#qi-hardware IRC log for Wednesday, 2011-11-30

wpwrakhow long does the RTC actually maintain time ?03:34
kristianpaulwell, seconds i guess03:48
wpwrakthere's a circuit that's supposed to power it for a bit: D1 and C20 (47 uF). but i'm not sure this does much good.03:51
kristianpaulbit? like the first palm m100 ;)03:52
wpwrakhmm, the diode has a reverse current of about 2 uA03:55
wpwrakhmm, simulation says that this should be sufficient for about 4 seconds04:01
wpwrakkristianpaul: your estimate wasn't so bad :)04:02
wpwrakactually, it's worse. since the diode drops some 200-300 mV. that's just about the voltage range the cap could buffer over. so it's ~1 second, worst-case.04:21
wpwrakmaybe in real life this translates to up to 10 s. perhaps that's what causes the ben to lock up if you do a very quick power cycle.04:22
rjeffrieshas this been discussed? has the right license it woudl appear. nice design http://teholabs.com/docs/procyon:overview04:41
qi-bot[commit] Werner Almesberger: m1/patches/rtems/: added Milkymist-specific USB-MIDI forwarding (master) http://qi-hw.com/p/wernermisc/a946aac09:47
wolfspraulkristianpaul: you are too fast! [spam] :-)11:25
wolfspraulthanks for helping11:25
kristianpaulhehe :-), just getting ready to go work :)11:27
kristianpaulbtw is taking so long the package with stickers/brochure to arrive11:37
kristianpaulbut i guess no rush, a package from Artyom took near 3 months to get here..11:38
wolfspraulwho knows11:38
kristianpaulindeed 11:38
kristianpaulmust go, read you later11:39
wolfspraulin the last 2.5 years that I've sent hundreds of packages I think I can remember loosing one, and that was to Colombia11:39
wolfspraulMirko dropped a letter (unregistered, untrackable) into the mail but it never arrived11:39
wolfspraulthat was the only case ever11:39
wolfspraulso let's see11:39
wolfspraulmaybe it arrives, maybe not11:39
wolfspraulif not, then we need to send another one...11:39
kristianpaullets wait11:40
kristianpaulwill be the first package i fail to receiver, hope not11:40
antoniodariushwpwrak, hi, if "dmesg | grep at86rf230" does not return anything in the nanonote, does this mean that the kernel is not configured for the atben?14:37
blogic"lsmod | grep at" ?14:39
antoniodariushblogic, nothing14:40
blogicthen i would assume that the image lacks atben support14:40
antoniodariusheven though I have these http://pastebin.com/5TnB5Fkk14:44
blogicnot sure14:44
wpwrakmaybe the module isn't loaded ?14:46
antoniodariushwhat u mean by loaded 14:47
blogicantoniodariush: find /lib/modules/ | grep at14:49
blogicare they even installed int he rfs ?14:50
wpwrakheh :)14:50
wpwrakand if they exist, lsmod will show is they're being used14:50
jivsI get this /lib/modules/3.0.0/at86rf230.ko14:59
jivs for the find command above14:59
jivsbut nothing for lsmod |grep at15:00
blogicthere will be some missing package then15:00
blogiccausing those packets to not be installed15:01
blogiccausing those packets to not be inserted15:01
jivsokay i will try to reflash by adding more packages as wpwrak suggested..15:02
blogicxiangfu: is this the latest http://patchwork.openwrt.org/patch/1439/ ?15:02
blogicah he is not online15:02
blogicdoes xiangfu have access to target/linux/xburst in the owrt repo ?15:03
blogicit seems weird that i need to merge this for him15:03
jivswpwrak, is IEEE802154_DRIVERS same as PACKAGE_kmod-ieee802154 . I don't see any package with ieee on its name in my toolchain15:06
wpwrakjivs: i don't know how those packages translate to kernel config options. but there is also a CONFIG_IEEE802154, so chances are that CONFIG_IEEE802154_DRIVERS and PACKAGE_kmod-ieee802154 are not the same15:34
blogicwpwrak: i asusme he uses this http://phrozen.org/wpan.mk15:35
blogici gave that link to xiangfu and i assume he integrated it to the qi tree15:35
jivsokay 15:36
wpwrakthis looks like a typo: CONFIG_IEEE802154_DRIVER vs. CONFIG_IEEE802154_DRIVERS15:36
wpwrakalso, no =y needed for CONFIG_MAC802154 ?15:37
Action: blogic looks15:37
blogicthat is correct15:38
blogicthe CONFIG_IEEE802154_DRIVER=y vs CONFIG_IEEE802154_DRIVERS=y  looks like the candidate for the borkage15:39
blogic(missing S at the end)15:39
blogicjivs: i am almost certain this is the case15:45
blogicthe dependency chain of at86rf230 is broken without that symbol15:45
blogicjivs: did you build the image yourself ?15:45
jivsyes blogic 15:45
antoniodariushso i guess we need to wait for xiangfu ?15:46
jivshow can we fix this on our image? if we can15:46
blogicyou need to rebuild the image15:46
blogicvi package/kernel/module/wpan.mk15:46
blogicgo to the line that has CONFIG_IEEE802154_DRIVER=y \ and add a S before the =y15:46
blogicthen do15:46
blogicrm -r tmp/ && make menuconfig      (save and exit to update the config15:47
jow_laptop(make defconfig)15:47
blogicmake package/kernel/clean && make V=9915:47
blogicjow_laptop: ok15:47
blogicrm -r tmp/ && make defconfig && make package/kernel/clean && make V=9915:47
blogicthat will build an image with the fix applied15:47
blogicbefore flashing it you can see if the file is now included15:48
blogicyou can do so by calling15:48
blogicls  build_dir/target-mips_r2_uClibc-0.9.32/root-xburst/lib/modules/15:49
jivsthank u blogic and jow_laptop , i am building new image now15:49
blogicthe path might be slightly different to build_dir/target-mips_r2_uClibc-0.9.32/root-xburst/lib/modules/15:49
blogicis the nanonote mips or mipsel ?15:49
blogicit will be  build_dir/target-mipsel_r2_uClibc-0.9.32/root-xburst/lib/modules/... in that case15:52
blogicthere will only be one folder of the format target-mipsel_r2_uClibc-0.9.32 so just use ls to check please15:52
Action: blogic might continue to integrate 802.15.4 in owrt tonight15:52
jivsthe folder now had these files af_802154.ko  crc-ccitt.ko  fakelb.ko      mac802154.ko  spidev.ko15:54
jivsat86rf230.ko  fakehard.ko   ieee802154.ko  spi_atben.ko  usbcore.ko15:54
blogiccheck 1 more thing15:54
jivs20-crc-ccitt  20-usb-core  90-ieee802154  91-mac802154  92-fakehard  92-fakelb  93-spi-dev15:55
blogicyou should corresponding files there that tell the system to insmod the drivers that will now be part of the image15:55
blogicok at86rf230 is missing15:56
blogiclet me look at wpan.mk 1 moment please15:56
blogicok i will need to look indetail15:58
blogicyou can flash the image but will need to manually insmod the modules atm15:58
blogici will ping you when i have a update wpan.mk15:58
blogici will build an image in a sec to play around15:58
jivsokay :-)15:59
qi-bot[commit] Werner Almesberger: m1nor: accept multiple file arguments; clarified diagnostics (master) http://qi-hw.com/p/wernermisc/037be5023:05
qi-bot[commit] Werner Almesberger: m1/BUILD-CHEAT-SHEET: updated for new RTEMS patch and local changes (master) http://qi-hw.com/p/wernermisc/9343b6123:11
--- Thu Dec 1 201100:00

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