[00:25] bjf: am here now, lemme know if you needed something === dileks is now known as Guest43017 === mfisch` is now known as mfisch === mfisch is now known as Guest29044 [01:40] ogasawara: nah, thanks, vanhoof and hugh took care of it [02:05] ogasawara: still around, for the bug 1084000 [02:06] https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1084000 [02:06] cooloney: Error: bug 1084000 not found [02:07] i believe we can point out to ubuntu server image builder and add a small post-install script to add 'elevator=deadline' to default grub kernel command line [02:14] xnox: hey, are you still around? [02:40] ogasawara: When do you think the quantal tree with by tracking 3.5? [02:40] *will be tracking [03:03] BenC: I wanted to get it rebased today but got sidetracked, so hopefully tomorrow. [03:03] ogasawara: sweet, my work pre-depends on that, so if you could just ping me when it's done, I'd greatly appreciate it [03:03] BenC: will do [03:04] ogasawara: I have my tree on github ready and cloned from ubuntu-quantal [03:05] BenC: you're quick, I haven't heard back about the zinc access yet. [03:07] ogasawara: It's kind of moot at this point, but I guess for future reference it will give you a litmus test of how likely someone else is to get access :) [03:07] BenC: it's definitely not the first time it's come up, so will indeed be good to know what the actual policy is [03:32] cooloney: while I agree that having that bug reporter use elevator=deadline is a valid workaround in that bug scenario, I don't think it's a convincing enough argument to default *all* server image installs to it. the fact that the CFQ scheduler is resulting in disk errors and hangs seems like a legitimate issue that needs to be addressed. Additionally it won't be obvious for anyone examining our kernel config that in so [03:32] me cases the schedule will be set to elevator=deadline. do you know if we've been able to reproduce? The bug report doesn't have any logs files of the actual errors to confirm the issue. [03:34] ogasawara: i can't reproduce it on my side, I just confirm CFQ is our default IO scheduler for server image now. [03:34] ogasawara: and i don't think we need any change in our kernel side,since we have good reason to keep it as the same as desktop kernel [03:35] so might need talk with server folks about this issue, or add elevator=deadline in server image after fully installation [03:37] cooloney: yah, probably best to raise with the server team for feedback. maybe try and get some benchmarks ran. I thought we'd done some in the past but can't seem to find the results. [03:38] ogasawara: no problem, i will handle that and might discuss this with smb firstly. [03:38] cooloney: sounds good [07:16] cooloney: well I am now. [07:16] xnox: cool, i saw your update about that usb 3.0 bug [07:16] looks like some works some not [07:16] quite weird [07:17] cooloney: well all that tested got a 'P' - pass. Others were not tested. So no fails. [07:17] it did take between 47-83s to spin up the drive and make it appear [07:18] the external drive has it's own power, so I'm not sure why it was taking so long. [07:38] Is it a known problem that installing the 3.5rc quantal kernel on precise makes an usb keyboard nonworking? and are there any workarounds? [07:40] xnox: i will take a look at the kernel change based on your testing result [07:43] cooloney: they are inconclusive. I have tested 4 kernels and all 4 of them passed [07:44] both the reported kernel by the user, kernels before & after it. As well as the current 3.4 kernel. And all works. [07:44] APART from that it takes ages for the device to appear (upto a full minute sometimes) [07:45] xnox: did you met this issue on some old ubuntu release before? like 11.10 [07:46] i mean the issue about long enumeration of USB 3.0 devices [07:49] cooloney: I got this laptop with USB 3.0 port this christmas and it has been running precise until precise release, and quantal since then. My first USB 3.0 device arrived yesterday. [07:50] Previously I thought that my 3.0 usb port was 'backwards-incompatible' as I would get frustrated and plut the stuff into 2.0 port, as that one 'was not working', well it actually was just enumerating slower. [07:50] * xnox