#qi-hardware IRC log for Monday, 2010-10-25

qi-bot[commit] Werner Almesberger: Cleaned up component values in cntr schematics. http://qi-hw.com/p/ben-wpan/021bb0000:01
qi-bot[commit] Werner Almesberger: Assorted termination simulation cleanup. http://qi-hw.com/p/ben-wpan/2e742d900:01
qi-bot[commit] Werner Almesberger: bom/Makefile generalization http://qi-hw.com/p/ben-wpan/6e0abca00:01
qi-bot[commit] Werner Almesberger: Added BOM data for "cntr" sub-project. http://qi-hw.com/p/ben-wpan/69ddbf600:01
qi-bot[commit] Werner Almesberger: Take "pads" and "stdpass" modules from kicad-libs and remove local copy. http://qi-hw.com/p/ben-wpan/a2e91cf00:01
qi-bot[commit] Werner Almesberger: Revert TVS size reduction. We're not ready for it yet. http://qi-hw.com/p/ben-wpan/57874cf00:01
qi-bot[commit] Werner Almesberger: Great renaming: atrf/wpan-atrf* becomes atusb/atusb* http://qi-hw.com/p/ben-wpan/b493f0900:01
qi-bot[commit] Werner Almesberger: Cleanup of values in atusb/*.sch http://qi-hw.com/p/ben-wpan/9d6906d00:01
qi-bot[commit] Werner Almesberger: atusb/atrf.sch (X1): don't explicitly say that the tolerance is a maximum http://qi-hw.com/p/ben-wpan/a0f078100:01
qi-bot[commit] Werner Almesberger: Added BOM data for "atusb" sub-project. http://qi-hw.com/p/ben-wpan/01a9f2c00:01
wpwrakwolfspraul: here's one for the schematics history: atrf/ben-wpan.{pro,sch} are now atusb/atusb.{pro,sch}00:03
wolfspraulyou mean I just rename it?00:03
wpwrakyou just change the path in the schhist2web invocation, yes00:04
wpwrakand then let's see if the "follow across all crazy renames" code works :)00:04
wolfspraultitle is ben-wpan/atusb?00:08
wolfspraulso let's see, I have00:09
wolfspraulrun_schhist2web -S --title ben-wpan/atrf --order wpan-atrf:USB:RF ben-wpan atrf/wpan-atrf.sch output/atusb00:09
wolfsprauland it will become00:09
wolfspraulrun_schhist2web -S --title ben-wpan/atusb --order wpan-atrf:USB:RF ben-wpan atusb/atusb.sch output/atusb00:09
wolfspraulhow about the --order parameter?00:09
wolfspraulI would like to change the output directory to output/ben_wpan.atusb or so, what do you think? or ben_wpan_atusb?00:10
wpwrakthe wpan-atrf in other should change, too. otherwise, the order comes out a little wrong.00:12
wpwrakthe rest looks good00:12
xiangfuaisa: Hi you there00:12
wolfspraulso it's --order wpan-atusb:USB:RF ?00:13
wpwrak(output dir) dunno. i would keep on using atusb/00:13
wpwrakno, --order atusb:USB:RF00:13
wolfspraulok00:13
wolfsprauloutput stays, will change it now00:13
aisaxiangfu: I am!00:13
aisaI managed this weekend to catch up on my life in a serious way.00:14
aisaI'm starting to feel normal again.00:14
wpwraklet's keep our fingers crossed :)00:14
aisaI had to work this weekend,00:14
xiangfuaisa: http://projects.qi-hardware.com/index.php/p/openwrt-xburst/source/changes/backfire/00:14
aisafinishing a project I planned on having done a week and a half ago.00:14
xiangfuaisa:  this branch is your LOCAL branch right??00:15
aisaoh, hmmm...  I may have accidentally done that.00:15
aisaI meant that to go to tracking_backfire.00:15
aisaAnd I have planned early this week to bump the original patch,00:16
xiangfuaisa: ok, I will remove that one. :)00:16
aisaon the OpenWrT list.00:16
aisagood, ok.00:16
aisaI'll get this comitted to tracking_ and reping openwrt about it.00:16
aisaHaving been on the openwrt-devel list, I think I understand why it was ignored.00:16
wpwrak(crazy renames) not quite sure if i also cover directory changes. well, we'll see soon enough :)00:17
xiangfuaisa: ok.00:17
xiangfuwe will create a new branch name "master". which is our working on. then try to cleanup the 'tracking_backfire',00:18
Action: xiangfu will create the new branch afternoon.00:18
wpwrakhmm .. You don't have permission to access /schhist/atusb/ on this server.00:18
wolfspraulwpwrak: yes, wait a second00:19
wolfspraulI just renamed it all to .obsolete, now it's being picked up from clean00:19
wpwrakah, i see00:20
wolfspraulwe can always go back if you want to stress test your caching etc. as well00:20
wolfspraulthe server runs a bit slower than usual, it looks like both mirko and xiangfu are running builds on it :-)00:22
