbrain: how is it connected?04:37
tuxbraintrough serial port, using test points tp4 and tp504:41
tuxbrainyou can see detailed pics of the wiring here http://en.qi-hardware.com/wiki/Serial_console#Hardware_mod_to_have_a_pluggable_serial_port04:42
Action: tuxbrain is cleaning the "studio" to make the video demo04:45
emdete:D great - pics & video rule!04:51
tuxbrainBNN hardware modified... debian running... arduino flashed... I feel an emptiness  in my geeksoul .... once the tutorial/video was done ....what outofbounds goals I should point now? :P04:53
emdetetuxbrain: wee need some case modelling...! :D04:54
tuxbrainemdete: have you see this one? http://www.youtube.com/watch?v=4F5g559TMrE04:54
tuxbrainemdete: if BNN can control arduino, it can control a RipRap or Makerbot :P04:56
tuxbrainmmmm if they not where so f%&ing expensive using a BNN with some dedicated prog as interface to those 3d printers sould be a interesting project to get into04:59
emdetetuxbrain: nice music...05:03
tuxbrainyeah nanowar are great musicians with a greater sense of humor :), and also GNU/Linux and free knowledge advocates05:06
zeartuxbrain, wait, a metal band interested in GNU/Linux?05:08
emdeteso there is direct wirering between nano & arduino? no additional converter?05:08
emdetetuxbrain: amazing..05:09
tuxbrainzear:yep, and also they licence they music under CC :P05:10
zeartuxbrain, so technically we could put some of their music as ogg samples or use it as a starting tune for the nanonote?05:11
tuxbrainemdete:yes in spite of they work in diferent voltages, BNN 3,3 Arduino 5 , Arduino is interpreting 3,3V as logical 1 and it really does not output 5V is more a 4-4,2V son Ben can handel it :P05:12
tuxbrainzear:yep :) I have some mails with gatto (the *leader* of the band) and he is totally happy to be related with NanoNote project :)05:14
zeartuxbrain, that might be very handy when we develop an alarm clock app05:15
zearso their guitar solo could be used as an alarm tune :)05:15
qi-commitsJonathan Nieder: update changelog http://qi-hw.com/p/xburst-tools/b65e32105:16
qi-commitsJonathan Nieder: avoid accidentally tracking the xbboot stage1 firmware http://qi-hw.com/p/xburst-tools/63d8f0805:16
qi-commitsJonathan Nieder: debian/rules: fix clean target http://qi-hw.com/p/xburst-tools/dbe9c8b05:16
qi-commitsJonathan Nieder: update .gitignore http://qi-hw.com/p/xburst-tools/e6a1f0f05:16
qi-commitsJonathan Nieder: include entire source tree in source package http://qi-hw.com/p/xburst-tools/78f38d405:16
qi-commitsJonathan Nieder: unset executable bit on firmware image http://qi-hw.com/p/xburst-tools/e14041105:16
qi-commitsXiangfu Liu: update INSTALL, add how to get cross-toolchain http://qi-hw.com/p/xburst-tools/4ffa69605:16
tuxbrainhe :D yeah they have some noisy themes to weak up anyone05:16
zearbtw http://code.google.com/p/grafx2/ might be worth porting this app05:17
tuxbrain[...]and generally any detailed graphics painted with a mouse[...] mouse? in BNN?05:18
zeartuxbrain, they made a port of it to the gp2x05:19
zearso i believe we could use some code from it for mouse emulation05:19
tuxbrainops! gp2x doen't have touchscreen?05:19
zeartuxbrain, only the F200 model05:19
tuxbrainok :)05:20
tuxbrainis should be nice to have a image editor on BNN :) but I'm still waiting for just a viewer :P05:21
zeartuxbrain, oh, it has mouse emulation support already compiled in05:22
zearjust hold ctrl and move it with arrows05:22
tuxbrainthis will easy a lot the work :P05:23
tuxbrainzear: do you know zLouD? he has a very simple command line fb viewer Dinimavie http://www.gp32x.com/board/index.php?/topic/52364-dinimavie-ver-20100303/05:26
zeartuxbrain, yeah, i think he hangs out at #dingoonity sometimes05:27
tuxbrainI have tryied to reach him without success05:27
