| ppisati | moin | 06:48 |
|---|---|---|
| === smb` is now known as smb | ||
| smb | morning | 07:19 |
| === deffrag_ is now known as deffrag | ||
| ericm|ubuntu | ppisati, what's the solution for irq_to_gpio() being messed up in 3.2 kernel, which may have impact on several ARM flavours? | 07:28 |
| ppisati | ericm|ubuntu: messed? | 07:47 |
| ppisati | ericm|ubuntu: IIRC every platform provides its own implemantation | 07:47 |
| === dholbach_ is now known as dholbach | ||
| dholbach | hey | 08:25 |
| dholbach | how do I debug a kernel which crashed (no magic sysrq, had to turn off and back on again) - nothing in /var/log/syslog? | 08:25 |
| ppisati | smb: apw if off today, right? | 08:26 |
| smb | ppisati, and tomorrow | 08:26 |
| smb | dholbach, not very much | 08:26 |
| dholbach | that sucks :-( | 08:26 |
| smb | dholbach, You had flashing caps (if there is still a led for it)? | 08:26 |
| dholbach | nope, it didn't flash | 08:27 |
| smb | So it might not have gone to the crash handler. Did you try to ping/ssh into it and it was dead, too? | 08:28 |
| dholbach | no, I didn't have another machine to ping it handy | 08:29 |
| dholbach | but changing to the console, magic sysrq and ctrl-alt-del weren't available | 08:29 |
| smb | Hm, ok. Wait you were able to change to the console? | 08:30 |
| dholbach | no | 08:30 |
| dholbach | none of the above worked | 08:30 |
| dholbach | I had to power off and back on again | 08:30 |
| smb | dholbach, ok, so it still could be a lock up which affected the input handler as well. | 08:31 |
| dholbach | right, I guess so | 08:31 |
| smb | But the problem is without anything else it could be that or something else. Just not possible to say anything for sure... :/ | 08:32 |
| smb | dholbach, So the only thing you can do is trying to observe whether it repeats and if it does whether there is any pattern to it. | 08:34 |
| dholbach | ok, let's say I have a machine to ssh/ping it with next time | 08:34 |
| dholbach | what are the next steps, if I 1) can ssh into it, 2) can't | 08:34 |
| dholbach | it doesn't happen very often, so bisecting kernels will be hard to do | 08:35 |
| dholbach | is there any debug information I can try to dig out? | 08:35 |
| dholbach | (this time I didn't do anything out of the ordinary: IRC, some terminals) | 08:35 |
| smb | If it pings, it is at least alive on a low level. If ssh succeeds its a lockup somewhere in X/input, ssh responding but login not succeeded, may point to a low-memory issue | 08:36 |
| dholbach | low memory meaning I ran out of memory or that there's a bug in memory handling somewhere? | 08:37 |
| smb | right | 08:37 |
| dholbach | I can't imagine I ran out of memory this time | 08:37 |
| dholbach | but I'll check the next time | 08:38 |
| dholbach | is there a way to have a kernel with more debug output or something? | 08:38 |
| smb | That was more related to the case that you can ssh but the login does not work. | 08:38 |
| dholbach | ah ok | 08:39 |
| smb | The problem with that kind of bug is that if shutting down is the only solution then more debug does not help either | 08:39 |
| dholbach | at least some of it might have gone to /var/log/syslog, no? | 08:39 |
| smb | You can boot with "debug" (instead of quiet) in the kernel commandline from grub | 08:39 |
| dholbach | alright, will do | 08:39 |
| smb | Most of the times if there is a hard crash, writing to a filesystem stops before that (its not synchronous) | 08:40 |
| dholbach | ok | 08:42 |
| ericm|ubuntu | ppisati, omap provides one? | 08:43 |
| dholbach | thanks smb | 08:44 |
