**** BEGIN LOGGING AT Sat Dec 08 02:59:59 2012 Dec 08 10:37:53 infinity: I see you fixed tcl crossing. What did you do about multiarching the tclconfig.sh path? Dec 08 10:38:07 wookey_: I didn't touch tcl, just tk. Dec 08 10:38:11 Or did you just fix tcl bulding itself, not anything building against it? Dec 08 10:38:32 wookey_: Though, tk builds against tcl, so maybe someone already fixed the thing you're talking about. Dec 08 10:38:41 aww, I want to run 13.04 on the nexus, but I need Bluetooth... Dec 08 10:39:01 OK. That would be good if true - I've beenwondering what to do about it for months :-) Dec 08 10:39:30 I see you had a go at apr too - is there a bug for that? It was an abomination when we looked at it a couple of years back Dec 08 10:39:54 somebody tried to add cross-support but did it in a completely broken way Dec 08 10:40:02 wookey_: I'm holding off on filing bugs (or committing directly) to apr/apr-util/apache2 until I'm happy with my "solutions" (and I use that term loosely). Dec 08 10:40:37 wookey_: apr and apache2 work in Ubuntu now (apr-util needs some twiddling with libraries it links to), but neither is pretty, so I'll likely revisit. Dec 08 10:40:46 I have a really long mail for the apache people that we weren't allowed to send a couple of yrears back, asking what tey thought they were _trying_ to do Dec 08 10:41:04 so we could work out how to fix it properly Dec 08 10:41:24 http://launchpadlibrarian.net/124947530/apr_1.4.6-3_1.4.6-3ubuntu1.diff.gz Dec 08 10:41:36 ^-- The ugly is just a bunch of autoconf cache preseeding. Dec 08 10:41:43 But there's got to be a less hideous way. Dec 08 10:42:23 As for apache2 itself, it's all fixed upstream, I just had to backport patches. Dec 08 10:42:30 Ah cool Dec 08 10:42:33 (And fix debian/rules to be less retarded) Dec 08 10:43:30 So, I'll probably commit the apache2 stuff to Debian (or file a bug and make sf do it), but the apr/apr-utils things need some thought, I think. Dec 08 10:43:38 And, really, apache is useless without apr, so... Dec 08 12:04:49 mooo Dec 08 12:11:32 * Tassadar hides the gnomes Dec 08 12:12:43 i'm not here to gorge on gnomes today Dec 08 12:22:18 yargh Dec 08 12:22:26 what is the magic to get sound out of this little beastie Dec 08 12:24:50 You mean n7? Dec 08 12:28:21 yeah Dec 08 12:28:28 i have gotten it to hiss at me Dec 08 12:28:31 and thats about it Dec 08 12:28:37 yes - i know - suspend and resume Dec 08 12:28:40 doesnt do squat Dec 08 12:31:09 i've been swizzling everythng randomly in alsamixer for a while to try see what low-level knob makes it work Dec 08 13:22:29 nvidia just published tegra boot process description: ftp://download.nvidia.com/tegra-public-appnotes/tegra-boot-flow.html Dec 08 13:22:42 (for those who didn't got it yet) Dec 08 13:23:53 nv is getting better at this Dec 08 13:24:12 sadly, it's not much use on most 'real' devices (given they are locked mostly down), but it's a nice move either way Dec 08 13:26:03 not much nv can do there Dec 08 13:26:07 thats oems Dec 08 13:26:17 indeed Dec 08 13:26:29 * RaYmAn looks at google Dec 08 13:26:45 i have a working nexus7 Dec 08 13:26:47 minus sound Dec 08 13:26:57 i cant divine how to make it peep Dec 08 13:28:16 raster: it's just a pity e.g. google makes the n7 unlockable and then locks down the bootloader still. :/ Dec 08 13:29:00 well to some extent thats to avoid bricking Dec 08 13:29:32 oems are paranoid about having hoards of users send in bricked devices under warranty Dec 08 13:29:32 tegra is unbrickable if you have access to APX mode (e.g. tegrarcm) Dec 08 13:29:36 well, there is still the rom bootloader, which cannot be bricked by nature Dec 08 13:30:07 on the other hand, who cares if the locked bl can load an unlocked one? Dec 08 13:30:14 if a user