zeari myself ported another image viewer for dingux, but it couldn't load anything greater than 320x24005:27
zearso not good for hi res pics05:27
tuxbrainIt will be awesome if he release the the code under GPL , it also work with big image and rescale is very fast05:28
tuxbrainand is simple to use05:28
tuxbrainbinary works out of the box on bnn but without the code can't be integrated in the main distro05:29
zeari understand, will ask him for the source then05:30
tuxbrainthanks a lot man05:31
tuxbrainsure you have more chances to be listened than me :)05:31
tuxbrainmust leave for a moment now05:32
emdete:D tuxbrain leaves and silence comes into this channel...08:18
joschis this good or bad news? :P08:29
Action: tuxbrain_away is always watching08:30
zeartuxbrain_away, i asked zLouD about dinimavie, and currently the code isn't available08:33
zearbut he's working on a new version which will be gpl'ed08:33
tuxbrainwtf has done with the code to make it unaviable?08:34
zearit never was available, was it?08:34
tuxbrainnop, but if he is the author?08:34
zeari guess it's too shitty to be published or something ;)08:34
zearbut he said he's working on a new - gpl'ed version, so it's only for the better08:35
tuxbraindo him does directly in 0 1 0108:35
tuxbrainjust joking it's good news all in all08:35
zearoh he has the code, i mean the code is not open08:35
zearso not available form the public08:35
tuxbrainthanks a lot zear, ping me if you have more news about this08:36
tuxbraingoing to eat, emdete you can continue missing me :P08:37
kristofferanyone actually tried the debian port? Everything went well, but nanonote is unresponsive afterwards :)08:45
vegyraupekristoffer: yes and it worked for me otherwise I wouldn't have posted the info08:54
vegyraupedavid also reproduced it08:54
vegyraupeand sent a mail about it yesterday08:54
kristoffervery weird then08:55
vegyraupecheck david's email, he found the combination it works in08:55
kristofferwhat combination? I just downloaded all the files from his site08:55
vegyraupeI'd also suggest erasing the NAND before flashing the kernel and uboot08:56
kristofferthat I didnt do, so might been the cause08:57
vegyraupemy wild guess is, that the problem is not debian itself08:58
kristofferno I dont think so either, since I atleast would have gotten something on screen (or U+PWR would work)08:58
kristofferso its bootloader and/or kernel08:58
vegyraupeit is the different size of the kernel08:58
kristofferit blew the barrier size?08:58
vegyraupeI do remember that we had problems if the kernel being flashed was smaller than the one being replaced08:59
zearkristoffer, for some reason when i try to flash my nanonote with the flash script, it never correctly flashes the bootloader09:00
zearand have to manually flash it again09:00
zearthat's probably what happened to you09:00
kristofferIt also had some issues when the kernel was too large, atleast early on. I remember hitting that a couple of times.09:00
kristofferdoh, now gotta spend 1-2 hours getting into usbboot mode :P09:02
zearkristoffer, i know how it feels ;)09:02
zearthough, i noticed that after i hold the pins shorted for a longer time (2-3sec) it usually boots into usb mode09:02
kristofferoki got bootloader going (no more short circuiting), so going to try replace kernel (it blacks out after loading kernel).09:21
emdetekristoffer: i soldered it to keep it in usbboot while testing my scripts ;)09:25
kristofferemdete, its an good idea. What I really miss is the serial output, so used to use it for kernel debugging.09:26
emdetekristoffer: yes, i have no serial here. that's why i switched u-boot to tell the kernel to log to the lcd09:27
emdetekristoffer: but i don't do really kernel debugging anyway...09:27
kristofferemdete, yeah same here. I think the bootmessages should automaticly come up on lcd. But I can understand that windows users hate that09:28