| dholbach | I'll see what I can dig out | 08:44 |
| ericm|ubuntu | ppisati, git grep 'irq_to_gpio' arch/arm/mach-omap2/ shows no result | 08:45 |
| smb | dholbach, Welcome. The primary step right now would be to find out whether it was really a crash (able to ping/ssh or not). The next steps would then depend on that outcome | 08:45 |
| * dholbach nods | 08:45 | |
| ppisati | ericm|ubuntu: sorry, omap3 in master is busted... :P | 09:08 |
| ericm|ubuntu | ppisati, which ARM flavour is not busted so far? | 09:08 |
| ppisati | ericm|ubuntu: err.. i think you are right, and i'm pretty sure i hit this in the omap4 branch too | 09:08 |
| ppisati | ericm|ubuntu: :) | 09:08 |
| ericm|ubuntu | ppisati, I think the issue could have been fixed in Q (3.4) by the way, we may have to identifiy and backport them | 09:09 |
| ppisati | ericm|ubuntu: uhm | 09:11 |
| ppisati | ericm|ubuntu: drivers/gpio/gpio-ep93xx.c:#define irq_to_gpio(irq) ((irq) - gpio_to_irq(0)) | 09:13 |
| ppisati | ericm|ubuntu: and gpio_to_irq() is defined for mach-omap2/$board | 09:13 |
| ericm|ubuntu | ppisati, that's actually not a correct fix | 09:13 |
| ericm|ubuntu | ppisati, but not irq_to_gpio() | 09:13 |
| ericm|ubuntu | ppisati, all those drivers making use of irq_to_gpio() now failed in 3.2 | 09:13 |
| ppisati | ericm|ubuntu: do you have the commit where they retired irq_to_gpio for omap? | 09:14 |
| ericm|ubuntu | ppisati, I think it's a global one, removing all the specific APIs to asm-generic | 09:15 |
| ericm|ubuntu | ppisati, wait | 09:15 |
| ericm|ubuntu | ppisati, 7563bbf gpiolib/arches: Centralise bolierplate asm/gpio.h | 09:15 |
| ppisati | and i just found out another case where having vfat + nls_iso8859_1 builtin is mandatory... | 09:16 |
| ericm|ubuntu | ppisati, actually when doing the config comparison, I came up with the table - https://wiki.canonical.com/EricMiao/Highbank/ConfigReview/Ubuntu-3.2.0-25.40/5/PurpleLinesClassified | 09:19 |
| ericm|ubuntu | ppisati, where we do have many inconsistent config options between ARM flavours and X86 ones | 09:19 |
| ericm|ubuntu | ppisati, some are for no obvious reason | 09:20 |
| ppisati | ericm|ubuntu: i know, it's on my todo list | 09:20 |
| ericm|ubuntu | ppisati, ok - just let know what I can help on that | 09:20 |
| ppisati | ericm|ubuntu: https://wiki.ubuntu.com/Kernel/Dev/pisosandbox | 09:20 |
| ppisati | ericm|ubuntu: but this one is only for omap4 | 09:20 |
| ericm|ubuntu | ppisati, and btw - apw is working on config alignment between flavours | 09:20 |
| ppisati | ericm|ubuntu: yep | 09:21 |
| janimo | apw, do you know if adjustments were made to kernel package builds to cope with some headers moving to multiarch dirs? I get sys/types.h not found when building scripts/basic/fixdep in the armadaxp kernel | 09:42 |
| orated | Hey ppisati | 09:49 |
| orated | I'm using Beagleboard B4 and for getting ethernet working, a usb-ethernet adapter is connected. I'm not able to get ethernet working on it. Is there any specific kernel module required to be loaded for it to work? My connection setup - Both beaglebaord and USB hub are externally powered with 5V supply. USB ethernet adapter connected to the HUB | 09:51 |
| ppisati | orated: if the adapter is supported, the module should be autoloaded | 09:53 |
| ppisati | orated: have you tried this adapter with another pc running ubuntu? | 09:53 |
| orated | Yes | 09:53 |
| orated | Hold on, I'll give you the details | 09:53 |
| ppisati | orated: does it work? which module does it use? | 09:53 |
| orated | ID 0fe6:9700 Kontron (Industrial Computer Source / ICS Advent) DM9601 Fast Ethernet Adapter | 09:54 |
