=== David-T is now known as Guest16283 [01:50] this is the 3rd time an update to the Linux kernel messed up either Xorg or my video card driver. I filed a bug report for both of the first bugs and it took several months to fix it, although many people reported the same bug. and now, after the recent update, I had to reinstall the graphic card drivers and xorg is eating up CPU again. serious question: is it so hard to get it right before officially releasing an update and not beta test it on [01:50] the user base? [01:50] hello [01:51] any chance anyone in here can point me in a direction for help in porting drivers? [01:51] maybe to someone who is amaising at it [01:51] or maybe just a great developer [01:51] i promise there IS no port of this driver, im not just over looking something thats allready done [01:55] ravenxbishop1: The Kernel Driver Project. See http://www.linuxdriverproject.org/foswiki/bin/view [01:59] thanks i'll check it out [02:01] I'm not sure you can really "port" a driver though [02:01] oh that wont work tho [02:01] because theres no specs [02:01] and no way to get them [02:02] if you don't have the specs, how do you know how to write a driver? [02:02] well, its called reverse engineering [02:02] so do that [02:02] then write the spec as a result [02:02] then write the driver [02:02] nod i need someone who is good at it [02:02] i didnt say I WAS [02:03] What does *he* need *you* for? :-P [02:03] depends, he doesnt actually, but it could be something we both want to do for any number of reasons [02:03] why so much animosity? [02:04] there's no animosity [02:04] well ion sounded... uhm [02:04] mean [02:04] i just want some help, so i can use linux [02:04] sure, but writing a driver is *hard* [02:04] and i dont really know where to start [02:04] sure it is [02:05] the people who can do it are already hard at work writing drivers [02:05] and it's still hard even if you've got the full specs [02:05] oh, so im out of luck? [02:05] and i cant get any help [02:05] and i should just give up? [02:05] what's the device? [02:05] It’s easy to interpret tones incorrectly on IRC. I wasn’t being hostile. [02:05] its a sound card [02:05] a professional sound card [02:06] and there's no extant driver for the sound card? [02:06] your best bet is to find someone in the ALSA project [02:06] bribing them by sending them the sound card and letting them keep it often works [02:06] i see no indication that there is any way to get it working in linux [02:06] especially if it's a good one [02:07] well then i wouldnt have it :) [02:07] if you gave us the model name/number/etc. I could actually look it up for you [02:07] AArdvark q10 [02:07] ravenxbishop1: they can't write drivers blind ;) [02:07] of course not [02:07] i didnt think they could [02:08] that's what ion meant [02:08] unless there's something in it for them (e.g. getting a neat sound card) a developer is unlikely to be interested [02:09] since they're probably doing linux in their spare time [02:09] but that cant be negotiated with you [02:09] becuase you cant do the job [02:09] i didnt ask anyone here, to do it [02:09] i asked to be directed to someone who might be able to [02:09] and instead [02:10] sure, and we've told you the two groups ;) [02:10] Linux Driver Project and the ALSA Project [02:10] i saw one [02:10] ah [02:10] alsa [02:10] so they work without specs? [02:10] no [02:10] because the first says "dont bother asking" [02:10] they *might* work without specs and with the card [02:10] (ALSA might) [02:10] but you're basically asking for a year or two's work [02:10] well, maybe [02:11] typically speaking, if i can get in touch with the right person [02:11] i can make things happen [02:11] another good bet is to speak to the manufacturer and try and persuade them that it's in their interests to assist in writing a linux driver [02:11] e.g. by giving the LDP the specs [02:11] thats impos [02:11] why? [02:11] because it isnt something that can be done [02:11] we've got specs out of *NVIDIA*, anything is possible [02:12] of course it is [02:12] well, no, its not [02:12] i can get specs from nvidia [02:12] but i cant get them from aardvark [02:12] if everyone returned their Q10 cards, because they didn't have linux drivers, and nobody bought them, then they'd pony up the specs pretty quick :p [02:12] well [02:12] If you can get in touch with the right person with Aardvark, you can make things happen. :-P [02:12] there would be no where to return them to [02:12] so the company who makes the device has gone bust? [02:12] I think the Q10 is already a discontinued product... ? [02:13] sort of went bust [02:13] being realistic, I suspect your chances of getting a driver are basically zero [02:13] especially if it's not a production device anymore [02:13] the company is closed. [02:13] theres NO one