emdetei think we are not in a state that we do enduser cosmetics, even a win user has to read whats coming up there to get help.09:28
emdetebut i understand you point. if everything works fine, i prefere a splash too09:29
kristofferI agree with you, I would really prefer the standard kernel/u-boot spitting out info while booting. And thats what Ive said on the mailinglist09:32
kristofferhmm, reflashed the kernel and still black09:42
emdetekristoffer: the u-boot starts with output on lcd - do u see that?09:44
kristofferyeppers, so should be kernel09:54
kristofferunless kernel is feeding output into oblivion and userland takes forever to load09:54
emdetenonono, i mean if you dont see anything we shouldnt talk about kernel, we should look why the u-boot does not load/start09:57
kristofferyeah agreed :) Maybe I should try and erase09:57
kristofferI see the bootloader so thats fine09:58
emdetei made me a script doing /all/ steps and always start with a full erase when i played with a problematic device09:58
emdetesee http://pyneo.org/downloads/nano/nano-1GB/reflash.sh (which does not fit for you!!) - but as an example...09:59
kristofferroger thx10:00
emdetedon't use it without adjusting the settings... all numbers are wrong for a ben...10:00
kristofferyeah, was thinking that I should erase from kernel start and a couple of megs "up"10:01
kristofferjust to make sure that its clear for when I reflash the kernel10:02
emdetei never liked that aproach... i always flash /all/ parts all the time i flash10:02
emdetetoo much possibility for errors to come in otherwise10:03
kristofferhmm agreed, I guess.10:05
kristofferjust a pity when bootloader is working, to risk messing it up10:09
emdetenonono, when you use a script and that manages to flash it once it will do every time. trust me10:14
kristofferdoh, now bootloader went away again10:35
kristofferand getting some flash errors :(10:35
tuxbrain_awaykristoffer:carbonized rubber time! :P10:40
emdetekristoffer: can you cut&paste the flash errors you got?10:41
emdete(not here... :D )10:41
tuxbrainI also recomend emdete aproach, erase all, uboot, kernel, rootfs10:41
tuxbrainyou will save a lot of time of trys10:41
emdeteplus: do it in a script and fix the script as long as it fails...10:42
emdetekristoffer: i really would like to see the error message10:42
emdete(i want to enhance my page about taht and did not get any errors anymore lately)10:43
kristofferone sec then10:45
kristofferemdete, this is when I do usbboot -c "boot".10:48
kristofferdoh, I mean when flashing bootloader10:48
kristofferI dont know if that "no check!" is an error or warning10:49
kristofferEarlier I got alot of bad sector shit, let me see what kernel flashing says10:49
emdetemom, one step by another10:50
emdeteyou did a boot, a nerase, a nprog, right?10:50
kristofferhttp://pastebin.com/BebFYnuV this is kernel flashing, only see that it skips an bad sector at end.10:50
emdetehave you read my page about nano flashing? :D10:51
kristoffernot sure, got a link?10:51
emdeteif a message from usbboot has no newline (as yours) the flashing did /not/ succeed :)10:52
kristofferyes that I know, but what can one do? If the flashing continues to fail..?10:52
emdetecould you retry flashing the u-boot?10:54
emdetewhat happens?10:54
kristoffersame output10:54
kristoffererased it and it marks the bad sectors but get same result when flashing10:55
tuxbrain_awaykirtoffer: I call it "usb port fatige", try change usb port on host computer or restarting host10:56
tuxbrain_awayalso be sure the cable is good10:56
tuxbrain_awaythe one that comes witn BNN quite good10:57
tuxbrain_awaybut sometimes the UPF happens sometimes10:58
tuxbrain_awaywhatever cable you use10:58
kristofferrebooted, changed cable and changed usb port. Getting same issue.11:18
