#qi-hardware IRC log for Wednesday, 2010-09-15

qwebirc85021Hi, I've just installed alsa-utils alsa-base etc in debian but when I try to use the sound system I get cannot open mixer: No such file or directory, any ideas03:42
qwebirc85021this is on the bnn btw03:42
mthis the program trying to use the alsa or the oss mixer?03:53
mthyou can check with strace which device node it tries to open03:53
qwebirc85021well I was trying to use alsamixer so I guess alsa03:57
rafakristianpaul: still traveling03:57
qwebirc85021do I need to copy the modules from the official release?  I have debian on the nand03:57
rafakristianpaul: i will upload the stuff when at hoome03:57
rafakristianpaul: also i think that we will add the wikireader plugin for the gtk1 dictionary in beta4 (without the wikipedia dump, the user will need to download that file)03:58
mthit depends on whether the kernel you're using has alsa in modules or built in03:58
mthdoes /proc/asound exist?03:58
rafaqwebirc85021: aplay -l  should tell you if you have some sound system in your linux. You can check that first04:00
qwebirc85021aplay: device_list:235 no soundcards found. . .04:01
rafaqwebirc85021: then you need to check your kernel .. which kernel are you using?04:01
rafaqwebirc85021: do you have extra modules which you can insmod?04:02
mthif there is no /proc/asound, the alsa drivers are not loaded04:03
rafaqwebirc85021: so before to try the alsa tools, etc.. you need to be sure that sound is built inside the kernel or as modules, AND (important) you have the /dev/snd/ /dev/.. etc sound device files created04:03
qwebirc85021I'm using the kernal from the tutorial to install for the adruino sketch04:04
mthrafa: I think in most distro's the modules get autocreated by the hotplug daemon when the drivers are loaded04:04
qwebirc85021there is no /dev/snd04:06
qwebirc85021I don't have connectivity on my nanonote right now but later I was going to do a apt-get dist-upgrade I wonder if that will install the necessary stuff04:07
rafamth: modules? or you mean device files>04:09
qwebirc85021oh whell in the directory /dev/ there is no file or directory called snd04:09
mthrafa: sorry, yes device nodes are created04:09
qwebirc85021for those of you whose modules are built into the kernal where did you get your kernal04:11
qwebirc85021I mean who have04:11
qwebirc85021alsa built into the kernal04:11
qwebirc85021I have to go because I am at internet cafe and have to go home before weather gets too nasty because I forgot my coat today04:12
qwebirc85021you have all given me some good ideas and things to reaearch04:13
rafaqwebirc85021: i have sound modules04:13
rafaqwebirc85021: if you wantget back i can tell you how to build the kernel only if you need that04:13
qwebirc85021oh yes I think thats what I need to do04:13
qwebirc85021is there a wiki entry I can look at about building the kernal?04:16
qwebirc85021oh Really gotta go now,  I'll be back another time thanks for your help04:17
rafaqwebirc85021: http://jlime.com/downloads/releases/muffinman/documentation/kernel-nn.txt04:17
rafaqwebirc85021: i have some notes more for jlime.. but those are general .. so you can use any toolchain04:17
qwebirc85021I will track you down if I I have any more qs04:18
rafaqwebirc85021: no problem04:18
wolfspraulxiangfu: sometimes my builds are failing because a source tarball cannot be downloaded04:44
wolfspraulI noticed OpenWrt uses a mirror system, but that won't help for the packages in the openwrt-packages respository04:45
xiangfuwolfspraul: yes. because you are in China.04:45
wolfspraulso I suggest this: we add a file scripts/localmirrors to openwrt-xburst, and point it to http://downloads.qi-hardware.com/sources04:45
wolfspraulxiangfu: no, not china. I also have this problem on the buildhost machine in Bavaria :-)04:45
wolfspraulthe problem is that for all new applications we are bringing into openwrt, the ones in openwrt-packages.git on the qi server, there is no functioning mirror04:46
wolfspraulso my plan is to upload all sources from the dl/ folder into downloads.qi-hardware.com/sources (same as OpenWrt)04:46
wolfsprauland then add a scripts/localmirrors file pointing to that URL04:46
wolfspraulwhat do you think? does this sound good?04:46
wolfspraulwell we have to be careful that it is in line with the OpenWrt policies, general free software ideas and the GPL04:49
wolfspraulbut I think it should be OK. if I misunderstand something in the mirror system let me know, I'll set it up like this for now04:49