wpwrakyeah ! didn't miss a beat :-)00:25
wolfspraulwpwrak: ok looks good I guess. I will delete the .obsolete dirs on the server.00:48
wpwrakyeah. worked flawlessly00:51
wolfspraulserver doc updated, so all done. thanks for the heads up.00:52
wpwrakthanks for updating !00:52
nitin_guptawhat is hardware usbboot ?01:00
nitin_guptaxiangfu told me that as your bootloader is flashed so go for hardware USB Boot01:00
qi-bot[commit] kyak: gitignore vim temporary files (*~) http://qi-hw.com/p/openwrt-xburst/51502be03:23
qi-bot[commit] Werner Almesberger: modules/Makefile (MODULES): belatedly remove "pads" and "stdpass". Oops. http://qi-hw.com/p/ben-wpan/e3c850707:45
qi-bot[commit] Werner Almesberger: Added Molex 0480371000 USB A plug (SMT) http://qi-hw.com/p/ben-wpan/f73f0a907:45
qi-bot[commit] Werner Almesberger: SOT-323 footprint for MMST3904 NPN transistor (use NPN standard symbol) http://qi-hw.com/p/ben-wpan/7dd3d1a07:45
qi-bot[commit] Werner Almesberger: Major redesign of "cntr" circuit. http://qi-hw.com/p/ben-wpan/e9147b007:45
viriclarsc: did you ever have kernel hang problems if you connect the USB in the middle of the kernel boot? (before mounting the rootfs)08:14
larscviric: i don't think i ever tried that08:15
viricok08:17
viric2.6.35 hangs here, every time I try08:17
viric(even the cursor stops blinking)08:18
dandondoes anyone of you know of a website that indexes reviews of Intel ixxx M processors?09:09
dandonforum, blog w/e09:09
B_Lizzardlarsc, it happened again but I don't have my System.Map :/09:13
B_LizzardI think it happens when you remove and reinsert the USB cable09:15
larscB_Lizzard: the same error?09:18
B_LizzardYeap09:18
B_LizzardBacktrace from IRQ185 and voltage timeout09:18
larscand your kernel was compiled with kallsyms disabeld?09:19
B_LizzardYeah09:19
B_LizzardI'd have to rebuild with kallsyms again?09:20
lekernelhttp://www.lucidscience.com/gal-rock%20disaggregator-1.aspx09:20
B_LizzardI thought it might've been my build because I tried crashing the driver yesterday09:20
B_LizzardBut this happened with a different build09:20
B_LizzardThis time I'll test more thoroughly09:20
B_LizzardMaybe it happens randomly, maybe not09:21
B_LizzardI removed my AC adaptor to move it to a different plug and the driver crashed09:23
B_LizzardCould be related, I'll try in any case09:23
B_LizzardIs the screen supposed to blank to white?09:25
larscnope09:28
B_LizzardAh, it does09:29
B_LizzardWhen left alone for a while09:29
B_LizzardNot via /sys/class/lcd/ili8960-lcd/power-something09:30
B_LizzardUh, the sys thing actually powers the screen down so I wouldn't know either way, would I? :)09:30
B_Lizzardlarsc, it happened09:40
B_LizzardWait, I'll upload the stuff09:40
B_LizzardI can replicate it if I remove and reinsert the AC adapter with the usb part plugged in09:41
B_Lizzarddmesg: http://pastebin.com/JbzyC4EH09:42
B_Lizzardinterrupts: http://pastebin.com/NRPE4X1N09:43
B_Lizzardsystem.map: http://pastebin.com/pGndKKSZ09:44
larscthanks09:46
B_Lizzardnp09:48
B_LizzardLet's hope this stuff is useful09:51
larscyes it is. it seems as if two jz_battery_read_voltage are running concurrently, otherwise i wouldn't know how to explain an unbalanced enable_irq09:54
larscalthough i cant see yet what could cause that09:58
B_LizzardIf you can somehow combine that with the fact I can replicate it by removing and reinserting my AC adaptor with the USB part plugged in.09:59
B_LizzardNot the other way around.10:00
B_LizzardRemoving and reinserting the USB cable doesn't do anything10:00
B_LizzardLemme check something10:01
B_LizzardActually, I just have to remove the AC adaptor10:05
B_LizzardDifferent AC adaptors don't seem to change anything10:05
B_LizzardMaybe it has something to do with the gradual tapering of voltage by the capacitors or something like that?10:05
larsccan you try and see if you can reproduce the problem by doing a (cd /sys/class/power/battery; cat voltage health)?10:07
B_LizzardOK10:07
B_LizzardDo you mean /sys/class/power_supply/battery/voltage_now?10:10
B_LizzardI get 400131210:10
larscyes10:10
B_LizzardAC plugged10:10
B_LizzardLemme unplug the USB cable10:10
larscno wait10:10