cant boot it to a state of a ui withotu some cruptic pc tools Dec 08 13:30:15 then its a brick Dec 08 13:30:20 form an oem pov Dec 08 13:30:32 fastboot is not an answer in their eyes Dec 08 13:30:34 write a simple tools then Dec 08 13:30:49 it has to be unbrickable without tools Dec 08 13:30:57 ie withotu needint to plug it into a pc and "use tools" Dec 08 13:31:11 same with plugging in special mucro-sd cards with rescue images etc. Dec 08 13:31:22 but how is that different from now? users could easily perceive it as bricked even if they have fastboot. Dec 08 13:31:27 and that is possible now Dec 08 13:32:00 dont they stil lock the bootloader down Dec 08 13:32:05 but the rest u can play with as u like Dec 08 13:32:09 flash a non working kernel to both boot and recovery Dec 08 13:32:09 so long as u use the bootloader to boot it Dec 08 13:32:14 ala the ubuntu nexus77 images Dec 08 13:32:19 lilstevie: isn't that fixed on n7? Dec 08 13:32:26 and it will be in the state you just called a brick Dec 08 13:32:58 RaYmAn, yeah, n7 also uses scratch rather than msc for bootloader messages Dec 08 13:33:17 hm Dec 08 13:33:22 RaYmAn, but I am referring to " fastboot is not an answer in their eyes" Dec 08 13:33:23 did you test if that works on tf201? Dec 08 13:33:32 RaYmAn, I did not Dec 08 13:33:56 RaYmAn, it is in the shutdown kernel code Dec 08 13:33:58 lilstevie: i work for an oem Dec 08 13:34:28 if ti requires anything more than absolute basic knowledge to fix Dec 08 13:34:36 raster, still doesn't change the fact, even unlocking a device will cause the definition you just called "brick" Dec 08 13:34:48 (press a reset button for example) it's undesired and everything is done to avoid it Dec 08 13:34:56 problem is a reset button is also aloiability in normal use Dec 08 13:35:16 lilstevie: yes - its a risk google has pushed asus to do Dec 08 13:35:36 raster, that asus also offer on their entire tegra 3 range Dec 08 13:35:42 prbably so google can use the n7 itself for hacking as well as allow others to do custom android roms and otherwise spread android around in general Dec 08 13:36:47 raster: tbh, if oems wanted to get around that, it *should* be possible to hook up boot strap pins in such a way that it can boot a special (signed) image incl. bootloader from sdcard. Then a win app that writes the sdcard Dec 08 13:37:31 also, they could just charge to fix bricks that could be fixed with e.g. nvflash if it was available Dec 08 13:38:05 RaYmAn: you expect them to figure out all for a tiny percentage of their customerbase Dec 08 13:38:19 when they can just continue with the old solution they have used for many years Dec 08 13:41:18 nah, I've lost faith in them bothering to do such stuff long ago Dec 08 13:41:32 so have most Dec 08 13:41:33 :) Dec 08 13:41:46 though, it would probably cut down on software repair costs =P Dec 08 13:42:21 anyways, your argument was that e.g. nvflash was too complex for normal users (I disagree) - there's plenty of ways they could make it simpler Dec 08 13:43:56 u are free to sit in meetings with some taiwanese, korea, etc. ppl and make your argument Dec 08 13:43:57 :) Dec 08 13:44:34 I don't expect an invitation any time soon Dec 08 13:44:34 ;) Dec 08 13:44:39 i've been there, done that... many times over the years :) Dec 08 13:45:06 and i get the bonus of the "listen to me" badge onmy chest Dec 08 13:45:12 even then its insanely hard Dec 08 13:45:22 although they may have to soon if the sony tool thing catches on Dec 08 13:45:56 the things that make themj move.. is if someone else moves Dec 08 13:46:06 and if it means market share gain, profits or somethnig Dec 08 13:46:09 that gets attention Dec 08 13:46:32 yeah, that's a pretty decent way of doing it - nice and semi-userfriendly tool that