wolfspraulxiangfu: hmm. maybe localmirrors is not good because it will always try that first, thinking 'local' == 'fastest'04:51
wolfspraulwell there is no other way to get it in, unless we patch download.pl04:52
wolfspraulI will try scripts/localmirrors for now, I think the mirrors are only invoked if the true upstream URL fails, so might as well try the Qi server first...04:53
xiangfuwolfspraul: I found there is a "LOCALMIRROR" in 'menuconfig'04:59
xiangfuwolfspraul: yes. seems it will alwyas try localmirror first.04:59
DocScrutinizerwell, isn't that the primary purpose of local mirrors?06:41
DocScrutinizerplus I seem to remember you have to rsync on a regular schedule06:42
DocScrutinizerweren't there extensive discussions about that on #openmoko-cdevel, about how to mirror the bearstech servers as they frequently were down?06:44
tuxbrain2countdown for NanoNote Nanoware Especial edition, 1h 15m left for launching.07:20
tuxbrain2you can ejoy 20mins demo video here http://en.qi-hardware.com/wiki/File:Walktrhough_QHBNNNWTEE.ogv07:21
kristianpaulrafa: sure i have lots of dumps here :)07:23
Action: jluis looking the video :)07:37
lekerneltuxbrain2, nice :)08:08
lekernelgood that it has a GUI in the end, even if Xorg-based08:09
tuxbrain2hehehe, thanks lekernel :) comming from you this words are music to my ears :)08:11
tuxbrain2count down 20 mins08:12
tuxbrain2Tuxbrain<->Nanowar in sync for the launching08:13
virictuxbrain2: felicitats! T'ha quedat de collons09:16
tuxbrain2gracies :)09:17
kristianpaultuxbrain2: bien  por el video :)09:39
tuxbrain2must leave! spread spread10:31
aisabartbes: are you online?  I want to ask you about the wordgrinder port in qi-packages.10:42
bartbesepic timing10:42
bartbesyes, I am10:42
bartbesor am I?10:42
aisaI see wordgrinder marked broken for 30 days.10:42
bartbesoh heh10:42
aisathe commit message says waiting for an upstream patch to curses.10:42
aisacan you give me the backstory?10:42
bartbeswell, that upstream patch is there, I believe10:42
bartbesbut in some mysterious way my patchfile broke10:43
bartbesand... I got distracted10:43
aisaMay I offer to work on this port for you?10:43
bartbeswell, why not10:43
bartbesyou only need to fix the patch file10:43
bartbesand then it runs10:43
bartbesthe menus themselves would need to be optimized10:43
bartbesbut otherwise, it works10:44
aisathey are too large on this device?10:44
bartbesit has been on my nn for a month now ;)10:44
bartbesthough maybe the new console font fixes a lot of that10:44
aisacan you tell me what problem you had that caused you to mark it broken?10:44
aisaso I can be aware if I see it.10:44
bartbeswell, I was waiting for that upstream patch10:44
bartbeswhich was applied about 3 days later, I believe10:44
aisaI'm running an old image, I may not have this fix.10:45
bartbesthe fix is in one of the packages10:45
aisaI won't be able to update my image until next week, alas.10:45
aisaoh, better.  Which one?  curses?10:45
bartbesncursesw was needed10:45
bartbesso I did some poking around10:46
bartbesgot it to build10:46
bartbesworked on it with one of the openwrt guys10:46
bartbestil it integrated nicely with ncurses itself10:46
bartbesand then he committed it, though delayed10:46
bartbesiirc, that is10:46
bartbesif you want I can look into the patch file myself10:48
bartbes(basically I'd just create a new one)10:48
aisaI started already.  I remember reviewing the patch and seeing it was reasonably simple.10:48
bartbesit sure is10:49
bartbesI just needed to change the build configuration a little10:49
aisaAfter today I'm going to be away from my computer for four days.  I'll only have my NanoNote, and I'd like to have wordgrinder so I can experiment with writing on the NanoNote.10:50
aisaSo hearing that it is a simple fix, I'm going to work on fixing it today to meet my deadline.10:50
aisalook!  A patch failure!10:50
aisaI know I can run package/$PACKAGE/prepare10:50
aisabut I sure would like to split that into untar/patch10:51
aisaso I can do the patch stage myself at times like this...10:51
bartbesyeah, I know the patch failed10:51
bartbesthat's what I've been telling you10:51
bartbesalso, I saw no obvious reason why10:51
bartbesI succeeded!10:52
bartbesalso, I managed to make someone upset10:52
bartbestroll bonus points for me10:52
aisaDeadBob wins the subtle troll award...10:53
aisaOr the idiot award, it is never clear.10:53
aisabartbes: I think someone updated wordgrinder-0.3.3 without bumping the release number.10:58