larsccat /sys/class/power_supply/battery/voltage_now & cat /sys/class/power_supply/battery/voltage_now;10:11
B_LizzardWell it doesn't fluctuate with the AC adaptor plugged in10:11
B_LizzardI unplug the USB cable, it starts from 3961593 and goes down from there10:11
B_LizzardAs expected10:11
larscyes. but what happens if you open the file twice at the same time?10:12
B_LizzardAh10:13
B_LizzardWait a sec10:13
larsci guess thats the source of the problem10:14
B_LizzardYeap10:14
larscok. easy to fix. add a mutex around the code in jz_batter_read_voltage10:15
B_LizzardThree cat commands did the trick10:15
B_LizzardMaybe two would too10:15
B_LizzardCould you fix the blank to white too?10:16
B_Lizzard:/10:16
B_LizzardUgh, I know fixing this stuff is hard, sorry about that.10:20
B_LizzardDidn't mean to sound like an ass10:20
larscyou are using the upstream kernel, right?10:23
B_LizzardYeap10:24
B_Lizzard2.6.32 some of your patches10:24
B_Lizzard*with some of your patches10:25
B_LizzardAnd your defconfig10:25
larschm looks as if i forgot the patch that fixes the whitescreen to the qi 2.6.36 tree10:27
B_LizzardAh, OK.10:27
B_LizzardOne other issue, bringing the volume down to 0% for Master in alsamixer doesn't actually mute the channel10:28
B_LizzardActually, the changes between 100% and 0% aren't huge10:29
bartbesdoes the openwrt build chain support hg?10:29
larscbartbes: i think it does10:30
larscB_Lizzard: i know. but that is all the hardware supports10:31
B_LizzardAh, I see.10:32
B_LizzardI think that software volume would require dmix which might be prohibitive10:32
larscB_Lizzard: here is the patch that should fix the whitescreen bug https://dev.openwrt.org/browser/trunk/target/linux/xburst/patches-2.6.36/420-fb-notifier-pre-post.patch10:32
B_LizzardThanks, sir.10:32
larsci'll add it to the qi 2.6.36 repo later when i'm home10:33
B_LizzardGreat, thanks.10:36
B_LizzardThe battery thing too?10:37
larscyes10:37
B_LizzardThanks, larsc.10:42
larscthanks for testing10:47
wpwraklekernel: (high voltage generator) nice. now i know what to use if we ever need a range extender for wpan ;-)11:05
B_Lizzardhttps://dev.openwrt.org/browser/trunk/target/linux/xburst/patches-2.6.36/500-modifier-keys.patch11:33
B_LizzardWould this patch add proper keyboard support under the console?11:33
larscyes11:34
B_LizzardOK, thanks.11:35
wpwrakrafa: [continuing here] then, anything that got flagged in OpenWRT as "patented". not sure if there's a list12:01
rafawpwrak: what is flagged in OpenWRT?12:14
wpwrakthe have a flag that indicated whether a package contains patented material12:15
wpwrakso that one can avoid that package12:15
rafaah.. for every package?12:15
wpwrakbut i don't know the details12:15
rafaah.. okey.. let me check12:15
wpwrakwell, as i understand it, it could be set on any package12:15
rafawpwrak: I hope that we will not get something like (OE repository)-(flagged packages)=openwrt repository :)12:17
wpwraknaw, don't worry. i think it's only very few12:18
wpwrakbesides, once you have a procedure for excluding in packages in place, and excluded the obvious ones, you may very well leave the detail search to others12:22
larsca hackish solution could be to create a no-patents package and let packages with patented technologies conflict with it12:26
wpwrakliberty.ipk ;-) it would then even make sense in a philosophical sense ...12:32
rafalarsc: yes, but we need still to know which are the packages with patented technologies12:40
larscthats true12:48
kristianpaulwich kind of patented techonology are you lookign for besides multimedia codecs?13:53
kristianpauli guess ubutu and fedora guys managed that a bit13:59
qbjectHey, all. Has anyone else had trouble installing hnb?15:31
qbjectI get "* deb_extract: hnb_1.9.17_xburst.ipk: invalid magic"15:34
qbjectDrat. I hoped it was maybe my FAT32 uSD mucking up the .ipk so I downloaded the .gz and didn't unzip it until I had copied it to the Ben's NAND. No change.16:08
qbjectCould this have anything to do with the fact that tar has been hanging?16:20
kristoffersounds more like corrupted package/binary16:27
qbjectkristoffer: Hrm. I've downloaded twice from the link in qi-hw.com. Could WinXP/FAT32 screw up a .ipk.gz?16:31
