=== shuduo is now known as shuduo_afk [07:50] moin === smb` is now known as smb [08:24] morning [09:09] * apw yawns === defiantredpill is now known as user8765 === henrix_` is now known as henrix === lool- is now known as lool [09:48] Hello, I am trying to build a driver and I get an implicit function declaration error for __netdev_printk [09:48] I think this function has been removed from the 3.7 kernel and I'm curious if it has been replaced or changed. [09:58] Deleted the function call and the driver compiles fine. It sucks that the driver hasn't found its way into the kernel. It is the ethernet driver for newer atheros cards. === yofel_ is now known as yofel === melkor_ is now known as melkor [10:54] brb [12:39] Hi Guys, How to find one particular process how many open files are there, any way to identify? [12:40] dhanasekaran: seems like this might not be the correct channel for that question. Did you mean #ubuntu? [12:42] melkor: it's process related right? process guys talk to libc. The kernel must know the list of open files right. [12:43] dhanasekaran: I'm not arguing with you I'm just offering you a suggestion that might improve your chances of getting help. [12:43] melkor: thnaks. === bambee is now known as rperier [13:19] smb, rebooting gomeisa for _something_, but can't remember what. [13:19] rtg, ok, I am off [13:21] apw: yesterday I asked one question, How to find particular path, available or not? You said no available in Linux tree. I agree. I want to know how to find it. It's helpful my further searches please guide me, I already download code from github [13:22] dhanasekaran, from what i could see that patch was only on the mailing list, i think your link was to patchworks [13:23] dhanasekaran, though it was also 3/3, ie the third one of a set and likely no use on its own [13:23] can i find patch history from gitk [13:26] if a patch is only on the mailing list, then it is not in git, therefore not in gitk === jdstrand_ is now known as jdstrand === henrix is now known as henrix_ === henrix_ is now known as henrix [14:06] rtg, did the something turn out to be a pain? [14:06] smb, gomeisa not back ? [14:07] rtg, sort of... just not letting me in [14:07] crud, better go check === ghostcube_ is now known as ghostcube [14:21] smb, gomeisa is back [14:21] rtg, thanks === henrix_ is now known as henrix === henrix is now known as henrix_ === henrix_ is now known as henrix === kentb-out is now known as kentb === henrix is now known as henrix_ === henrix_ is now known as henrix [15:41] this smatch run is taking forever === henrix is now known as henrix_ === henrix_ is now known as henrix [16:14] ** [16:14] ** Ubuntu Kernel Team Meeting - Today @ 17:00 UTC - #ubuntu-meeting [16:14] ** [16:35] rtg, so you can't hear me, eh? [16:35] jsalisbury, nope [16:36] rtg, whoops, I was wondering why it was so quiet :-) [16:36] jsalisbury, still no sound [16:36] rtg, ok, let me see what broke [16:40] jsalisbury, you are screwed today [16:41] rtg, yup. I'll try to figure out what happened after the IRC meeting. [16:42] 'ubuntu-bug linux' fails on raring, claims -5-generic is 3rd party, known? [16:45] check aptitude and see if it's in the obsolete or locally created packages section, if it isn't, i dunno :] [16:47] ohsix: indeed, -6 is out :) [17:00] ## [17:00] ## Meeting starting now [17:00] ## === jsalisbury changed the topic of #ubuntu-kernel to: Home: https://wiki.ubuntu.com/Kernel/ || Ubuntu Kernel Team Meeting - Tues February 19th, 2013 - 17:00 UTC || If you have a question just ask, and do wait around for an answer! [17:17] * ppisati goes for some sweating - later... [17:50] henrix, ok ... looks right now [17:50] apw: \o/ [17:51] apw: any other thing i missed there? [17:51] nope, the ' thing was the primary problem, then the raring things were simply sha1s coming available later [17:52] apw: ah, ok. thanks a lot for your help [18:02] bjf, henrix, herton bug 1119809 is a regression in 3.2.0-38.59 that is affecting allot of people. The issue doesn't happen in 3.2.0-37.58 [18:02] Launchpad bug 1119809 in linux (Ubuntu) "Desktop login freeze with kernel 3.2.0-38.59" [Medium,Confirmed] https://launchpad.net/bugs/1119809 [18:03] jsalisbury, ack [18:03] bjf, I haven't started a bisect for it yet, but I can do so [18:04] jsalisbury: looking, thanks [18:27] * rtg -> lunch [18:45] jsalisbury, it seems I'm able to reproduce here, I'm starting a bisect here [18:46] herton, cool, thanks! [18:54] slangasek, we loaned you a mac. i thought it was so something would be fixed so we didn't have to produce +mac isos. i just noticed we are still producing them so i'm assuming you have not finished that work? [18:54] bjf: correct, I have not :/ [18:54] you need the macs back? [18:55] slangasek, nope, all good, just making sure we were not building +mac isos unnecessarily [18:55] sadly no [19:20] * henrix -> EOD === henrix is now known as henrix_ === e11bits_ is now known as e11bits [20:13] * rtg -> EOD === hggdh_ is now known as hggdh [20:58] stgraber, can you try out the kernel at http://people.canonical.com/~sforshee/lp1098216/linux-3.8.0-6.11~lp1098216v201302121943/ on your x230? [20:59] this