kristofferThere been any recent changes to usbboot so I should upgrade it?11:19
emdetekristoffer: which version are u using?11:20
emdete0.0+200906-1 is mine11:20
kristofferusbboot version: 20090630-111:20
emdeteyes, same11:20
emdetedo u use a script?11:20
kristofferand it worked the first time I reflashed device (bootloader but not kernel)11:20
kristofferno, but unless its some sort of timing thing I cant see how it matters.11:21
emdeteto exclude errors introduced by humans sitting at the kbd... even me succeded with that aproach ;)11:22
kristofferI can agree on that, but Im only aiming for the bootloader currently, which is only boot, delete, flash11:24
emdetedo as you like.11:27
kristofferI just want some options, currently it seems like bad sectors in wrong places are messing things up.11:30
tuxbrain_awaykristoffer:just one thing, have you enabled nand erasing in the usbboot.cfg12:33
tuxbrain_awaysed -i 's/NAND_FORCEERASE = .*/NAND_FORCEERASE = 1/' /etc/xburst-tools/usbboot.cfg12:33
kristoffertuxbrain_away, yes12:57
kristoffertuxbrain_away, I ran benflash script and got bootloader working, but kernel got bad CRC12:58
kristofferConsider an bad block inside the kernel area, how can I move beyond that? Map the kernel area somewhere else?13:48
emdetekristoffer: that is done automaticlly if you proper nerase13:56
emdeteplease read my page... ... ... ...13:56
kristofferIve done that, but cant see that it offers anything extra that Im not already doing13:57
qi-commitsJuan64Bits: Fixing expansion board for sakc. http://qi-hw.com/p/nn-usb-fpga/978a55514:55
KristofferI believe my nanonote is giving up.15:22
emdetekristianpaul: hm, we have to improve all this, it's anoying. it worked for me using the script15:23
Kristoffergetting more and more bad blocks :(15:23
emdetekristianpaul: nonono!15:23
emdetekristianpaul: thats all wrong usage of usbbot... trust me, i had that too15:23
Kristofferemdete: you talking to me?15:24
emdetewe still just don't have a reliable way to flash. let me prepare a script and you try /exactly/ with that script. would you do that for me?15:24
Kristofferyeah sure15:24
emdeteKristoffer: sure oups.. :D15:24
emdeteKristoffer: was some tab-auto-expand here...15:24
Kristofferbeen at this for hours now so Im open for anything15:25
KristofferI guess as much :)15:25
emdeteharhar, were does kristianpaul come from?!?15:25
emdeteKristoffer: i have been there too, ask vegyraupe... ;) he suffert from that alot15:25
emdeteKristoffer: give me a bit, i will put a script somewhere...15:26
kristoffer_back, had to reboot15:29
kristianpaulemdete: hello15:31
emdetekrist* have problems :D :D15:32
kristoffer_Oki, got it inside usbboot mode now15:36
kristianpaulemdete: np15:36
emdetehttp://pyneo.org/downloads/nano/reflash.sh is there. i can't test it now, cause i send my nn to josch15:36
emdetekristianpaul: sry waking you up.. ;)15:37
emdetekristoffer_: it curls the version that work for me. you need curl..15:37
kristianpaulemdete: ohh you are from pyneo?15:39
emdetekristianpaul: yes, is that good or bad?15:39
kristianpaulemdete: let me think ;)15:39
kristianpauli dint knew it about that script15:40
kristianpauli had some pain trying to install debian on my nano, until tuxbrain_away giveme some answer15:40
kristoffer_its transfering15:40
kristianpaulabout why it dint booted15:40
kristianpaulemdete: so how usefull is debian on a nn right now?15:41
emdetekristianpaul: i just put it there, i managed to get debian on the 1GB model today it the script was he outcome ;)15:41
kristoffer_hmm, when I erased I did it with tailing -n15:42
kristoffer_but you dont do that inside the script, whats the difference?15:42
emdetehm, i use -n for nprog only... hm15:43
kristoffer_just asking since thats the difference I can see atm.15:44
kristoffer_60% transfered of userland so far15:44