| orated | driver:dm9601(configuration: autonegotiation=on broadcast=yes driver=dm9601 driverversion=22-Aug-2005 duplex=full firmware=Davicom DM9601 USB Ethernet ip=192.168.10.110 link=yes multicast=yes port=MII speed=100Mbit/s) | 09:54 |
| orated | Same setup on host system with the hub | 09:54 |
| ogra_ | do you see it with lsusb | 09:54 |
| ogra_ | ? | 09:55 |
| ogra_ | (on the beagle that is) | 09:55 |
| ogra_ | note that nobody in the ubuntu dev team has tested on beage B models in years (they dont fulfill the minimal reqs.we usually only test XM) | 09:55 |
| orated | On the beagle, no | 09:56 |
| orated | ah | 09:56 |
| ogra_ | a server install should still be possible on the rev. B though | 09:56 |
| orated | Could you guide me on that? | 09:56 |
| ogra_ | (dont expect a great experience with 600MHz and 128MB though) | 09:57 |
| orated | I only need network connection to install and update packages | 09:57 |
| ogra_ | hmm, https://wiki.ubuntu.com/ARM/Server/Install used to have a section for omap3 too, seems it is gone, talk to #ubuntu-server, seems they removed that section (especially all the special setup you hgave to do with old beagles) | 09:59 |
| orated | I tried sudo dhclient usb0 and then ifconfig usb0 192.168.1.5 netmask 255.255.255.0 and they ran without errors | 10:00 |
| orated | ok, I'll try | 10:00 |
| ogra_ | well, does a plain ifconfig show the usb0 device ? | 10:01 |
| orated | yes | 10:01 |
| ogra_ | so your network should work then after you set up proper routing | 10:02 |
| orated | proper routing, like? I'm new to nwtwork related tasks | 10:02 |
| ogra_ | look on your desktop computer (if its in the same network) with "route -n" what the default GW points to | 10:03 |
| orated | Yes, in the same network | 10:03 |
| ogra_ | then use "sudo route add efault GW <IP you found above>" | 10:03 |
| ogra_ | *default | 10:03 |
| ogra_ | probably gw needs to be non capitalized ... | 10:04 |
| * ogra_ doesnt use this very often | 10:04 | |
| ogra_ | after that try to ping the gateway IP | 10:04 |
| orated | ogra_: http://paste.ubuntu.com/1022830/ | 10:12 |
| orated | Is that correct? | 10:12 |
| ogra_ | well, are you sure a plain ifconfig (without options) shows usb0 on the beagle ? also, if you use dhclient you dont need to set the IP manually with ifconfig afterwards | 10:14 |
| ogra_ | (and if the dhclient command worked you shouldnt need to set the route either) | 10:14 |
| orated | ogra_: Yes, http://paste.ubuntu.com/1022834/ | 10:16 |
| orated | There are some garbage character coming on the serial terminal though | 10:16 |
| ogra_ | looks all fine, you should be able to ping the GW right after the dhclient command, weird that you cant | 10:17 |
| orated | Alright, now I would like to know if there is any need to load module like g_ether ? | 10:17 |
| ogra_ | no, g_ether is for USB->USB networking through an USB cable | 10:18 |
| orated | Ethernet gadget, ok | 10:18 |
| ogra_ | (it emulates the ethernet connection between two USB devices) | 10:18 |
| orated | Then is there any module required for usb-ethernet adapter? | 10:18 |
| ogra_ | no | 10:18 |
| ogra_ | not apart from the one that brings it up as usb0 | 10:19 |
| ogra_ | which apparently is already loaded on your beagle (else ifconfig wouldnt see a device) | 10:19 |
| ogra_ | so there might be a bug with that driver or a wrong/outdated firmware etc ... i'll leave you with ppisati to debug that part :) | 10:20 |
| orated | In desktop with USB hub it uses driver:dm9601 and without usb hub, its not getting connected. So in beagleboard, is there any module to load to be use that driver? | 10:20 |