it's not very scientific, and I'm not sure how to numerically test this. [07:53] based on you testing, "P 3.2.0-23.363.2.14" means your devices can be enumerated and works fine, right? only it will take about 1 min to see the device appear [07:54] xnox: if so, i'm going to check the change between 3.2.0-23.36 to 3.2.0-24.37 [07:57] cooloney: well, all of them were slow. But I was not using, e.g. a stopwatch, nor multiple tests to see if there is a consistent trend/correlation. E.g. there was still a lot of cpu activity while desktop was loading and I would plug my device in. [07:59] cooloney: i am dogpiling on the bug report now. I think it is best to wait for the original reporter to come back and find out if it's "working slowly" or "not working at all" [07:59] hence the status incomplete. [08:01] xnox: thanks a lot. i got you [08:01] cking: morning, how's the celebration for Queen [08:02] cooloney, morning; the celebrations were a load of fun [09:18] "For the fist time in my memory, today's linux-next build required no manual intervention i.e. there are no new conflicts or build problems." [09:18] linux-next: Tree for Jun 7 === yofel_ is now known as yofel [10:55] ppisati: I have A1 desktop image not booting on panda [10:55] ppisati: it's stuck on the Uboot , keeps rebooting [10:55] ppisati: how do I enable the debug info? [10:56] ppisati: I'd like to give some more info than just "it doesn't boot" on the bug report [11:01] gema: which board rev? [11:01] ogra_: didn't you try A1 installation on panda? [11:01] ppisati: A4 [11:02] gema: i've a panda a4 [11:02] gema: where di you get the img? [11:03] ppisati: from the tracker [11:03] ppisati: http://iso.qa.ubuntu.com/qatracker/milestones/221/builds/16974/downloads [11:04] gema: ok, i'm downloading it [11:04] gema: i'll let you know [11:04] ppisati: thanks [11:04] ppisati: can I enable serial console on the early boot ? [11:04] ppisati: so that I can see something else [11:05] gema: on the installation image? uhm [11:05] yep [11:05] gema: yes, but you'll have to tinker with uboot [11:05] gema: or reroll boot.scr [11:05] I'd like to see the crash [11:06] ppisati: ack, will try [11:06] gema: what's the content of boot.scr on the first partition of the sd card? [11:09] ppisati: Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(179,2) [11:10] gema: so you installed the img somewhere [11:10] gema: and it's unablt to mount what root= says [11:10] I dd'ed the image to an mmc and then tried to boot the load [11:11] gema: did you have any video output? [11:11] ppisati: no [11:12] gema: did you modify boot.scr? or did you manually override u-boot vars? [11:12] gema: anyway, just finished downloading the img, let me try by myself [11:12] ppisati: I unset the bootscr var [11:13] ppisati: I will raise a bug, now I have something to put on it :D [11:14] gema: can you past the boot messages somewhere? (http://paste.ubuntu.com/)? [11:15] yep [11:16] ppisati: http://paste.ubuntu.com/1028413/ [11:20] ppisati: the /dev/mmcblk0p2 in the card is not ext3 [11:20] ppisati: that is probably why the board cannot load it [11:20] ppisati: so maybe the image is wrong or .. ? [11:20] i'm dding it now [11:20] wait [11:20] ok [11:21] ogra_ tested the installation img yesterday (or two days ago) and he said it was ok [11:21] is this img different in any way? [11:22] ppisati: this image is from yesterday [11:22] ppisati: and it is an A1, so I am retesting [11:22] no result for it on the iso tracker [11:23] ppisati: breaking for lunch now, back in a bit [11:23] still dd-ing.. [11:39] ppisati: back [11:41] gema: ok so [11:41] gema: i just tried your img on my panda rev a.4 here [11:41] gema: and it's booting [11:42] gema: how did you dd the img? [11:42] ppisati: the image itself is broken [11:43] gema: but did you rech the installation? you installed and then it paniced or what? [11:43] ppisati: let me verify I downloaded it correctly [11:43] ppisati: it never booted the installer [11:43] gema: works here [11:43] choosing the timezone now [11:44] ppisati: ack , I will have to figure out what's wrong with my setup [11:44] ppisati: thanks :D [11:44] gema: zcat $imgfile | sudo dd of=/dev/sdX bs=64k [11:44] gema: try this [11:45] gema: sd cards are picky