=== ian_brasil_ is now known as ian_brasil === sumitsemwal is now known as sumits === lilstevie is now known as lilstevie|ZNC === lilstevie|ZNC is now known as lilstevie === JaMa|Zzzz is now known as JaMa|Off [08:38] morninge [10:03] ogra: Hi Oliver, you came back after all :) [10:03] ogra: I find something puzzling regarding the PPA. [10:04] I wouldn't trust much communication until there's been no netsplit traffic for a while: sometimes stuff gets lots in the backscroll. [10:04] persia: May I please ask _you_ about this PPA thing, then? [10:05] I generally recommend asking questions generally of the channel. I'm unsure that this is the best channel for a PPA question, but I'll try. Otherwise, you can ask in #launchpad. [10:05] persia: Ok; I move there. [10:17] so the sound stuff doesnt work at all anymore :/ [10:18] What have you done? [10:18] lag, trying mathieus kernel, putting all new files in place [10:18] Hmm [10:19] all muted [10:21] urgh [10:21] ? [10:21] a) i tried remotely .... my screen is all blue if i switch the monitor on [10:22] Nice [10:22] not sure what he did to that kernel package [10:22] b) calling alsactl init makes it work [10:22] I'm just looking now [10:22] Is that correct? [10:22] no [10:22] * ogra reverts everything, it worked yesterday with distro kernel and my file without alsactl init [10:23] What does cat /proc/asound/cards produce? [10:23] ogra: panda sound still has problems? [10:23] What was the link to the kernel he gave you? === XorA|gone is now known as XorA [10:24] http://people.canonical.com/~mpoirier/linux-image-2.6.35-903-omap4_2.6.35-903.13+release3_armel.deb [10:25] lag, before i had http://people.canonical.com/~roc/kernel/audio/linux-image-2.6.35-903-omap4_2.6.35-903.13+audio1_armel.deb installed [10:25] just reinstalling that one [10:25] That's Bryans kernel [10:26] right [10:26] and that one worked after the alsa-lib change [10:26] i dont get why i would get that bright blue screen [10:26] the patches dont touch anything related [10:27] I'm trying to find the server which he built on [10:27] But to no avail as yet [10:28] sound/soc/omap/sdp4430.c , include/sound/soc.h and sound/soc/soc-core.c [10:28] nothing should have any influence on my display [10:28] I can't find any of his builds [10:28] fun [10:29] * ogra reboots into old kernel with old init files [10:30] yup, login sound here [10:30] * ogra checks the screen [10:30] and netbook ui [10:33] I can't find any up-to-date build? [10:33] weird [10:33] well, we need cooloneys kernel plus the bits from lrg [10:34] i just need the names [10:34] and i know cooloney built on one of your servers [10:34] How different is cooloney's tree to the mainline? [10:35] What if I applied lrg's patches to the current tree? [10:36] you would be missing a ton of audio fixes [10:36] afaik [10:36] Ah [10:36] oh, no, wait, i think they were applied [10:36] please check the tree [10:36] Yep [10:36] not sure there was an upload with thgem though [10:38] http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-maverick.git;a=shortlog;h=refs/heads/ti-omap4 [10:38] Is that all the patches? [10:38] Or are there more? [10:38] its like 70 patches and i cant find the merge request on the kernel ML [10:39] RAAAAAHHHH !!!! [10:39] * ogra freaks out about being CCed on all the kernel mails [10:39] I HATE THAT ! [10:39] We've had this discussion [10:39] thanks to that they dont match the ML sort filter and end up in places i cant find them [10:39] Not everyone reads the mailing list [10:39] thats so annoying [10:40] i do [10:40] Good for you :) [10:40] and i am sure that other people filter on ML headers too [10:40] Perhaps you shouldn't [10:40] ;) [10:40] I have working filters [10:40] ;) [10:40] Anyway ... [10:41] That current tree looks like it has loads of new