| orated | hmm, thanks for your help :) | 10:20 |
| ppisati | orated: i bet you need the dm9601 kernel module | 10:21 |
| ogra_ | well, if ifconfig sees usb0 that driver should be in use already | 10:21 |
| ogra_ | lsmod|grep dm9601 | 10:21 |
| ogra_ | that should show the module being loaded on the beagle | 10:21 |
| ppisati | ogra_: isn't it the usb network? | 10:21 |
| orated | no, nothing ogra_ | 10:22 |
| ppisati | orated: we had the usb0 as a nic till maverick | 10:22 |
| ppisati | orated: sudo modprobe dm9601 | 10:22 |
| ppisati | orated: which kernel are you using? | 10:22 |
| ogra_ | (and which image did you install) | 10:22 |
| orated | omap 3.2.17-11 #1 SMP Tue May 15 12:16:43 UTC 2012 armv7l | 10:22 |
| ppisati | orated: find /lib/modules/`uname -r` -name \*9601\* | 10:23 |
| orated | one moment please | 10:26 |
| orated | ppisati: /lib?mdules?3.217-11?kernel?drivers?net?usb?dm9601.k | 10:28 |
| orated | There are still garbage issues, I restarted for the same | 10:28 |
| orated | Maybe its the cable or loose connection | 10:28 |
| ppisati | orated: ok, then you have the module | 10:28 |
| ppisati | sudo modprobe dm9601 | 10:28 |
| orated | Yes, I did that | 10:28 |
| ppisati | uhm | 10:29 |
| ppisati | can you bypass the usb hub> | 10:29 |
| ppisati | ? | 10:29 |
| orated | And its listed now in lsmod | 10:29 |
| orated | bypass as in? | 10:29 |
| ppisati | and you didn't get anything in console, right? | 10:29 |
| orated | nope, just the prompt after the command | 10:30 |
| ppisati | ack | 10:30 |
| ppisati | then unplug it from the usb hub and attach it directly to the board | 10:30 |
| orated | One moment, I think the adapter got disconnected | 10:30 |
| orated | Yes, restarting the whole setup. .. One minute | 10:31 |
| === orated_ is now known as orated | ||
| orated | ppisati: Regarding OTG port on BB, I've connected a OTB to USB which is then connected to USB female to female adapter and then it goes to USB hub or USB storage device/usb-ethernet. The problem of connecting it directly without the hub is that, it doesn't get powered up. .. | 10:44 |
| orated | Regarding the USB hub, its Belkin self powered 4 ports USB2.0 if that helps | 10:45 |
| ppisati | orated: well, the problem is that your board doesn't sense any "usb thing" attached to it wrt to your usb nic | 10:47 |
| orated | And I tried it again with the hub, its still outputting nothing on sudo modprobe dm9601, falling back to prompt | 10:47 |
| ppisati | orated: so we need to get rid of the usb hub | 10:47 |
| orated | It requires to be independently powered | 10:47 |
| ppisati | orated: wait | 10:47 |
| ppisati | orated: how do you power the board? | 10:48 |
| orated | Also , other storage devices comes up with same behaviour. The BB is powered using 5V adapter | 10:48 |
| orated | And the hub is also powered using another 5V adapter | 10:48 |
| orated | BTW I've noticed that if I connect any storage device to the hub, it doesn't show up in /media | 10:49 |
| ppisati | ok, the if the board is powered indipendently, why can't you unplug the usb hub and attached the usb nic to the board? | 10:49 |
| ppisati | *attach | 10:49 |
| orated | Oh, you wanted to know if OTG is unused or not. Yes, I can unplug the usb nic from hub and conenct to board but its not getting its not getting sufficient power to start it. There is a LED on the nic which glows when RJ45 is connected | 10:51 |
| orated | Anyway, I'll try connecting it directly again | 10:51 |
| orated | Nic removed from the hub, connected to board, no led on nic | 10:52 |