installs drivers and reflashes devices. It would be fairly easy for e.g. asus or google to do something similar with a custom nvflash for n7 Dec 08 13:51:18 raster: incidentally, it's amusing they spend so many resources on locking down things. You'd think it would be less resources to just make it open :) Dec 08 13:51:19 i flashed raring using the tool in ubuntu, but i only get a 8GB partition on my 32GB nexus7 Dec 08 13:53:18 RaYmAn: the locking down has already been done and it was a product requirement from the product specs team to begin with Dec 08 13:53:21 or something like that Dec 08 13:55:47 and i have a black bar under the notification bar in raring Dec 08 13:57:30 raster: it's a choice e.g. google makes. Of course it has already been done, but it's still a choice they made Dec 08 13:57:52 and then you have stuff like the arm chromebook which is completely open/unlockable. Dec 08 13:58:00 a choice made by a product specs dept that doesnt knowq nor care about hackers/development Dec 08 13:58:02 as they are not Dec 08 13:58:14 they expect the o9nly development happens inside this engineering dept they have Dec 08 13:58:17 and nowhere else Dec 08 13:58:27 I somehow doubt google is that naive Dec 08 13:58:29 people wanting to hack on their PRODUCTs is a bizarre foreign concept to them Dec 08 13:58:44 raster: did i not mention oems above? Dec 08 13:58:46 not google? Dec 08 13:58:48 :) Dec 08 13:59:45 google is essentially an oem too :) Dec 08 13:59:55 well since they bougt moto - sure Dec 08 14:00:03 tho within large companies its not so simple Dec 08 14:00:11 its a lot of hops from left to right hand Dec 08 14:00:12 :) Dec 08 14:02:11 is everyone sleeping? Dec 08 14:06:11 gurgalof_mobil, NO Dec 08 14:06:23 :D Dec 08 14:07:22 feels like no one notices me, but ill ask again later... Dec 08 14:07:57 hmm i didnt even write a question, well well Dec 08 14:11:58 gurgalof_mobil, the current ikmmge only uses a 6G partition, we're working on getting it to work for bigger devices too, the broken panel shadow is known and reported as a bug already Dec 08 14:12:07 *image Dec 08 14:12:26 okay, thanks Dec 08 14:13:40 i'll revert to quantal for the moment then Dec 08 14:14:40 raster, do you use pulse in youir setup ? iirc fixing the sound was a combo of alsa UCM profiles and pulse profile changes Dec 08 14:14:46 yup Dec 08 14:14:49 pulse is there Dec 08 14:14:49 oh and the slideshow in raring says 12.10 Dec 08 14:15:00 gurgalof_mobil: It usually does for the first few months. Dec 08 14:15:02 i use it on all y setups Dec 08 14:15:04 hell Dec 08 14:15:12 my mier eeven detects pulse and is happy Dec 08 14:15:19 gurgalof_mobil: It'll say 13.04 before release. :P Dec 08 14:15:25 raster, also note that the master is muted by default ... only suspend/resume isnt enough Dec 08 14:15:29 yup Dec 08 14:15:30 i checked Dec 08 14:15:31 :) Dec 08 14:15:33 :P hopefully Dec 08 14:15:45 pulse is unmuted Dec 08 14:15:53 rhythmobx doest play Dec 08 14:15:59 hmmm, then i dont know Dec 08 14:16:00 aplay doesnt play (bypassing pulse going right to alsa) Dec 08 14:16:08 it works flawless under unity for me Dec 08 14:16:40 weird Dec 08 14:16:46 boot, suspend/resume, hit vol-up omce (easier than unmuting) always gets me sound Dec 08 14:16:48 i didnt get to test audio until i had the rest settled Dec 08 14:17:20 ogra_: That's an odd definition of "flawless". ;) Dec 08 14:17:39 heh Dec 08 14:17:57 well, "as described" :) Dec 08 14:18:15 Thou shalt swing the mutilated chicken thrice above thine head, whilst chanting "ALSA, our lord, hear my prayers." Dec 08 14:18:30 *giggle* Dec 08 14:18:55 ogra_: yaaargh Dec 08 14:19:10 any info as to.. what this magic pulse + alsa magic was? Dec 08 14:19:26 ask diwic during the week Dec 08 14:19:29 ok Dec 