[00:14] lilstevie: uhm, is your script on xda supposed to work on a unrooted transformer? [00:21] it will as long as you don't have an SBKv2 device [01:00] I've got a Microsoft LiveChat LX-3000 headset that x64 has a driver for. The ARM build seems to be missing the driver, anyone know where I should look to get it compiled? [01:02] It sees the headset, the vol up/down, mute buttons on it work, pulseaudio just doesn't see it. === lopi is now known as Lopi === zyga is now known as zyga-gone === GrueMaster_ is now known as GrueMaster [06:06] time to test 11.10 on a panda [06:07] lilstevie: still remember me? :P [06:08] nope :p [06:09] :-( === zyga-gone is now known as zyga === zyga is now known as zygqa === zygqa is now known as zyga [13:04] ogra_: ping [13:10] berco, yes ? [13:10] ogra_: have you tried the oneiric ubuntu-core minimal fs on panda? [13:10] ogra_: I'm having issues to boot it... [13:11] boot it ? [13:11] hwo do you boot it ? its an unconfigured base system to buuild other images on top [13:12] yes, I put my mlo, u-boot and kernel. [13:12] not intended for usage without a lot of configuration or modification [13:12] I was expecting to get at least up to a console [13:12] i.e. if you would build a settop box or IVI image you would use it to put your user-less UI on top for example [13:12] I'm stuck with last message "mountall: Disconnected from Plymouth" [13:13] how did you configure it ? [13:13] did you install a kernel packge, roll an initrd etc ... [13:13] it is not intended to work OOTB at all, unless you look for a cheap way to get a chroot [13:14] okay, so it's not at all intended to be a basic fs. I need to customize it first [13:14] it is just enough OS to execute apt, nothing more [13:15] no users, no root access etc [13:15] you should get a console if you configure it right though [13:15] on framebuffer ... [13:15] okay, I don't event get this [13:16] it should get you a console on screen [13:16] but there is no user and root is locked by default [13:16] I only have the 2 pinguins displayed on the screen, so it starts... [13:16] it *wont* get you any login on serial since it is (on purpose) not configured for that [13:17] no [13:17] your kernel starts [13:17] yes [13:17] penguins have nothing to do with userspace [13:18] I agree. I see the FS is mounted in the debug output and nothing else then [13:18] on hdmi, right ? [13:18] yes, I'm on HDMI [13:18] then you should eventually see a login prompt [13:19] with no option to log in though, unless you configured that [13:19] I have the latest kernel from sebj with drm, etc.. maybe I should try with an older kernel [13:21] likely ... what initrd are you using, the mountsall issue looks like your initrd is messed up [13:22] I need to check for initrd. [13:23] also if you dont use an initrd or a messed up one, make sure to not use UUID for root= on your cmdline [13:23] the kernel isnt capable of using that, it needs the initrd [13:24] okay, I just checked: my last test was without initrd and I use root=/dev/mmcblk... [13:24] k [13:24] I also removed spash from cmdline [13:24] spash=splash [13:24] *theoretically* it should work without inird, but since there are no initrd less images in ubuntu,. nobody actively tests that [13:25] so i would always recommend to use one ... [13:25] okay, I'll test to add it [14:04] ogra_: with uInitrd, I get to the login screen now :) [14:05] great [14:06] so now as I understood from you there is no user or root login possible. I need to customize the fs, right [14:07] yes [14:14] this news tech article says late 2012 early 2013 for A15's to hit the market [14:14] I don't think I buy that [14:15] broadcom has licenses for a tablet already [14:18] maybe it's true [18:44] lilstevie: do i need to root my transformer to use your scripts? [18:55] XorA: hey hey [18:55] XorA_: hey hey [18:56] hey [18:57] forgetting I already had an xchat :-) [18:57] XorA: you going to make it to ELC-E? [18:57] prpplague: no :-( [18:57] XorA: doh [20:09] prpplague, are you going to ELC-E? [20:09] davidm: yea [20:09] davidm: got something that needs to go? [20:09] I'll see you there then [20:39] shadeslayer: for you to be even asking that means you need to do more research [20:40] lilstevie: well, your instructions don't even mention rooting, which is why I'm a bit lost, I rebooted into recovery and the device doesn't show up ... [20:40] I can't even figure out how to use the mmc on the device when it's booted in Android [20:41] ( I don't get the option to mount the mmc over USB like I do on my phone ) [20:42] I haven't had alot of time to poke around which is why I'm asking [20:42] shadeslayer: what I mean is you need to do more research, like learn about nvflash [20:42] the very fact you are asking if you need to be rooted is telling me you haven't even looked into what nvflash is [20:43] I did go through it, which means I obviously missed something [20:43] I'll look at it again [21:07] ogra_ / GrueMaster for 11.10, is there a command line change i need to do tell the install to use the DVI as primary instead of hdmi? [21:08] not that i know of, we dont maintain the kernel anymore so i could only guess [21:08] prpplague: Not sure. [21:08] hmm [21:09] robclark may know. [21:11] I'd say ask on #pandaboard, but then you'd probably smack me. :P [21:11] GrueMaster: hehe, indeed i would [21:13] prpplague, drm will light up all monitors that it detects