| ppisati | any msg on console? | 10:54 |
| ppisati | dmesg? | 10:54 |
| orated | No, the serial console hanged | 10:54 |
| orated | Its not accepting any input. Hold on please | 10:54 |
| orated | ppisati: Back, no no change in dmesg on connecting/disconnecting nic directly to the board | 11:01 |
| orated | ppisati: Do you think its due BB being not sufficiently powered? | 11:04 |
| ppisati | orated: could be | 11:04 |
| ppisati | orated: i've a dm9601 adapter, let me try it with my bbxm | 11:05 |
| orated | ppisati: But then what's the problem when USB hub is used? | 11:05 |
| orated | ppisati: And secondly, I don't understand why usb storeage devices don | 11:06 |
| orated | n't show up in lsusb//media when they are connected to the hub | 11:06 |
| orated | storage* | 11:06 |
| orated | Is 12.04 LTS server image supported for BB or should I try 11.10? | 11:16 |
| ogra_ | as i said before, we all only test on beagle Xm since it is the only beagle that fulfills the minimal requirements for ubuntu | 11:17 |
| ogra_ | so getting it to run on a beagle rev. B is really a matter of luck | 11:18 |
| jMCg | I got 10.10 in a KVM in an init=bash: http://dpaste.com/755097/ -- can someone help me debug and make it boot? | 11:20 |
| jMCg | s/10.10/10.04/ | 11:20 |
| ogra_ | seems to boot just fine | 11:22 |
| ogra_ | (for booting with init=bash at least) | 11:23 |
| jMCg | ogra_: that's true, but when I take that away, it chokes. I've pasted --verbose and --debug in this channel, let me reproduce those. | 11:26 |
| jMCg | http://sprunge.us/iGaU << --debug | 11:27 |
| orated | ppisati: Is dm9601 adapter working for you in BB xm? | 11:29 |
| orated | Unfortunately, xm is on back order here and can be available only after a month | 11:37 |
| ppisati | [ 3757.129302] usb 1-2.2: new full-speed USB device number 8 using ehci-omap | 11:52 |
| ppisati | [ 3757.365447] dm9601 1-2.2:1.0: eth1: register 'dm9601' at usb-ehci-omap.0-2.2, Davicom DM9601 USB Ethernet, 00:10:13:50:a3:43 | 11:52 |
| ppisati | [ 3757.373413] usbcore: registered new interface driver dm9601 | 11:52 |
| ppisati | orated: ^^ | 11:52 |
| ppisati | works here | 11:53 |
| ppisati | orated: what's the output of your power adapter used with BB? | 11:54 |
| ppisati | orated: i'm using a 5v 3.8A adapter | 11:54 |
| orated | 5V 900mA for BB, 5V 2.6A for hub | 11:55 |
| ppisati | ah | 11:57 |
| ppisati | orated: i doubt that's enough | 11:57 |
| orated | current rating, yes | 11:57 |
| ppisati | but when you attach stuff, it drains power | 11:58 |
| ppisati | orated: try with a beefier adapter | 11:58 |
| orated | ppisati: Yes, one moment. brb | 12:00 |
| bjf | ppisati: bug 951043 needs verification, can you take care of that ? | 12:27 |
| ubot2` | Launchpad bug 951043 in linux-ti-omap4 "Port OOM changes into do_page_fault for arm" [Medium,Confirmed] https://launchpad.net/bugs/951043 | 12:27 |
| ppisati | bjf: ack | 12:28 |
| orated | ppisati: Is it possible to install xfce/gnome and opencv offline on SD card and run them on BB? | 12:48 |
| ppisati | orated: an a BB board? no way, it'll swap like hell | 12:49 |
| ogra_ | on a BB rev. B ! | 12:49 |
| ogra_ | (600MHz/128MB ram) | 12:49 |
| ogra_ | probably lxde or a standalone window manager might work ... | 12:50 |
| ogra_ | xfce and gnome are surely to big for that | 12:50 |
| ppisati | bjf: there's no mention of any test to validate that bug, i can confirm the patch is there and kernel boots, but not much more | 12:50 |
| orated | Afaik, its lxde>xfce>gnome>KDE in incresing order of memory consumption | 12:52 |