to contact [02:13] and no place to contact them [02:14] they didnt fail they were forced out by competition [02:14] through an unrelated legal issue with a board member [02:14] the competition probably had Linux drivers ;) [02:14] no, you dont understand [02:14] its alot of irrelivant things to explain [02:14] before you would [02:14] but something unrelated to the product, closed the company [02:15] Bottom line: They aren't likely to sell many more Q10 cards... so the work to create a Linux driver is out of all proportion to the benefit. Buy a new card. [02:15] there IS no new card [02:15] it does things other things CANT do [02:15] From a different company. [02:16] if i spent $10,000 i could get something similer, using more space that still wasnt as good [02:17] would you pay someone $10,000 to write a driver for your card? [02:17] hrm. [02:17] hard to say [02:18] though, honestly, I suspect the rate would be more like $100,000 [02:18] Old ads for it do not seem to proclaim that it does things only $10K cards can do... makes me wonder what these things are and where they are documented... [02:19] it was a $1k card [02:19] it contains a word clock thats ALSO worth $1k [02:19] but it is a clearer card than 10k cards [02:20] first try to find a 10 in 10 out card with 8 preamps in BOTH 1/4 inch and xlr. [02:20] once you do. [02:21] find one that has midi too [02:21] and a word clock generator [02:21] not just having word clock [02:21] but having a generator for it [02:21] you still wont have as GOOD of a word clock generator because there ISNT a better one [02:21] its EXTREMELY low latency [02:22] i get 1ms [02:22] sounds like it's missing a key feature though - a driver [02:22] it has drivers. [02:22] just not one for linux [02:22] has osx and windows [02:22] Does it have one for FreeBSD? or OpenSolaris? [02:22] shake, osx and windows [02:23] and only 32 bit windows [02:23] Other professional audio interface manufacturers just don’t bother to use good enough components? [02:23] correct [02:24] the guy that invented it, is a nasa guy, who is just better than everyone else [02:24] igor levin [02:24] and so far i'm yet to get any contact with him [02:24] tho i've had ... really important people try [02:24] people just dont FIND him [02:25] for instance... we've called "antalope audio" quite a bit, no one ever answers the phone the website stays up, but i cant imagine they do much business when they never pick up [02:27] if i had one of the cards sooner i probably would have GOTTEN the source when they closed down [02:27] i have a tendancy to do things like that [02:28] I guess the fact i demand less from my gear makes my life easier. :-) [02:28] it seems like walking me through a clean room reverse engineering process wouldnt take too much time or resources. but i'd have to know the right linux developer. it doesnt seem like THAT person hangs out in here [02:28] thats probably true [02:29] maudio has linux drivers but their cards are shit [02:29] ravenxbishop1: usually the approach is debug the windows driver [02:29] send commands, intercept the pci stream, then figure out what it's actually doing [02:29] replicating it in your own driver [02:29] nod keybuk... but do you have experience? [02:30] I've done it before, not for a sound card though [02:30] (and I'm not interested, btw :p) [02:30] because you arent really the guy for the job. [02:30] but someone is [02:32] my guess from years of practical experiance is... that you probably havent REALLY done it, that you've done parts of the actual process and feel you know the theory. i could be wrong. but someone around has done it a few times and would probably help [02:32] infact the guys that do it alot, do it for fun [02:32] usually the people asking for money or saying how hard it is... arent really the guys that know how [02:33] like the friend of a friend of mine that just cracked the ps3. [02:33] he doesnt charge :) [02:34] he's the one for the job, when you start bringing up anything to do with it, you cant even shut him up [02:34] not that i expect anything for free. === vorian is now known as obama === obama is now known as vorian [03:01] are there any official ubuntu sparc tftp images? === Richie is now known as WelshDragon === Guest16283 is now known as David-T [10:30] hi all [10:30] I am new to ubuntu but I want to contribute as a developer [10:39] Liquid-Silence: hello. did you need something? [10:40] nope [10:41] just looking @ contributing to ubuntu as a developer [10:41] Liquid-Silence: have you seen this already? http://www.ubuntu.com/community/participate [10:41] yup [10:41] going through that and some other things atm [10:41] ah, cool [10:41] got launchpad etc.. [10:41] but more interested in what development tools I require etc... [10:43] Liquid-Silence: that will depend on the package you