if you don't specify dd bs= [11:45] ppisati: I used bs=1M [11:46] finished netering my user details, it's installing now [11:46] *entering [11:46] ppisati: the second partition, in my image, is ext4 yet the bootargs in uboot tell the kernel it's ext3 [11:47] ppisati: any idea how could I do that to an image? [11:48] ppisati: also, what is the correct type for mmcrootfstype? ext3 or ext4? [11:53] gema: did you modify boot.scr? [11:53] gema: because i don't have that info in mine [11:53] gema: setenv bootargs vram=40M mem=456M@0x80000000 mem=512M@0xA0000000 root=/dev/mmcblk0p2 fixrtc quiet splash [11:53] gema: that's what you img says [11:57] ppisati: http://paste.ubuntu.com/1028461/ [11:57] ppisati: I never get to the installer, those are the env vars for uboot [11:58] gema: wait [12:04] gema: i've that too, but it shouldn't use that info [12:05] ppisati: then why is it failing there for me? [12:05] gema: did you dd it again with bs=64k? [12:05] ppisati: not yet [12:05] gema: try it [12:06] and the loader prompt, execute 'printenv' and paste it somewhere so i can compare it [12:06] *at [12:09] ppisati: changing the mmcrootfstype to ext4 makes it boot (I tried before rewriting the card) [12:10] gema: so you've a different boot environment [12:10] where does your board come from? [12:11] ppisati: http://paste.ubuntu.com/1028479/ <- the only thing removed is bootscr [12:11] my board came from pgraner, on the same batch as yours, I believe [12:12] ppisati: are we happy for me to try with bs=64K ? [12:12] no [12:12] Iok [12:12] ok [12:12] don't do that [12:12] ok [12:13] I only have a working SD [12:13] why you removed bootscr? [12:13] that's the bootscript present on the sd [12:13] ppisati: to be able to see dmesg [12:13] serial output [12:13] ok, but then you are using the env of you board that is wrong [12:13] if you want serial output [12:13] you need to change the file boot.scr on the first partition of the sd [12:14] taking out 'quiet splash' [12:14] and adding console=ttyO2,115200 [12:14] boot.scr is created with mknewimage [12:14] wait [12:14] ppisati: ok [12:14] ppisati: I am going to document that in our wiki [12:15] gema: mkimage -A arm -O linux -T script -C none -a 0 -e 0 -n "Ubuntu 10.10" -d ./boot.cmd ./boot.scr [12:16] gema: and boot.cmd is plain text file containing onle the text part of the boot.scr you find on the sd card/first partition [12:16] gema: so, open a text editor, cut&paste from boot.scr ONLY the text part [12:16] gema: delete 'quiet splash' [12:17] gema: add "console=yyO2,115200" [12:17] gema: write the file, close the editor [12:17] gema: execute [12:17] gema: mkimage -A arm -O linux -T script -C none -a 0 -e 0 -n "my Ubuntu boot.scr" -d ./$TEXTFILE ./boot.scr [12:18] gema: copy the resuling boot.scr to the sd card/first partition [12:18] gema: of course, do a backup oif the boot.scr found there [12:18] gema: just in case [12:19] ppisati: ack [12:20] gema: try by yourself, so we can double check it [12:20] * ppisati is hungry [12:21] ppisati: ack, go for lunch, it is going to take me a while... where do I do the mkimage? [12:27] ppisati: ok, on it, follow your instructions helps x) [13:09] ppisati: care to send a quick pull request for your omap4 config sync? [13:12] ppisati, all is good, trying to write the boot.scr to my card it started to behave weirdly [13:12] ppisati: camera to the rescue, all it good I can boot the image and install it [13:13] ppisati: thanks ! [13:16] ogasawara: i'll do [13:23] ogasawara: i did some more config sync, let me test that and then i'll pull req [13:23] ppisati: ack === mfisch` is now known as mfisch === mfisch is now known as Guest83117 [14:04] hi, kentb would need ALSA 1.0.25 linux-backports-modules on top of oneiric, any chance we can provide that to him? [14:12] ppisati, fyi, omap3 fails with an USB oops and i dont get any display output once the kernel started (i get the yellow u-boot screen though) [14:12] (note also that i'm on vacation, just doing some quick image tests) === Guest83117 is now known as mfisch === mfisch is now known as Guest28186 [14:14] bug 1010024 [14:14] Launchpad bug 1010024 in linux "omap fails to boot with oops in qantal alpha1 " [High,New] https://launchpad.net/bugs/1010024 [14:16] ogra_: i know video is broken on omap3 [14:17] ogra_: let me try the daily img for it [14:17] well, apparently USB too :) [14:17] at least that oops comes up right after USB is initialized it seems [14:19] uhm [14:19] let's see [14:32] * cking gives his network a poke [15:14] * ogasawara back in 20 [15:17] network bites cking for poking it [15:17] yep, it's a yappy little fella [15:20] ogra_: actually boots here, it reaches the installer [15:20] ogra_: but since video is broken, can't go ahead [15:21] ogra_: server images shoudl be fine (if we ever build it for omap3) [15:21] actually the installer dies cause X can't open a screem [15:21] and then drop you in # shell [15:46] ppisati, did you happen to see this one: bug 1010024 [15:46] Launchpad bug 1010024 in linux "omap fails to boot with oops in qantal alpha1 " [High,Incomplete] https://launchpad.net/bugs/1010024 === dileks_ is now known as dileks [16:18] jsalisbury: kind of saw it, it actually boots here (i was talking about it with ogra above) [16:18] ppisati, cool thanks. Just wanted to give you a heads up. [16:18] jsalisbury: but yes, video is broken on omap3, so omap3 desktop image for A1 are dead [16:18] ppisati, i'll do a bit research tomorrow when i'm not in vacation mode [16:19] ogra_: don't worry, A1 is done [16:19] ogra_: A1 is the target now [16:19] not worrying about A1 but about the images in general :) [16:19] A1 is done indeed, we'll only release omap4 [16:20] omap3 in the past didnt init the graphics when you didnt have the right boot options, thats what i planned to play with ... and i didnt do more than one test, might be that the oops doesnt show on every boot [16:21] ack [16:22] go have a Weizenbier for me [16:30] * ogra_ will do :) [16:47] jsalisbury: ping [17:09] henrix, pong [17:10] jsalisbury: hi! i was trying to add the series (precise and quantal) to bug #1010022 [17:10] Launchpad bug 1010022 in linux "Ext3 filesystems converted to ext4 can get erroneously reported as containing errors" [Medium,Triaged] https://launchpad.net/bugs/1010022 [17:10] jsalisbury: any idea how to do that? [17:10] infinity, apw: bah, so as I'm checking the linux-firmware 1.81 build that I've just uploaded to quantal-proposed I see I managed to typo the bug # in the debian/changelog (the git tree is correct). what's the preferred way to resolve? upload a new 1.82 version which only corrects the changelog and then make sure we just copy the 1.82 version to the release pocket? [17:10] jsalisbury: i guess i've done that before, but can't remember how :-/ [17:11] henrix, I see your nominations. I can approve them. I think I can approve them because I'm in the release team, which you may be able to subscribe to [17:11] ogasawara: Nah, just fix it in git retroactively, and it'll be right on the next upload. [17:11] ogasawara: And go close the bug by hand. [17:11] infinity: ok cool [17:11] henrix, I clicked approve, can you check on your end to see if it worked? [17:12] jsalisbury: oh, so i need approval for that! that means i've never done that before :) [17:12] jsalisbury: will check [17:12] ogasawara, yeah what infinity said. once that version is in -proposed its used up anyhow [17:12] ogasawara: (If it were a stable release, I'd punt it from proposed and make you try harder, but this isn't really the devel workflow) [17:12] jsalisbury: yeah, thats it. thanks :) [17:12] henrix, np [17:47] Oh cute, my ~ was never removed from zinc, I just lots access when I was gone. [17:47] Not that there was anything in it other than dotfiles... [18:01] ogasawara: Am i right in saying a new snapsnot of linux-firmware will resolve.. [ 21.388655] microcode: failed to load file amd-ucode/microcode_amd.bin [18:02] ie, should i not bother raising a bug or investigating [18:39] ogasawara: how goes the rebase? [19:43] RAOF: hi! do you know if there's any progress on making DP output work under nouveau with NVD9 based GPUs (e.g., the 4200M in the T*20 thinkpads)? [22:40] htorque: As far as I'm aware, that should be fixed somewhere in the nouveau git pipeline. My T420s died before I could test out darktama's patch, though. I'd guess it should work in 3.5, if it doesn't already work in 3.4. [22:42] RAOF: thanks for the info, i'll give 3.4 a go then.