| orated | So, is it possible to install lxde and openCV offline on SD or atleast openCV? | 12:52 |
| ogra_ | yes, install qemu-user-static on your host PC ... | 12:53 |
| ogra_ | then *manually* mount the SD, copy qemu-arm-static from /usr/bin of the host into /usr/bin of the SD and you can chroot into the SD card | 12:53 |
| orated | ogra_: Does chroot requires that the host and guest should be of same architecture... arm-arm, x86-x86 ? | 12:54 |
| ogra_ | not with qemu-user-static | 12:54 |
| orated | Ah-ok | 12:54 |
| ogra_ | but you need to copy the binary from the host to the same place on the SD | 12:55 |
| === yofel_ is now known as yofel | ||
| tgardner | ogasawara, whats the story on 3.4.0-4.9 ? Is that the version you intend to upload ? | 14:11 |
| ogasawara | tgardner: it is, just wanted to get a quick powerpc build and will then upload. | 14:12 |
| tgardner | ogasawara, ok, I'll start the precise backport | 14:12 |
| ogasawara | tgardner: I just wanted to shove the closing commit in there in case I/you wanted to get a jump on the 3.5-rc1 rebase | 14:12 |
| ogasawara | tgardner: I have not yet pushed it to master | 14:13 |
| tgardner | ogasawara, nah, I think we can wait until -rc2 | 14:13 |
| ogasawara | tgardner: ack | 14:13 |
| tgardner | ogasawara, 'sides, I'm bailing for a week and don't wanna start something I can't finish today. | 14:14 |
| ogasawara | tgardner: I was also looking at the highbank meta package patch, is there a specific reason we mark the linux-headers-<flavor> packages as "Section: devel" rather than "Section: metapackages"? | 14:14 |
| ogasawara | tgardner: I notice it's only the headers meta packages which are marked devel, all other meta packages are marked as metapackages | 14:15 |
| tgardner | ogasawara, missed that. I'll fix it. | 14:19 |
| tgardner | ogasawara, ok, now I've read your comment for content. I don't think 'Section:' has any real impact in the Ubuntu archive, but I'll ask just to be sure. In the meantime I think its OK as its been that way for multiple releases. | 14:22 |
| ogasawara | tgardner: indeed, I don't think it should hurt anything and should be ok to fix up later if we want to | 14:23 |
| tgardner | ogasawara, its a simple change, so I might as well get it in before A1. gimme a minute. | 14:23 |
| tgardner | ogasawara, hmm, I wonder wtf 'Section: restricted/metapackages' means for powerpc* ? | 14:26 |
| ogasawara | tgardner: hrm, no idea there | 14:27 |
| tgardner | why not just 'Section: metapackages' ? | 14:27 |
| ogasawara | tgardner: that sounds fine to me | 14:27 |
| ogasawara | tgardner: I've been trying to find a defined list for Section: but have been unsuccessful in my searching | 14:28 |
| tgardner | ogasawara, ok, I pushed both changes, so I guess we'll just see what breaks. | 14:29 |
| ogasawara | tgardner: ack | 14:29 |
| infinity | Anyone know why the ti-omap4-tools package was dropped from linux-ti-omap4? | 14:40 |
| infinity | tgardner / apw: Also, apparently it's our PlusOneMaint month. Had you guys sorted out how you plan to split your time? | 14:40 |
| tgardner | infinity, all of the omap tools should come from the main kernel. | 14:42 |
| tgardner | infinity, apw will get started on +1 maint after the UK holiday. perhaps wednesday ? | 14:43 |
| infinity | tgardner: I thought tools had to precisely match ABI or some such, hence the extra packages from out-of-tree builds? | 14:43 |
| infinity | tgardner: (This was always the case in previous releases, but I'm happy to be told that was a mistake) | 14:43 |