want to hack on [10:43] Well in general I am not going to focus on one package [10:43] I want to contribute in many ways [10:45] Liquid-Silence: Id say join #ubuntu-motu they have some help for people to get started [10:45] Liquid-Silence: I would just start with one and learn from there [10:45] !motu [10:45] motu is short for Masters of the Universe. The brave souls who maintain the packages in the Universe section of Ubuntu. See http://wiki.ubuntu.com/MOTU [10:46] I am in that channel [10:46] :) [10:46] ahh, so you are. best to start on the basics, I can assure you that there is more than enough work to go around. [10:50] jussi01: yip [11:41] chrisccoulson: The version of Transmission I pushed last night does work. I'm now working on creating a patch for the latest version. [11:42] qense - thanks. i will try and set aside some time later to look at it [11:42] chrisccoulson: thank you [11:44] i don't know why people keep changing the status and assignee on bug 452208 [11:44] Launchpad bug 452208 in devicekit-disks "devkit-disks-daemon crashed with SIGSEGV in dbus_g_method_return_error()" [High,Fix released] https://launchpad.net/bugs/452208 [11:44] it's getting ridiculous now [11:44] when you look through the history of it, it's just people having to constantly revert unnecessary changes [11:49] qense - which branch did you base your work on? [11:49] chrisccoulson: lp:ubuntu/lucid/transmission [11:49] and that still uses 1.80~b1 [11:49] ah, yes [11:49] whereas the repos have got 1.83 [11:49] the packaging branch we use is lp:~ubuntu-desktop/transmission/ubuntu [11:50] that one is always up-to-date [11:50] ok [11:50] it should be listed in debian/control [11:50] I'll merge with that branch [11:50] (if it's not, then i need to add that) [11:50] I'll have a look at it [11:50] cool, thanks :) [11:51] chrisccoulson: I can just merge that autoreconf patch into this branch, it just contains the directory. That's correct? [11:52] btw, yes, it contains the Vcs-bzr entry [11:52] qense - thats right. our packaging branches only contain the debian/ folder [11:52] but once you have the branch, you can type "bzr bd-do", and get the full source directory [11:52] ah [11:53] that's a nice feature I had never heard of [11:53] it's actually a bit of a pain, but it does save on bandwidth ;) [11:53] qense: the lp:ubuntu/transmission packaging branch might stay outdating for some time as the package importer failed over the .orig.tar.bz2 from Debian (and this might have stopped imports of Ubuntu uploads too) [11:54] geser: yeah, I saw that. A lot of imports failed recently. === elpargo_ is now known as elpargo [14:23] aloha [14:23] Hi [14:23] at fosdem great turn out, interesting talks, one talk this morning, ubuntu and debian was soo full they stopped allowing folks in [14:24] tweet came in afterwards [14:24] RT @rdicosmo: Debian and Ubuntu .... After the talk in H1309 at Fosdem, I think Debian really needs to push Ubuntu to behave as a proper dowstream!!!!! [14:24] er? [14:26] ho? [14:26] I can imagine that, I was at FOSDEM already twice (the last time last year) and the Debian dev room was always full [14:26] debian needs to push ubuntu to behave as a proper downstream? what's up with that? [14:26] have we not been a proper downstream? [14:27] I guess the opinion depends on who you ask [14:27] or are those rogue debian developers getting ahead of themselves claiming that ubuntu never contributes to debian again? [14:28] hyperair: it was one comment that just caught my eye on the #fosdem search on twitter [14:28] czajkowski: i see. [14:29] * hyperair sighs. i thought that whole war was behind us [14:29] you should just drag the original poster in middle of Ubuntu crew there (; [14:30] i agree [14:30] Tm_T: there are more debian here [14:30] 18 ubuntu folks went for dinner last night [14:30] czajkowski: doesn't matter (: [14:30] http://fosdem.org/2010/schedule/events/dist_debian_ubuntu was the speaker [14:30] Debian is to inhomogeneous to come just to *one* opinion on anything [14:33] czajkowski: no material available? ): [14:33] Tm_T: pardon? [14:34] i'm curious to see the actual speech [14:34] czajkowski: nothing, I cannot build reasonable sentences currently (: [14:34] hyperair: me too [14:35] it was wedged, looked interesting but it was just too warm to stay [14:35] * hyperair wants to see a video or hear a recording =\ [14:35] aye before commenting [14:36] i was just curious tbh, i'd have prefered to see a more larger presence from ubuntu tbh. [14:36] I guess you talk to lucas also when he is here again [14:36] his first slide was why he's qualified to give this talk, and said jokely it was being recorede so didnt want to offent anyone or something like that [14:36] lucas did the recent archive rebuilds for us (and is also a MOTU) [14:37] i dont think there was anyting