aisaDoes openwrt track md5/sha1 sums on upstream data?10:58
bartbesthat may very well be the case10:58
bartbesbut that guy should be kicked in the teeth10:58
bartbeswait, what?10:58
aisaSince teeth kicking is illegal in some places and inconvenient most others, I propose instead we verify incoming data.10:59
aisaif we have an upstream tarball, compute the md5/sha1 and stick that in the Makefile.10:59
bartbeswell, yeah10:59
bartbesit's standard practice10:59
aisaand then compare the download in the future.10:59
bartbesbut since it was broken I didn't bother to put it in yet10:59
aisaah, wonderful.11:00
aisathe number of context lines in your udiff is unusually large.11:01
aisaDo you set a high default, or are you using a diff tool different from the diff tool I'm using?11:01
bartbesI just use diff -Naur11:01
bartbesthough that might have been for something else11:01
bartbesdude, you don't want to rely on my memory ;)11:02
aisaHaha, developing my own memory is why I bought a NanoNote:11:02
aisayou need not read that if you're not engaged, I found your comment funny so provide context.11:02
bartbesI can only reply with11:03
bartbesI read something about oral11:03
bartbesso now I have reason to derail this entire channel11:03
bartbesyou should've seen what I did to #omg!ubuntu! yesterday11:04
bartbesI took over the entire channel11:05
aisaIn theory my extreme focus on being serious prevents me from giving you support with that much information.11:06
aisaAt the sime time, I'm sure it was fun.11:07
bartbesit sure was11:08
aisaI've updated your pmfile patch on wordgrinder, it applies cleanly again, and I'm running the build now.11:11
qi-bot[commit] Alan Post: update the wordgrinder-0.3.3 patch. http://qi-hw.com/p/openwrt-packages/77694b311:35
bartbesaisa: I guess that means it works?11:37
aisaI haven't run it on my device yet, doing that now.11:38
aisaIt compiles :-)11:38
bartbesalso, how easy is it to run a gp2x game on the nn?11:40
aisabartbes: I get a segfault somewhere initializing the terminal.11:48
bartbesdid you install the new ncurses package?11:48
aisait leaves the terminal a mess, so I have to log out.11:48
aisaI installed the one I built, libncurses5w.11:48
bartbes(and ncursesw)11:48
bartbesI mean plain ncurses too11:48
aisaI probably didn't get that one.11:49
bartbes(it built too, because it is a dep of ncursesw)11:49
aisaok, libncurses updated from 5.7-2 to 5.7-3, same error.11:50
aisathough I should perhaps try rebooting the machine...11:50
aisabartbes: it is the same, alas.11:52
bartbesI guess I should rebuild and see for myself then11:52
aisaThank you, I would appreciate knowing how it behaves for you.11:52
bartbeswell, if upstream did in fact change it in place11:53
bartbesthe changes could be the thing killing it11:54
aisaLooking at the differences between your patch and mine,11:54
aisathere are only whitespace changes.11:54
aisathe reason your patch wasn't applying cleanly was difference in whitespace in context lines.11:54
bartbesI got that from reading the diff, yes11:54
aisaIt doesn't prove this isn't true, but I do wonder if that was why it was updated in place,11:55
aisato fix source code problems that didn't show up in the executable.11:55
bartbes(not to be taken seriously)11:57
bartbesit would be weird, yes11:57
aisait is a guess, and not a particularly useful one.11:57
bartbeswell, it's compiling11:59
bartbesor well, preparing to compile11:59
bartbeswhich always seems like that's the thing taking the longest11:59
aisaI'm glad it is one step further than it was yesterday.11:59
bartbes... I forgot to select it11:59
aisaha!  right.  That annoys me, that you say "do this to this package" and it comes back and says "you couldn't possibly mean that..."11:59
bartbes"WARNING: Totally ignoring what you said"12:00
bartbesnow ehm12:00
bartbeswhere in the menu was it12:00
aisautilities => editor12:00
aisaand remember lua too, in languages=>lua12:01
bartbesnot.. there12:01
aisa./scripts/feeds install -a ?12:01
bartbesyeah, i guess that is needed then12:01
bartbesaisa: gtg, sorry12:08
aisaThank you for your help today, I really appreciate it.12:08
bartbesthank you for allowing me to be productive without doing anything ;)12:09
wpwrak tuxbrain_away: heya ! i see that you got the counterweights ! :) one thing you want to check is the plastic glued on them. i'm not entirely sure of the glue i used. if any fell off, just glue them back on.13:00
rafaamaizing stuff and band15:55