a connection on.. [21:13] robclark: yea, but it seems that ubuntu is selecting the hdmi for some reason [21:14] prpplague, there is a (*detect) fxn ptr in omap_dss_driver.. you just need to make sure that is returning sane values [21:14] prpplague, it should select both.. in theory.. [21:14] (well, that is, if you have hdmi attached) [21:14] robclark: can i use the video bootargs? video=DVI-I-1:1024x768@85 video=TV-1:d [21:14] robclark: no hdmi attached [21:15] hmm.. that sounds suspicious.. [21:15] @ your desk? [21:16] yea [21:16] k, I'll swing by [21:23] GrueMaster / ogra_ any special boot args to use to get console working with the 11.10 prebuilt desktop images? [21:34] console? You mean serial console? [21:35] GrueMaster: yea [21:36] For serial console, you just need to create a /etc/init/ttyO2.conf with the proper settings. Use /etc/init/tty2.conf as a template, changing the tty port and speed. [21:37] If you are trying to figure out the dvi stuff, you might be better off with a server image. You can boot and configure it from the serial console, then tweak from there. [21:39] GrueMaster: hmm [21:39] * prpplague doesn't like this aspect of ubuntu [21:39] * prpplague does an install using hdmi first [21:39] Why not? [21:40] How many desktop image do you usually configure through a serial port? [21:40] GrueMaster, ogra_, it looks like a plymouth issue.. or perhaps plymouth vs omapdrm.. [21:40] but the splash screen is lighting up HDMI, but not DVI.. [21:40] That's possible. [21:41] so if you plugin just DVI, you see busybox instead of plymouth [21:41] GrueMaster: just prefer to have an easier way of enabling the serial console [21:41] You can mute plymouth by removing splash from the kernel cmdline. [21:41] is plymouth showing status msgs during the partition-resizing magic? [21:42] It should show a few. Not many. [21:42] robclark: none with just dvi plugged in, only with just hdmi plugged in [21:42] But it may also ve treating the dvi port as a second monitor source. [21:43] GrueMaster: right that is what it appears to be doing [21:43] GrueMaster, on my laptop, splashscreen is coming mirror'd if I boot up w/ 2nd display attached.. [21:43] robclark: Might need to talk to the omap fb driver developer. [21:44] Oh, wait.... [21:44] there were some new support in drm for unaccelerated "dumb scanout buffer", which might not be in the version of omapdrm in 11.10... but on other hand, I'm not sure if plymouth is using those yet.. [21:44] so big question is whether plymouth is using drm/kms kernel API or falling back to fbdev.. [21:45] Not sure. Let me see if I can get slangasek to answer. [21:45] could someone point me at what kernel you are using? I can check what revision of omapdrm you are using.. [21:46] 3.0.0-1205.10 [21:46] Is on the release image. [21:46] * robclark was looking for a git tree + tag.. [21:47] ok, think I found it [21:49] GrueMaster: when i use the prebuilt 11.10 images, what should i expect to see on the hdmi display? [21:49] ok, GrueMaster, I think I added support for the "dumb scanout API" (which libkms can use) at same time I added GEM support... so not in the version of driver you have.. [21:49] so plymouth might be falling back to some legacy fbdev mode [21:56] * prpplague grumbles [22:02] robclark, GrueMaster: plymouth requires a per-drm-driver backend; I don't think plymouth knows about omapdrm [22:02] slangasek, ok, it doesn't use libkms? [22:02] I have never heard of libkms, so I guess not :) [22:02] (which has backends for different drm drivers, and also a backend that can use the new "dumb scanout" API..) [22:02] it is in libdrm git tree [22:03] ok - nope, not being used, at least not in the version we have [22:03] Does anyone here know anything about getting ubuntu on mobiles/tablets? [22:03] ok [22:03] then I guess there may be some plymouth work to do at some point.. [22:03] * slangasek nods [22:04] or if it uses the "dumb scanout" ioctls, then it should work out of box with omapdrm which is going upstream... the slightly newer one with GEM support [22:04] I don't think we would want to refactor plymouth's drm support in 12.04; a bit risky for an LTS [22:04] * robclark would recommend to have the dumb scanout ioctls as fallback [22:05] but the current drm support is not great (we even have radeondrm blacklisted right now, due to multimonitor issues), so I would love to see this improved in the future [22:05] that won't work for omapdrm in 11.10 but will work in 12.04 [22:06] slangasek, http://lxr.free-electrons.com/source/include/drm/drm_mode.h#L347 (fyi) [22:06] robclark: have you spoken with plymouth upstream about this, by any chance? [22:06] those ioctls will give an API that can be used to allocate/mmap a buffer for any drm device (supporting GEM at least) [22:06] not really.. [22:07] * robclark has no idea who that is ;-) [22:07] #plymouth :) [22:07] ahh [22:07] best if the support could land there first [22:07] I guess they must know about those APIs.. they were added pretty much for the purpose of things like plymouth.. [22:07] but I think they are pretty recent [22:08] looks like there's preliminary support for libkms upstream since Aug 2010 [22:08] there hasn't been an upstream release since then, however [22:08] hmm.. well, if they finish that, they get the dumb-scanout buffer stuff for free.. === Jack87|Away is now known as Jack87 === Jack87 is now known as Jack87|Away