wrong wiht his talk, the perception on stream just got me wondering [14:55] does Ubuntu plan to apply as an Org for GSoC'10? [14:56] what would you work on under gsoc for ubuntu? [14:58] there's the whole Atayana thing, and notify-osd needs a lot of revamping. [15:00] Ayatana* [15:01] notify-osd eh.. [15:01] i'd think that'd be under notify-osd instead of ubuntu. [15:03] notify-osd won't make much of an organisation... [15:32] hyperair: Debian's conclusion about Ubuntu at FOSDEM - http://short.ie/fh36jd [15:33] * hyperair clicks [15:34] czajkowski: that sounds fine to me. [15:34] * Daviey expected a rick roll [15:34] czajkowski: but how did that come to "ubuntu needs to be a good downstream"? [15:34] Daviey: that was the first thing that crossed my mind as well =p [15:34] hyperair: that was whats someone took from the talk.. i really need the recordings [15:35] =\ [15:35] Daviey: i'm not going near rickroll that song does my noggin in [15:36] :) [15:37] hyperair: talk should be up next week once proceessed [15:38] czajkowski: cool [15:59] For LTSP development, I want to disallow upstart daemons from running while maintaining the chroot. Does upstart support some kind of "flag" to disallow daemon starting, like policy-rc.d does for invoke-rc.d ? === yofel_ is now known as yofel [16:24] Hmmm ok I found this: https://bugs.launchpad.net/ubuntu/+source/upstart/+bug/430224 [16:24] Ubuntu bug 430224 in upstart "misc: packages cannot be upgraded in a chroot" [Medium,Confirmed] [16:25] I'll try to make a wrapper for initctl, that either runs initctl or runs true, depending on if we're on a live system or in the chroot... [16:35] arg [16:36] need to go sit and watch movies with the wife [16:42] lamont: is there a specific reason why we don't ship /usr/lib/libuuid.la in uuid-dev anymore? [16:58] mdeslaur: There's a general push in Debian not to ship .la files. I think it should be the other way around, don't ship it unless there's a reason to. [17:00] ScottK: do you know of any webpages with info on transitioning packages to use the .pc file? [17:00] mdeslaur: No. Sorry. I haven't had to deal with it yet, I've just read about the goal to remove them on debian-devel. [17:00] ScottK: ok, thanks === dendro-afk is now known as dendrobates [18:03] mdeslaur: because .la files are evil and should die [18:04] lamont: so, if I get a FTBFS on another package because of it, what's the proper thing to do? === micahg1 is now known as micahg [18:04] bbl [18:04] slangasek, are you around I have a quick question for ya? [18:05] fix the other package to not need the .la file - a large part of the push to get rid of the .la files was that they introduced amusing(tm) build-deps that were 4 and 5 deep and totally undeclared [18:05] gnome in hoary was _such_ fun === dendrobates is now known as dendro-afk === dendro-afk is now known as dendrobates [18:22] akgraner: hi [18:23] hey [18:28] * ScottK thought maybe slangasek was still on his trip home from the sprint. [18:28] ;-) [18:28] it's a tiring journey [18:39] heh === ikonia_ is now known as the_tester [19:19] hello, how does ubuntu handle virtualbox? Any specific kernel settings or videodrivers? I see ubuntu doesn't use guest additions etc. [19:22] nick19691: If nobody gets back to you in a while, you might try #ubuntu-virt [19:24] thanks persia [19:27] nick19691, there are two versions of virtualbox. the open source edition thats included in ubuntu and the closed source edition that you can download from the virtualbox website, that has guest additions and such, but is closed source [19:29] gord: guest additions is in ubuntu [19:29] micahg, i think there is a few that are kept out or something, i'm not 100% on the details [19:29] gord: USB support [19:30] gord: and shared folder support [19:33] uhm [19:33] the one you can install via apt has folder support [19:33] and 3d [19:33] and the guest additions [19:33] I am running it atm [19:37] gord: I was mostly talking about ubuntu as a guest as it doesn't freeze as eg debian does [22:32] ScottK: postfix 2.7.0~rc1-1~1 bits at deb http://ppa.launchpad.net/lamont/ppa/ubuntu lucid main [22:32] or will be once the publisher runs [22:33] ScottK: feedback welcome [22:33] lamont: THanks. [22:34] lamont: Unfortunately I supposed to leave town for a week tomorrow, so tossing experimental Postfix packages on even my test server (which I do use for some actual stuff) is probably not prudent. [22:34] I'll have a look when I get back (although it's no doubt going to be OBE by then) [22:35] it's running on my primary mail server, but then, I'm _in_ town this week [23:54] 00:36 * bremner learns from an Ubuntu user that he is the maintainer of syncevolution in Ubuntu. [23:54] and indeed, he's mentioned as "Maintainer" if you input syncevolution in launchpad.net/ubuntu's search box and click the result