is a different approach to the backlight problem [21:00] sforshee: yep, will do [21:00] stgraber, thanks. Be sure you don't boot with acpi_osi="!Windows 2012". [21:05] sforshee: oh, I see, you're doing it the Windows way then ;) set all the values in the array from the current one to the one we actually want [21:06] stgraber, yep. That's what I was asked to do when I tried to get the quirks upstream. [21:07] stgraber, what you'll probably see is that it mostly works, but in one or two spots the brightness won't actually change when you adjust it through the desktop [21:07] g-s-d divides the range up into 20 steps but there are only 16 valid values [21:08] I rarely use the GUI tool, I usually just use the hotkeys which give you 8 levels IIRC [21:08] you only get the 20 if you go through the control center I believe [21:08] hmm, okay. Those must be generating backlight events rather than keypresses then. [21:09] stgraber, please try it both ways, I'd like to know the results [21:10] stgraber, also try directly writing the value in /sys/class/backlight/acpi_video0/brightness back into that file. That's an edge case I want to be sure gets tested. [21:11] ok [21:11] my laptop should be available for testing in a few minutes, finishing a couple of upstart and network-manager builds now ;) [21:15] stgraber, at your leisure ;-) [21:22] sforshee: bad news for you, doesn't work [21:23] sforshee: using the hotkeys in X (which are caught by g-s-d), I can't change the brightness level, it gets stuck at maximum [21:24] looking at the value of brightness, I can bring it down to 95 as the lowest value (actual_brightness remains at 100) [21:24] stgraber, drat [21:24] sforshee: going through the full gnome-control-center applet works though [21:25] hmm, okay [21:25] stgraber, when you say you can bring brightness down to 95, how are you doing that? [21:26] (note that reading actual_brightness is going to reset brightness to the actual value your firmware thinks the value is, which is the last valid value written) [21:26] fn+F8 which shows me the notify-osd brightness stuff. The slider appears as slightly less than maximum but there's no actual change in screen brightness [21:31] sforshee: I've been tracing gnome-settings-daemon a bit here and looking at the dbus calls, when going through gnome-control-center only SetPercentage calls are emited [21:31] when using the hotkeys, I see a GetPercentage call and a StepDown call instead [21:32] so depending what file GetPercentage is reading, this may explain why it gets stuck [21:32] stgraber, run acpi_listen and press your hotkeys and tell me what output you get [21:33] F8 is: video LCD0 00000087 00000000 [21:33] F9 is: video LCD0 00000086 00000000 [21:33] that's what I thought [21:34] so acpi_video handles those events internally to adjust the brightness [21:34] unfortunately when it does that it's calling _BQC to ask the firmware what it thinks the brightness is [21:34] so it's always getting the last value that actually applied [21:35] so it that explains why it's trying to set the same value over and over again without any actual result [21:35] s/it// [21:35] yeah, it always tries to get the next larger or smaller value, but when it writes that to the firmware the value just gets discarded [21:36] the next time it asks it just gets the same thing it got the previous time [21:36] stupid firmware [21:36] so the question then is does acpi_video really need to ask, or can it just use the cached value it already has [21:37] do we have firmware that messes with the value by itself? if not, then cached should be safe [21:38] (thinking of some of those machines with ambient light sensors, not sure whether it's just exposed to the OS or if the actual brightness change is done by the firmware) [21:38] I don't know, the code is way more careful than I'd think it needs to be [21:38] it doesn't even assume that the value it got from the firmware is one of the values the firmware said is valid [22:06] stgraber, even if I fix this bug your hotkeys are going to work terribly. They're going to cycle you through 100 brightness values, most of which do nothing. [22:06] great... [22:07] anyway I've got a patch to fix it already, so I'll spin a build to verify that I understand the problem correctly. Then I'll go back to the upstream folks and describe the problems with the solution they're asking for. [22:18] Hey guys, I want to add data from a test I did for the PowerManagementALPM. How can I add this to the wiki? [22:39] stgraber, http://people.canonical.com/~sforshee/lp1098216/linux-3.8.0-6.11~lp1098216v201302122215/ [22:39] let me know if that gets the hotkeys working, even if it is suboptimal [22:40] sforshee: ok. I'll get you some results tomorrow. I'm on 3G for the rest of the day so I'll avoid any massive download :) [22:41] stgraber, np === kentb is now known as kentb-afk [23:47] sforshee: hey, looks like your kernel bricked my machine... [23:48] sforshee: or rather, it apparently failed to suspend, when I noticed it was rather hot already and had a nice kernel panic on the screen. Since it won't boot at all, not getting anything from the firmware