audio patches [10:41] lag, based on ML headers (like you should for LP mail ?) [10:41] [Maverick] [ti-omap4] SRU: audio driver fixings [10:41] I'd say ~90 patches [10:42] applied and pushed by tim on the 6th [10:42] In the ML/ [10:42] ? [10:42] yes [10:43] Yes, they are applied [10:43] good [10:43] so only the naming patches from lrg are missing there [10:43] could you roll me a kernel deb ? [10:45] Of course [10:45] merci :) [10:45] * ogra meanwhile goes to find some breakfast [10:45] we'll need GrueMaster (once he gets up) to confirm the switching works, he has a blaze [10:47] There are only 4 patches right? [10:52] ogra, jasper-initramfs/scripts/local-bottom/jasper_setup runs in initramfs, right? [10:53] yes [10:55] So, I think we can't enable universe via the python API there. Is there a script that runs in userspace, or do you think unsafe sed is better? [10:58] you cant run apt-get update from there at all [10:58] no networking in initrd [10:59] so your apt-cache will be completely out of sync [10:59] Oh, I don't care about apt-get update. [10:59] That's fine. [10:59] i do [10:59] Everything will be able to tell the apt-cache is out of date because of timestamps. [10:59] if a user calls apt-get install i dont want him to be greeted with an out of sync cache error [11:00] thats about the first impression not about technical issues [11:00] * persia digs into oem-config code [11:01] i think apt-setup is only called from oem-config in case you also run tasksel [11:01] which we dont want in our setup [11:01] i'm not sure they are separate [11:52] ogra_ac: I also have Blue screen! [11:53] even with the new kernel ? [11:53] wow [11:53] The latest kernel [11:53] We're screwed [11:53] i dont get how the soc changes can affect hdmi [11:54] I don't think it was the SoC changes [11:54] especially since its only a few lines adding names [11:54] Let me roll there kernel without lrgs patches [11:54] well, the kernel on the release image works [11:54] Which was compiled when? [11:55] and i think the 90 sound patches are in there [11:55] and cooloneys test kernel works for me too [11:55] without any issues [11:55] there was one patch afterwards ... [11:56] one sec, let me find it [11:56] [Maverick] [ti-omap4] [SRU] UBUNTU: [Config] enable passing all kernel command line to init [11:56] 7 days ago Mythri P K OMAP4:DSS:HDMI:Fix for default boot on HDMI with ES2.0 [11:56] Thu, 7 Oct 2010 [11:56] lag, oh, wait [11:56] which image do you use ? [11:57] x-loader changed along with that patch [11:57] Ah [11:57] I'm using the daily build [11:57] you need to use the new image [11:57] But with my own x-loader [11:57] Whould the daily build work? [11:57] right, x-loader and the Mythri P K HDMI patch had to be applied at the same time [11:58] the daily from 7th should [11:58] Okay, re-DDing [11:59] (it should be the release image, we didnt re-roll) [12:15] ogra: Okay, my blue screen has disappeared momentarily [12:15] ogra: Why did you get a blue screen? [12:15] ogra: Were you using the old x-loader too/ [12:17] lag, yeah [12:18] i still use the same image, upgraded since RC and we dont upgrade x-loader (as you wouldnt upgarde a PC BIOS) [12:19] ok, updated x-loader [12:19] let me try with mathieus kernel again [12:20] If it doesn't work, I have a working one here [12:20] k [12:20] i suspect it will work now [12:21] *twiddle* [12:21] Which board do you have? [12:21] 8 layer? [12:22] 2.1 [12:22] I am sooooo behind on boards :( [12:22] and 8 layer 2.0 and a special 2.1 with DVI wiring added [12:22] 2.0 8 layer is fine [12:22] I only have a 6 layer [12:22] oh, yeah, we need to upgrade you [12:23] No kidding [12:23] Why have I been left behind [12:23] ask rsalveti, i thinnk he has a patched u-boot and kernel package [12:23] because we got limited amounts of