qbjectLike I said, I'm not unzipping it until it's on the Ben.16:31
kristofferfat shouldnt mess it up like that. At worst it would simply shorten the filename16:32
kristofferyou sure the package is working for someone else?16:32
qbjectkristoffer: no, I'm not. I just know that others are using hnb on their Bens, so I assumed.16:44
qbjectI'll ping jirka about it.16:46
kristofferqbject, roger16:47
qbjectkristoffer: you were right - corrupted file. didn't even have to bug jirka.17:41
qbjectChrome browser on Windows XP appends .gz to the filename (I didn't realize that Chrome was doing it) so I ran the hnb installer through gunzip before trying to install, which I think is what was corrupting it.17:42
qbjectI downloaded the installer with Firefox instead, and FF didn't muck it up. Installed just fine.17:43
qbjectsoo, thanks for listening. =D17:43
kristofferah, that explains it. good catch17:45
qi-bot[commit] Werner Almesberger: Cleaned up board outline and ground routing. http://qi-hw.com/p/ben-wpan/2ad54bf17:49
qi-bot[commit] Werner Almesberger: Use "standard" 2:1 pad to hole ratio for USB connector shield. http://qi-hw.com/p/ben-wpan/8cd7acd17:49
qi-bot[commit] Werner Almesberger: Corrected MMCX connector fields. http://qi-hw.com/p/ben-wpan/fca53a317:49
qi-bot[commit] Werner Almesberger: Added cntr's MMCX jack, USB A jack, and NPN transistor to BOM. http://qi-hw.com/p/ben-wpan/86b116317:49
rafaqbject: just a question.. that package you downloaded is not on some repository ? so you can use opkg install package.ipk from your nn?.. I am trying to understand how you get packages18:05
qbjectrafa: correct. I do not have my Ben set up to go online, so I download .ipk files from various places like http://fast10.vsb.cz/brozovsky/data/ports/ and http://wejp.k.vu/projects/gmu and http://downloads.qi-hardware.com/people/zear/games/18:10
qbjectI download those files to a desktop computer, then copy to one of my microSD cards with a µSD-USB adapter that came with one of them.18:11
qbjectthen copy to my Ben from the µSD card.18:11
qbjectOnce an .ipk is on the Ben (I usually just toss them in / and delete when I'm done) it can be installed with, for instance, "opkg install hnb_1.9.17_xburst.ipk"18:13
qbjectrafa: does that make sense?18:14
qbject(oops. scratch that last link to zear's stuff. I'm pretty sure those are static binaries that you just unpack and run.)18:16
rafaqbject: yes I understand.. a little hard to get packages :( but well, you are having fun it seems ;) (games, gmu, etc :) )18:21
qbjectrafa: I try to use my computers for as many things as I can. =D Why do you have trouble getting packages?18:30
rafaqbject: me?, I do not have problems, or what do you mean?18:37
qbjectrafa: I'm sorry, I misunderstood when you said "a little hard to get packages :("18:40
rafaqbject: ah.. that was for you :)18:43
rafabecause it seems a little hard (download on pc -> copy SD using adapter -> copy ben -> try install)18:44
qbjectAhh. Yes, it is true. But that is okay. It keeps me out of trouble when I play with the NN at work (can't concentrate on it for too long), and then I don't get distracted by emails when I should be playing games!18:45
rafaqbject: you are right. If mails are not letting you to play games then something is wrong there.. ;)18:47
qbjectrafa: I'm glad you understand. :) I find that the Internet is enormously valuable as a tool, but if it's still hooked up when it's not needed, it becomes a monumental waste of time.18:50
wolfspraulwpwrak: hmm. Yanjun Luo (the guy who designed the board) says R16 doesn't need 250V.22:08
wolfspraulI don't know why and I doubt he will come here into irc to discuss it :-)22:08
wolfspraulmeanwhile Sebastien found a serious bug on the board, C2, C4, R3 and R4 need to be removed, otherwise USB high speed won't work22:09
kristianpaul250V? >:|22:13
kristianpaulhey stardict got ranked in osalt.com as alternative to Babylon http://www.osalt.com/stardict22:21
rafakristianpaul: what is osalt.com?22:35
kristianpaulrafa: just other site to post floss alternatives to propietary sofware22:46
--- Tue Oct 26 201000:00

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