tuxbrainkristianpaul: the true kristianpaul have you achive to install debian?15:45
kristoffer_emdete, btw like your script, alot more informative this way.15:46
emdetekristoffer_: that's the cause i do scripts ;) it's informative for me too, say in some weeks when i have to redo all that stuff. i prefere clean script in favor to prosa wiki pages15:48
kristoffer_point taken, especially if this decides to suddenly work.15:49
emdete-n is probably ignored for nerase: image type must be: -n:no oob -o:with oob no ecc -e:with oob and ecc15:49
kristianpaultuxbrain: nope tried yet15:49
emdetekristoffer_: nono, point is: if it does no work we learn how to even improve it for the next guy asking. /this/ is the way open source works!15:50
Action: tuxbrain gonna be f%&/ing picky,15:51
Action: tuxbrain has think it twice15:52
Kristofferwhen I flashed last time (manually) I had approx 200 bad blocks out of 409615:53
tuxbrainKristoffer: how about jlime on BNN? any progress (I know you where too bussy with studys)15:53
Kristoffertuxbrain, jlime has bought an dev machine (nanonote) for another developer so we should be back in buissness shortly, Im tweaking the kernel and he will work on the GUI.15:54
Kristofferso we got 3 nanonotes in developer hands so far15:54
tuxbrain:) great , also have a meeting soon , isn't it?15:55
Kristofferyeah, the idea is to meet up in poland in august/september, will be fun.15:55
Kristofferemdete, oki, its done flashing now. So will see ...... *drumroll*15:56
Action: tuxbrain is holding his breath15:57
Kristofferwell, seems like bootloader is working but getting black screen after it loads kernel.15:57
Kristoffersure, but for how long?15:58
Kristofferafk, Ill let it work for awhile15:58
tuxbrainmmmm too much15:59
tuxbrainwhat rootfs are you using?15:59
emdeteKristoffer: any errors?15:59
tuxbrainwith official uboot you don't have log16:00
tuxbrainkristoffer if you are not pissed off yet, try my combination16:01
Kristofferits still black16:04
Action: kristianpaul dejaboo16:04
Kristofferyeah, I should try different bootloader to get output16:04
Kristofferreflashing with other bootloader16:07
Kristofferusing different bootloader, no change. Still black screen after bootloader16:36
Kristoffergoing to give it a rest for tonight.16:37
Action: kristianpaul dejaboo16:37
Kristofferthx for help so far, nite16:37
qwebirc21601is there a way to reflash openwrt-xburst-u-boot.bin when my u-boot is messed up?17:08
darfgarfqwebirc21601, should be, you'll need a copy of a working u-boot, usb-tool and you need to know what offset to flash it to17:09
qwebirc21601usb-tool, not usbboot ?17:11
darfgarfoh yeah, brain farted17:12
qwebirc21601when I connect the serial to the ben, then  the bootprocess ends with:17:13
qwebirc21601[U] pressedn17:13
qwebirc21601enter USB BOOT mode17:13
qwebirc21601Starting U-Boot ...n17:13
qwebirc21601but it does not activate the usb interface17:13
qwebirc21601at least I do not get any message from my systems kernel.17:13
darfgarfno idea really, but i've only ever got it to conenct properly by shortin ghte pins under the battery, not with power+u17:14
qwebirc21601hm. that does not sound good.17:16
darfgarfcould just be me, i tend to break things a lot.  but once i get mine into flashing mode, the reflash_ben script has worked flawlessly every time17:17
zeardarfgarf, same, power+u returns an error in usb-tool17:17
qwebirc21601<darfgarf> yeah, but that's if u-boot is working. I broke u-boot ...17:18
darfgarfqwebirc21601, so wouldn't puttin git into usb mode by shortin gthe pins be better? as that's a hardware thing built into the chip17:19
zearqwebirc21601, short the pins and you can flash u-boot again17:19
qwebirc21601ok, trying ...17:19
qwebirc21601thanks a lot, u 2 :-)17:21
--- Wed May 12 201000:00

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