HW [12:23] I should always have the most up to date board! [12:23] And who has priority? [12:23] and i think mathieu got the new one because he didnt have one at all [12:23] Surely the kernel is important to you? [12:24] Mathieu is normally OMAP3 [12:24] cooloney has the 2.1 and the 2.1+DVI [12:24] you should get his 8 layer [12:24] ogra: how old are 2.1 boards? [12:24] about a week [12:24] * hrw wonder which ver will land in my hands [12:24] we only got fout of them directly handed out [12:25] hrw, pray its not a 6 layer [12:25] lag, we'll get more borads within the next few weeks [12:25] ogra: I was told that it will es2.0 8layer minimum [12:25] enough for everyone i hope [12:25] I want 2.2 [12:25] lag, well, then get a soldering iron :P [12:26] and a list from TI what will be on 2.2 [12:26] good thing is that I will be able to run anything not exactly ubuntu on it D: [12:26] For future h/w releases I really should have the latest boards [12:26] they wont go to production for a while :) [12:26] It was a jk [12:26] 2.1 is what goes out [12:26] and i hope we wont see any HW upgrade for a while now [12:26] But as the OMAP4 kernel team representative I should be on the priority list [12:27] lag, well, cooloney is *the* OMAP4 representative i was told [12:27] fight, fight! [12:27] we can swap that indeed [12:27] Who told you that? [12:28] that was decided at UDS iirc [12:28] Before I joined :) [12:28] right === amitk is now known as amitk-afk [12:28] Well there's a new daddy in town ;) [12:28] yeah [12:28] rsalveti: let me know if my response about the P8 BeagleBoard-xM memory was at all confusing. you should swap out MLO and see if your problem goes away.. [12:28] lets shuffle responsibilities at UDS [12:29] By all means [12:32] hmm, i think i trashed my panda image :/ [12:32] ah, no [12:32] now it boots [12:35] lag and ogra, are you guys talking about my panda toy? -:) [12:35] cooloney, hehe, we do [12:36] Yep yep [12:38] yeah, I got 2 panda boards now. actually one is 2.1+DVI [12:39] oh, you dont have the 2.0 anymore ? [12:39] and one is 2.0 [12:39] yeah, the 8 layers 2.0 [12:39] i thought you got a normal 2.1 too [12:39] no sound here :/ [12:40] ogra: oh, i returned that to chris, as 2.1+DVI is enough [12:40] sigh [12:40] but i have to keep 2.0, since we need to support 2.0, right? [12:40] so with the old kernel it all worked [12:40] with the new kernel i have to call aslactl init *again* [12:41] * ogra doesnt get that [12:41] and i cant esaily downgrade, hrm [12:41] ogra: just before we left, i got 3 boards -> 2.0, 2.1 and 2.1+dvi [12:41] yeah [12:42] ogra: after sent your guys to airport, we were back to hotel [12:42] chris called GrueMaster [12:42] ah [12:42] chris showed up and i returned the normal 2.1 to him [12:42] so i just have 2 boards. one old 2.0 and a 2.1+dvi [12:42] yup [12:44] so lag, do you wanna i give you my board? i think you got 2.0 board [12:44] ogra: how come sound doesn't work? [12:44] ogra: doesn't work with new kernel? [12:44] cooloney, it never worked ? [12:44] only partially [12:45] we're trying to fix the last bits since beginning of the week [12:45] oh, i mean even using your alsactl trick, it doesn't work? [12:45] sure it does [12:45] ogra: ok, got it. [12:45] but the alsactl call is wrong [12:45] ogra: so any idea about our plan? [12:46] yes, look at the bug [12:46] lag, documented everything [12:46] ok, let me check [12:46] ogra and lag, thanks [12:50] GRRR [12:50] no sound at all [12:50] not even after alsactl init now [12:52] HA [12:52] putting my omap4 file back works [12:53] so one step at a time now [12:53] keeping everything as is and upgrading to mpoiriers kernel [12:53] :) [12:53] cooloney: I don't want your board [12:53] (and adjusting the name in 00main indeed) [12:54] cooloney: I'll wait