08 14:19:35 either way Dec 08 14:19:40 suspend/resume is all happy now Dec 08 14:19:41 he is our sound guru and added the fixes Dec 08 14:19:47 it was wifi simply not being shut off before suspend Dec 08 14:19:54 ah Dec 08 14:19:55 u have to rfkill it or it gens wake interrupts Dec 08 14:19:55 :) Dec 08 14:20:10 my screen is all nicely portraited now Dec 08 14:20:13 yeah,, the pm-utils scripts care for that iirc Dec 08 14:20:28 portrait with working input ? Dec 08 14:20:30 i tried the 6.1-7 xorg thing and it fixes some of the funkiness of the touch when in portrait Dec 08 14:20:37 yeah Dec 08 14:20:40 oh, sweet Dec 08 14:20:40 in quantal BT stays on while suspending as moving bt-mouse unsuspends it Dec 08 14:20:49 tho the pointer jumps around like its obsessed Dec 08 14:20:51 inptu works tho Dec 08 14:21:02 you have a pointer ? Dec 08 14:21:21 unity/compiz hides it if there isnt an actual mouse Dec 08 14:21:32 (that hides such issues) Dec 08 14:21:59 yeah Dec 08 14:22:01 i haveth one Dec 08 14:22:19 i can hide it if i want Dec 08 14:22:24 i just like seeing it Dec 08 14:22:29 its kind alike a nice debug thing Dec 08 14:22:30 :) Dec 08 14:22:53 i'm down to 180m of mem usage Dec 08 14:22:55 heh, yeah Dec 08 14:23:02 for some reaosn my mystery missing ram... comes back after some time Dec 08 14:23:21 i think there are some oddities with the kernel Dec 08 14:23:25 definitely Dec 08 14:23:33 like on boot - 420m ram gone Dec 08 14:23:36 i compile a bunch of stuff Dec 08 14:23:40 suspend/resume a few times Dec 08 14:23:41 it also shows some strange load values that i cant really confirm Dec 08 14:23:43 restart wm etc. Dec 08 14:23:54 and it goes DOWN.. to about 150-200 Dec 08 14:24:01 oh yeah Dec 08 14:24:05 and DONT use powertop Dec 08 14:24:06 WAT Dec 08 14:24:20 it somehow turns off the cores Dec 08 14:24:24 and makes u have only 1 core Dec 08 14:24:31 file a bug ! Dec 08 14:24:42 everything nicely slows down to a crawl Dec 08 14:24:43 :) Dec 08 14:24:46 that should definitely be fixed Dec 08 14:24:52 i'll kick arjan next i get the chance Dec 08 14:24:58 great Dec 08 14:25:02 :) Dec 08 14:25:11 rightr now.. i dont know what it did Dec 08 14:25:17 you should confirm that is powertop that did it Dec 08 14:25:17 i was compiling at the time Dec 08 14:25:35 lilstevie: certain - i was runing htop at the time Dec 08 14:25:37 and i rant it Dec 08 14:25:41 I've come across that exact issue as a kernel bug on the tf201 Dec 08 14:25:47 and then immediately everything jumpt to a single core Dec 08 14:25:54 it has never happend otherwise Dec 08 14:25:57 so far anyway Dec 08 14:26:13 raster, lather rinse repeat Dec 08 14:26:31 busy lathering a lot of other things Dec 08 14:26:32 :) Dec 08 14:27:03 I've never been able to trace the exact moment before, but I have never run powertop Dec 08 14:27:40 they were too suspiciously timed together Dec 08 14:27:44 i dont know if its pt's bug tho Dec 08 14:27:54 it may simply innocently do something that happens to trigger the kernel bug Dec 08 14:28:08 these things happen :) i'm rather used to them Dec 08 14:33:26 someone else should probably try the above on a n7, cause if it is the case, it is n7 specific and doesn't affect T3 in general Dec 08 14:33:30 * lilstevie just tested Dec 08 14:35:29 i only have the n7 Dec 08 14:35:35 so... :) Dec 08 14:37:03 well then :p reproduce Dec 08 14:38:43 \would love to Dec 08 14:38:48 but i'm busy with.. other fun Dec 08 14:38:55 like nixing backlight artifacts Dec 08 15:22:42 where do i find the 12.10 nexus7 image? Dec 08 15:24:59 or can i resize the partiton to 32GB? instead of 6? Dec 08 15:25:52 http://hwe.ubuntu.com/uds-r/nexus7/ Dec 08 15:27:26 hi all Dec 08 15:27:40 can someone give me a hint for alsa playback on pandaboard with ubuntu 12.04 