kristianpauland jlime :)16:00
rafayeh, hehe but it seems that the nanowar has a really complete multimedia stuff of the band16:02
rafaso it has a lot of hours of fun16:03
bartbesaisa: did I just read you have a problem with cmake?17:29
aisain a manner of speaking, yes.17:29
bartbeswell, I doubt I'm using any new features17:29
bartbesso please try lowering the version17:29
aisaI am running debian 5.0, which includes cmake 2.617:29
aisaand nlove asks for 2.8.17:29
aisaoh, wonderful, I will try this.17:29
aisaIt would be personally convenient to me if we could use a stable debian as a build platform,17:29
aisabut I realize this gets harder to do as the years go by.17:30
aisayears between releases, that is.17:30
bartbesthe biggest problem I have with 'the debian way' is that you always get stuck with old packages17:30
bartbeswhich is especially annoying when talking about stuff like firefox17:30
bartbes(or chromium, which is developing even faster)17:31
aisaI strongly suspect I'll be running testing because of this,17:32
aisaI can't imagine a group of package maintainers settling for debian's release schedule.17:32
elricsfatehey guys20:07
elricsfatetrying to reflash my ben using usb boot and I run to a problem on the last step20:07
elricsfatenprog 2048 openwrt-xburst-qi_lb60-root.ubi 0 0 -n20:07
elricsfatethat command gives me an error20:07
elricsfateUsage: nprog (1) (2) (3) (4) (5)  (1)start page number  (2)image file name  (3)device index number  (4)flash index number  (5) image type must be:  -n:no oob  -o:with oob no ecc  -e:with oob and ecc20:07
elricsfatecan someone PLEASE help??20:07
elricsfateanyone? :/20:13
rafaelricsfate: please20:14
rafagive us the complete command tht you are trying20:15
rafaelricsfate: i guess that it should be :20:15
rafausbboot -c "nprog 2048 openwrt-xburst-qi_lb60-root.ubi 0 0 -n"20:16
rafaare you using "?20:16
rafa"  "20:17
elricsfatejust noticed that20:50
elricsfateworking now20:50
elricsfateany OS suggestions for the nano note? Thinking about switching over to Debian20:53
wpwrakrafa: the problem with usbboot is that there seems to be a bug in the parser. i got the same error. i worked around it by renaming the file. not sure if this is fixed already. the usbboot i have is from march.22:22
wpwrakxiangfu: heya ! do you know who's in charge of usbboot ?22:28
xiangfuwpwrak: me. :)22:28
wpwrakgotcha ! ;-))22:28
wpwrakthere seems to be a problem with the parser. when someone follows the flash instructions, the last command, flashing the .ubi file, fails with22:29
wpwrak(looking up the error message ...)22:30
wpwrak"not enough argument."22:30
wpwrakinterestingly, it works after renaming the .ubi file, e.g., to "foo"22:31
wpwraknow, i used the may 5 version. i looked at the changelog, and i saw that there was some command line parser change later in may.22:32
wpwrakdo you think this solved also this issue ?22:32
wpwraki.e., are you using the process described on http://en.qi-hardware.com/wiki/Updating_Ben_with_usbboot yourself ?22:32
xiangfuwpwrak: I think  it solved this issue. I remember someone have report this issue to me. and we have fixed that in last version22:33
xiangfuwpwrak: can you test with : "xburst-tools_201007-1.tar.bz2"22:33
wpwrakhmm, that's the .deb i used22:34
wpwrak(or .tar, don't remember which one it was)22:35
wpwrakmaybe the precompiled packages need updating ?22:35
wpwrak"elricsfate" also ran into the same problem today22:35
wpwrakerr .. no, must have been the .deb that i used, obviously :)22:37
xiangfuwpwrak: ok. I will update the deb packge. when I have time today.22:39
wpwrakperfect. thanks !22:40
wpwrakah, and do you know who's in charge of gmenu2x ? that one is still messing with the clocks :-(22:42
xiangfuwpwrak: I don't know about gmen2x. seems no one in charge of it. :)23:08
wpwrakxiangfu: heh, that may explain a few things ;-)23:41
kristianpaulxiangfu: but you git it from upstream right?23:43
kristianpaulcause i noticed some improvements in the gmenu before the uclib reverse23:43
kristianpaulor may be they we're already there and i just discover it later..23:44
xiangfukristianpaul: yes. the gmenu2x is git clone from upstream.23:44
kristianpaulwpwrak: so you have somebody(ies) to ask finally :)23:44
wpwrakkristianpaul: hmm. then i need to find out who sent upstream patches to mess up the ben.23:58
wpwrakkristianpaul: these crude hacks are definitely ben-specific.23:59
--- Thu Sep 16 201000:00

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