for the next batch [13:00] lag: oh, panda 2.2? heh [13:01] cooloney: No, just the new batch of 2.1's [13:01] ogra: did liam provide the patch to export platform name from kernel to userspace [13:01] lag, confriming, with the new kernel it doesnt work at all [13:01] lag: yeah, indeed. [13:01] ogra: which kernel? [13:01] cooloney, yes, thats what i'm testing here [13:02] ogra: Well that's not good! [13:02] ogra: ok, where is the patch from liam [13:02] no [13:02] I don't know which kernel he rolled [13:02] Do you want to try mine? [13:02] yes [13:02] In case it's any different [13:02] Okay, wait one [13:03] i know which kernel he rolled ... 2.6.35-903-omap4 #13+release3 [13:03] :P [13:03] its release3 :) [13:03] (i dont know what release3 is supposed to mean though) [13:05] ogra: Uploading [13:05] My point exactly [13:05] I don't think it has any of the audio patches in [13:05] Doh! [13:05] The one he rolled was *.13* [13:05] Mine is *.15* [13:06] That would explain it [13:06] * lag is uploading [13:06] Well, my kernel is :) [13:08] ogra: I'll let you know when it's finished uploading: http://people.canonical.com/~ljones/lp637947-maverick/ [13:08] oki [13:09] the alsa-lib fix is uploaded btw [13:09] we'll need some testers to confirm it works and can go to -updates (its in -proposed now) [13:11] ogra: i can try that on my board [13:11] cooloney, use the 00main and omap4 files from the bug [13:11] ogra, Have you prepared the candidate alsa-utils bit yet (the new conf file and 00main changes)? [13:11] persia, not until i know they work [13:12] Ah, so you're waiting for the alsa-lib bit to be verified? Makes sense. [13:12] ogra: oh, if you didn't changed those 2 files since Dallas trip, I've already installed them on my board [13:12] persia, to test the switching i need the new kernel, but the one i got didnt include cooloney's 90 audio patches [13:12] cooloney, great, with the new alsa-lib it should work without alsactl init [13:12] ogra: i think lag's kernel contains those audio patches [13:13] Oh, I somehow missed that the alsa-utils bit depended on the kernel bit. Sorry for the poorly-timed poke. [13:13] cooloney, right, thats what i'm waiting for [13:13] ogra: ok, so i just need to update my alsa-lib on board? [13:13] persia, well, my alsa-libs bit doesnt but that works only for panda [13:13] what we need is the switching piece based on boardname [13:13] and thats what we try to test atm [13:14] cooloney, from maverick-proposed, yes [13:14] ogra: ok, cool. let me upgrade [13:14] cooloney, bug 652035 is the one waiting for test results [13:14] Launchpad bug 652035 in alsa-lib (Ubuntu Maverick) (and 1 other project) "libasound2 not finding usb sound card (affects: 2) (heat: 18)" [High,Fix committed] https://launchpad.net/bugs/652035 [13:15] ogra, Right. I just was chatting with crimsun earlier, and added the task to the other bug (637...) based on what he was saying. [13:16] right [13:16] i'll prepare my stuff today as soon as i can prove it works on the panda at least [13:16] we'll need GrueMaster for the blaze test [13:16] Right. [13:16] btw, i suspect the fix also fixes dove [13:17] we should ask NCommander to try [13:17] ogra: Done [13:17] lag, you guys dont need to upload the headers every time :) [13:18] i wont compile anything against the test packages :) [13:18] It's automatic [13:18] ah [13:18] scp linux-* [13:18] i thought you do that with scp === amitk-afk is now known as amitk [13:18] right, just use scp linux-image* [13:19] What does it matter? [13:19] saves time and bandwith [13:19] 100% 573KB 573.3KB/s 00:00 [13:22] 573KB? Are you sure that's the file you wanted? [13:22] well, at least it took no time :P [13:22] everyone cross your fingers :) [13:22] * ogra reboots [13:24] lag, hmm [13:24] lag, are you sure you applied