server? Dec 08 15:27:57 in syslog: OMAP ABE Media: asoc: OMAP ABE Media no valid playback route from source to sink Dec 08 15:28:11 in mplayer [AO_ALSA] alsa-lib: pcm_hw.c:1293:(snd_pcm_hw_open) open '/dev/snd/pcmC0D0p' failed (-22): Invalid argument Dec 08 15:28:20 and nothing is played Dec 08 16:04:24 one step foward Dec 08 16:04:30 now it is: omap_hwmod: aess: _wait_target_disable failed Dec 08 17:35:17 got it Dec 08 18:30:49 can i resize the partition on my nexus7 with gparted? Dec 08 18:39:36 or is there some other way? Dec 08 19:14:00 why do we have swap enabled in 13.04 on nexus7? Dec 08 19:16:25 I think everyone is sleeping again... Dec 09 00:16:42 Hey guys, anyone played with multitouch stuff on Nexus 7 running ubuntu? Dec 09 00:16:52 when i search the repos for multitouch i find: Dec 09 00:16:53 xserver-xorg-input-mtrack - Multitouch X input driver Dec 09 00:16:53 xserver-xorg-input-multitouch - Multitouch X input driver Dec 09 00:17:12 also wondering if magick-rotation is worth the trouble? Dec 09 00:17:16 would be siiiiiiick Dec 09 00:17:52 I've got things running pretty darn well with XFCE4, but it would be nice to actually use the touch screen effeciently Dec 09 00:18:28 one of the problems is by default you can drag the windows around or maximize or close the windows via touchscreen.... Dec 09 00:18:45 which is pretty lame considering it works so well in gnome-classic and lxde ect Dec 09 00:21:53 holla? Dec 09 00:25:02 don't worry, there are people still awake and coding, but as for me, I did not really tried multitouch on n7) Dec 09 00:25:19 *haven't...dammit 1AM Dec 09 00:37:04 lol Dec 09 00:37:08 right on thanks man Dec 09 00:37:39 Tassadar, it's usually 1am here and im the only one ranting! Dec 09 00:37:40 lol Dec 09 00:38:07 yeah i saw some sweet links of people using it, not on the Nexus7 but i thought id give it a try Dec 09 00:38:23 going to look up the system requirements to see if it's even doable Dec 09 00:38:27 but i figure it should be Dec 09 00:38:47 it must, on android it can handle up to 10 fingers Dec 09 00:39:09 or at least 10 fingers, because I don't have any more to test Dec 09 00:39:53 hahahaha Dec 09 00:39:57 totally Dec 09 00:40:18 i've been messing with a lot of different desktop environments trying to find one that's both stable and usable Dec 09 00:40:40 unity is just too damn slow, and i haven't been able to get unity2d to work Dec 09 00:41:13 Ethernin, there isn't really much point in those packages, evdev does multitouch, you just need a wm that supports it Dec 09 00:41:22 unity2d doesn't exist anymore Dec 09 00:41:48 i did however try, xfce4, lxde, and gnome-classic via fallback Dec 09 00:41:51 yeah no kidding Dec 09 00:42:01 not in raring that's for sure Dec 09 00:42:07 evdev does multitouch? but how to enable it? Dec 09 00:42:17 evdev? Dec 09 00:42:40 Ethernin: it's the main keyboard and mouse xorg driver Dec 09 00:42:54 Zero_Chaos, you don't need to enable it, it already tracks multiple fingers Dec 09 00:43:05 Zero_Chaos, you need a part of the windows manager to track Dec 09 00:43:11 lilstevie: ahh Dec 09 00:43:45 lilstevie, awesome what part of the windows manager ? Dec 09 00:44:07 Zero_Chaos, or something like utouch-geis Dec 09 00:44:17 Ethernin, I'm not sure, I don't care enough Dec 09 00:45:19 lilstevie, sweet, ill look around Dec 09 00:45:39 Ethernin: search for utouch and have fun. Dec 09 00:45:44 Ethernin: I'm going to five guys Dec 09 00:45:45 :-) Dec 09 00:45:52 hahaha nice Zero_Chaos Dec 09 00:46:00 bbiaf Dec 09 00:46:07 Zero_Chaos, yeah going to take a break myself Dec 09 00:46:13 Zero_Chaos, ill be back in a bit Dec 09 00:48:29 ah yes, xserver-xorg-input-multitouch - Multitouch X input driver comes up **** ENDING LOGGING AT Sun Dec 09 02:59:58 2012