| tgardner | infinity, perf is the only tool that is ABI sensitive. lemme review... its been awhile. | 14:43 |
| infinity | tgardner: Right, and perf is part of -tools. :P | 14:44 |
| * ogasawara back in 20 | 14:44 | |
| tgardner | I'm not sure we're even building perf 'cause it was causing FTBSs, but that was a bit ago during early -rc's | 14:44 |
| * infinity checks the linux-ti-omap4 changelog. | 14:44 | |
| infinity | I see no mention in the changelog of dropping the tools build, it just disappeared with no word. | 14:47 |
| tgardner | infinity, looks like I turned it off during the early dev cycle because of FTBS. 'UBUNTU: [Config] armhf: do_tools=false' | 14:50 |
| infinity | tgardner: Oh, was that in the changelog and I'm just blind? Or just in the git logs? | 14:52 |
| tgardner | infinity, just the git logs | 14:53 |
| infinity | Anyhow. Not world-ending for now, just makes the tools metapackage uninstallable, but would be nice to restore it if and when it can be made to work. | 14:53 |
| tgardner | infinity, working on it... | 14:53 |
| infinity | Or maybe it'll all become moot if omap4 upstreaming keeps going well, and we can build 3.5 from mainline. | 14:53 |
| * infinity crosses his fingers. | 14:53 | |
| tgardner | infinity, ain't gonna happen. there is still a pile of BSP patches for ti-omap4 | 14:54 |
| infinity | At which point, to keep ppisati from getting bored, we'll have to get him to do an mx6 or omap5 or something. ;) | 14:54 |
| tgardner | 5 is coming | 14:54 |
| infinity | tgardner: Shame. I was led to believe it was getting close. Though, maybe we're still carrying some "won't ever go upstream, don't even try" bits. | 14:55 |
| ppisati | it's already here, i didn't turn it on yet | 14:55 |
| orated | Sorry, got disconnected before. Thanks ppisati and ogra_ for help | 14:56 |
| infinity | ppisati: Does that mean you have a Panda5, or have you just been doing omap5 blind and hoping it will boot? ;) | 14:57 |
| ppisati | infinity: it means the BSP supports both omap4 and omap5 | 15:08 |
| ppisati | infinity: but i din't enable the omap5 side yet | 15:08 |
| infinity | ppisati: Ahh. | 15:09 |
| ogra_ | thats what they tell you :P | 15:10 |
| dreks | I'm trying to set up L2TP over IPSec and I've got the IPSec conn established. When I try to start the L2TP connection with xl2tpd, the kernel log starts filling and PC becomes unusable. I have tested the exact same setup on in virtualizatin and its flawless. I believe it has something to do with my ethernet hardware/kernel combination. | 15:11 |
| dreks | I've posted a portion of my kern.log here, http://paste.ubuntu.com/1023244/ | 15:12 |
| dreks | I've tried grabbing the Realtek driver for the integrated adapter straight from their site and using that. It produced the same result. | 15:12 |
| ogra_ | dreks, your message might become readable if you change your colors ... | 15:12 |
| dreks | Sorry. | 15:13 |
| dreks | That better? | 15:14 |
| ogra_ | yep | 15:14 |
| dreks | I'm trying to set up L2TP over IPSec and I've got the IPSec conn established. When I try to start the L2TP connection with xl2tpd, the kernel log starts filling and PC becomes unusable. I have tested the exact same setup on in virtualizatin and its flawless. I believe it has something to do with my ethernet hardware/kernel combination. | 15:14 |
| dreks | I've posted a portion of my kern.log here, http://paste.ubuntu.com/1023244/ | 15:14 |
| dreks | I've tried grabbing the Realtek driver for the integrated adapter straight from their site and using that. It produced the same result. | 15:14 |