the patches ? [13:24] my card is still named SDP4430 with your kernel [13:24] 2 secs [13:25] should be Panda [13:25] ogra@panda:~$ cat /proc/asound/cards [13:25] 0 [SDP4430 ]: SDP4430 - SDP4430 [13:25] TI OMAP4 SDP4430 Board [13:25] yeah, no name change [13:26] behind the colon it should be: OMAP4 - Panda [13:27] http://kernel.ubuntu.com/git?p=lag/ubuntu-maverick.git;a=commit;h=39f15b5e3ac8c486bcc8bcf4376b7d2e8122517b [13:27] They're applied [13:27] Maybe something went wrong with the build [13:27] Let me try again [13:27] err [13:28] whats these abe changes ? [13:28] ref: lrg's email 12/10/10 @ 11:45 [13:31] ah, k [13:31] he does weird testing though [13:31] pretty hackish :P [13:32] but ok, i'll trust him [13:32] he should know what the mixer values should be [13:32] though i dont get why the name change doesnt work now [13:33] As I said, it's probably a build error [13:33] I am re-compiling [13:33] ok [13:33] Shouldn't be too long [13:34] k [13:34] i'll go play with my ac100 [13:34] :) === ian_brasil_ is now known as ian_brasil === ian_brasil_ is now known as ian_brasil [14:13] rsalveti, hey [14:22] ogra: The latest kernel is up by the way [14:22] (and has been for a while) [14:26] pulling, thanks [14:32] ok, rebooting, lets see [14:32] (or hear rather) [14:36] the names changed [14:36] but it doesnt work :( [14:37] lag, could i get a build with *only* the name changes ? [14:39] You mean without all the other audio patches? [14:39] Or without his other 2 patches? [14:39] right [14:39] without any patches to the abe files === ian_brasil_ is now known as ian_brasil [14:40] i have to call alsactl init manually again with your kernel [14:40] then it works [14:40] and i cant imagine that to be caused by the name changes [14:41] sound/soc/omap/abe/abe_dat.c , sound/soc/omap/abe/abe_def.h and sound/soc/omap/omap-abe-dsp.c dropped please [14:41] (the changes, not the files indeed :) ) [14:43] oh, shriek ! [14:44] * ogra just found a spares alsactl init in one of the alsa files [14:45] *sparse [14:45] Does that mean you don't want me to re-compile yet? [14:45] yes, wait a sec [14:45] i have to test the old kernel again [14:45] GAH [14:47] lag, leave it i think i'll haver to additionally apply a hack to alsa-utils [14:47] seems the alsa-lib fix only fixed it partitally [14:47] Okay [14:47] Let me know if you need me [14:47] will do [14:48] but i think kernel wise we're fine now [14:50] ogra: hey === xfaf is now known as zul [14:52] * rsalveti reading backlog [14:54] rsalveti, not about the backlog :) [14:55] rsalveti, about the "oem-config doesnt start in my rootstock images OMG the world ends !!!" issue :) [14:55] haha [14:55] rsalveti, i ran into a similar issue with ac100 users ... [14:55] we need to use --numeric-owner in tar [14:56] sure, saw that while reading backlog [14:56] I have an exclusive todo list just for rootstock :-) [14:56] else it will replace the UIDs and GIDs with the ones from /etc/passwd and group from the host machine [14:56] but I can push at least this fix for now [14:56] ah, you already saw it, k [14:57] ogra: so, ac100 users are using rootstock? [14:57] no, but a tarred up rootfs [14:57] and i suddenly got users with exactly the same issues [14:57] using --numeric-owner on packing and unpacking fixed it [14:57] hm, ok [14:57] yeah [14:59] ogra: so, how is the audio thing going now? [14:59] bad, we still need the init, but at least it gets respected in the right place now [14:59] i'm preparing patches now [15:01] ogra: reading the above exchange, should I undertand the patched kernel that was produced for you didn't work ? [15:01] mpoirier, you based on the wrong version and were missing about 90 audio patches apparently [15:02] it didnt work for me for other reasons based on my own stupidity though [15:02]