| tgardner | dreks, try the quantal kernel from https://launchpad.net/~ubuntu-x-swat/+archive/q-lts-backport | 15:16 |
| dreks | ok, i can try that. Thanks. Do you see anything in the logs that might indicate kernel over driver problem? | 15:17 |
| tgardner | dreks, the r8169 driver has had lots of problems in the past, so I suspect you might be right. | 15:18 |
| dreks | Is there anywhere else I can look for driver related bugs? I'd like to find out whats to blame, kernel or driver | 15:20 |
| dreks | If I need to replace a ethernet card because the issue has not been resolved with the driver, I will do so, but I want to find evidence it's not the kernel | 15:21 |
| tgardner | dreks, you could try asking romieu@fr.zoreil.com (the maintainer). be sure to Cc: netdev@vger.kernel.org | 15:21 |
| dreks | thanks | 15:22 |
| === tgardner is now known as tgardner-afk | ||
| === tgardner-afk is now known as tgardner | ||
| * ppisati -> gym | 15:52 | |
| tgardner | ogasawara, so, you're happy with Ubuntu-3.4.0-4.9 ? I'll get the LTS backport started. | 17:45 |
| ogasawara | tgardner: I am, just pushed to master. Am prepping the upload right now. | 17:45 |
| tgardner | ogasawara, yep, saw that when I jst pulled. | 17:46 |
| tgardner | just* | 17:46 |
| === tgardner is now known as tgardner-afk | ||
| infinity | tgardner: So, linux-ti-omap4-tools-3.4.0-201 is now the only thing standing between me and beer on the archive health reports. ;) | 18:05 |
| infinity | ppisati: ^ | 18:05 |
| infinity | ppisati: Do you know if the tools can be re-enabled (or, if they need fixing, can we?) | 18:05 |
| tgardner-afk | infinity, I've got a test build going. if successful I'll upload later this afternoon | 18:06 |
| infinity | tgardner-afk: Ahh, fancy. Thanks. | 18:06 |
| * tgardner-afk is now really afk... | 18:06 | |
| infinity | ogasawara: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1005699 is preventing us from upgrading some of our buildds to precise. Is there a formal escalation process for these sorts of things? | 18:51 |
| ubot2` | Launchpad bug 1005699 in linux "tg3: reports "eth0: DMA Status error. Resetting chip.", fails to work" [High,Confirmed] | 18:51 |
| ogasawara | infinity: usually assigning it to the canonical-kernel-team in lp gets it on our radar, or pinging us directly like you've just done | 18:52 |
| infinity | ogasawara: Looks like it's fixed upstream, so should just be some tedious bisection back and forth. | 18:53 |
| ogasawara | infinity: ack, will get a resource on it and get it sorted | 18:53 |
| infinity | (Or perhaps not even that, if the changes to the driver are limited and obvious, I guess) | 18:53 |
| infinity | ogasawara: <3 | 18:53 |
| orated | ogra_: Still there? | 19:18 |
| Kano | hi | 20:59 |
| Kano | i see your quantal builds need libc 2.14 | 21:00 |
| Kano | is it possible to lower that to libc 2.13 as debian only uses 2.13 | 21:01 |
| jjohansen | ogasawara: so I looked into CONFIG_NF_CONNTRACK_PROCFS its obsolete, and should not be set (so no change) do you want an email to the list before I mark the wi done | 21:11 |
| ogasawara | jjohansen: if you could just send a quick email that'd be great. it'd be nice for historical tracking purposes. | 21:12 |
| infinity | tgardner-afk: Tools build for ti-omap4 seems to be missing a build-dep on flex. | 23:36 |
| infinity | (at least) | 23:36 |
| infinity | tgardner-afk: There's also the warnings about missing newt/gtk/python support due to not having those dev packages installed, but I'm guessing you don't turn those on for the master branch either? | 23:37 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!