=== JanC_ is now known as JanC === JanC is now known as Guest38106 === JanC_ is now known as JanC [13:25] Hello, is there any newbie bugs related to the kernel? something I need to start of with [15:13] apw: any ideas on the situation [15:18] green_, i have no scrollback, so i have no idea as to what we are talking about [16:34] apw: The whole output http://dpaste.com/2BDBVZW Do you think it's fine. [16:34] green_, that sounds like you enabled the src lines for the CD and not the archive [16:35] apw: I was trying to recompile kernel and run the command [16:35] apw: using instructions here https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel [16:36] apw: apt-get source linux-image-$(uname -r) [16:36] apw: I wanted to recompile the same kernel I am using and edit regd.c file in source [16:36] it is the ones with archive.ubuntu.com that you want to add/uncomment deb-src lines for for that to work [16:36] not the deb-src related to the non-existant cdrom [16:37] That line was not in the source file. There is only one line in the file [16:37] you then asked me to check whether there are other files [16:37] likely, if apt is finding our ubuntu repositories they must be listed somewhere [16:38] likely in /etc/apt/sources.list.d/* [16:39] I am running mint 18.1. there is /etc/apt/sources.list.d/official-package-repositories.list [16:39] That file has this line # Do not edit this file manually, use Software Sources instead. [16:40] you could do that then [16:41] all other lines are uncommented. None have deb-src; they all begin with deb and then an URL e.g. deb http://archive.ubuntu.com/ubuntu xenial main restricted universe multiverse [16:41] likely software source will add them if you enable sources [16:42] What would I do specifically in software sources? [16:43] green_, i think there is a 'source code' tickable, but i don't use it often [16:44] there is "enable source code repositories" [16:45] that sounds like the sort of ting [16:45] and then update cache? [16:47] apw: Thanks. that worked I think [17:02] apw, this is mint ... donnt expect something like "software sources" to work as in ubuntu ;) (or anything else for that matter) [17:05] it is all a guessing game that is for sure [17:05] ogra_: So it appears I downloaded the source for 4.8.0 and updates for 4.8.0-34.36 but I am running -28 specifically because it works with sound [17:05] ogra_: the idea was to enable chan 13 on the sound card by editing regd.c on this exact kernel which work correctly with the sound on a new kaby lake machine [17:07] In update manager, the -28 is listed as the first available. Would that mean that 4.8.0 source that I downloaded is -28 or could it be some other version [17:11] -28 who knows, that isn't anything recent [17:15] Is there any way to check in some file? Is this statement correct for mint also For example to obtain the source for the currently running kernel you can use the command: [17:15] apt-get source linux-image-$(uname -r) [17:16] doesnt mint have its own kernel package ? that would likely not be called linux-image-* [17:16] does "currently running kernel" mean the specific ver (-28) that I am running or just 4.8.0 without any specific version. [17:22] aha ... http://packages.linuxmint.com/pool/main/l/linux-kernel/ [17:22] so they call it linux-kernel- [17:23] cat /proc/version_signature [17:23] or it is unrelated and all bets are off [17:34] I am not really sure. I was pointed to Ubuntu when I asked about kernels. That list at http://packages.linuxmint.com/pool/main/l/linux-kernel/ ends at 4.4 while currently the latest are 4.8 [17:46] green_: I know little to nothing about how mint selects it's kernels [17:47] but the apt-get incantation gives you source for what you are running ... [17:49] I got this with the cat command Ubuntu 4.8.0-28.30~16.04.1-generic 4.8.6 [17:50] apw: so I don't need to check if the source is the same version as what I am running. I could just try to recompile and use those instructions [17:50] apw: for ubuntu [17:51] that is some achient version, but if the .DSC you have matches you are good [17:54] apw: I guess I could wait for 4.10 and try it instead since I am already recompiling. I am not sure what .DSC is [17:54] apw: or how I would check it matches [17:55] the versions is in the filename. [17:56] apw: You mean this linux-hwe-edge_4.8.0.orig [17:56] apw: of the downloades source? [17:58] there is one ending .dsc [18:01] apw: aha. actually no, it's 34.36 with .dsc and I am running 28. I also see this dpkg-source: warning: failed to verify signature on ./linux-hwe-edge_4.8.0-34.36~16.04.1.dsc [18:03] then this is not the one you want [18:03] apw: 3 files were downloaded: linux-hwe-edge_4.8.0.orig, linux-hwe-edge_4.8.0-34.36~16.04.1 and linux-hwe-edge_4.8.0-34.36~16.04.1.dsc [18:04] I think 36 did not work correctly with the sound, while 34 I did not try [18:07] well the command you are meant to run, should get the one you are running and no other [18:36] apw: ok so what do you think happened. I did run that exact command. Is it possible that the orig file is what I am running and the 34.36 are updates? [18:36] what was the command you ran [18:37] apt-get source linux-image-$(uname -r) pasted from https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel [18:38] and what does "uname -r" say [18:38] 4.8.0-28-generic [18:38] 4.8.0-28-generic is aslo in mint update manager as "active" [18:38] so run that in a new clean directory, and pastebin the output [18:40] I am running it. I see this line Picking 'linux-hwe-edge' as source package instead of 'linux-image-4.8.0-28-generic' [18:41] the source package for linux-image-* can be different, linux, liux-lts-*, linux-hwe, etc etc [18:43] is there a keyboard shortcut to select all text in terminal [18:44] http://dpaste.com/2JXCHTY [18:45] oh it is getting the latest version of the kernel on the assumption you are running the latest [18:47] apt-get source linux-image-$(uname -r)/$(uname -r) [18:47] green_, ^ does that get a different version ? [18:52] should I run this apt-get source linux-image-$(uname -r)/$(uname -r) [18:52] might work, you never knwo [18:52] i can't trivially test it here, though if it goes wrong it will get the same wrong one it did before [18:53] ok. I'll try it. I guess I could also update the kernel to 34 and see if it works with the hardware. I believe I tested 36 but not 34; I don'[t remember. But 34.36 is also confusing [18:54] What's the difference between the 2 commands [18:54] apt-get source linux-image-$(uname -r) [18:54] apt-get source linux-image-$(uname -r)/$(uname -r) [18:55] in theory i am requesting a specific version, no idea if it works, it would for a binary [18:56] apw: Reading package lists... Done [18:56] Picking 'linux-hwe-edge' as source package instead of 'linux-image-4.8.0-28-generic' [18:56] E: Unable to find a source package for linux-hwe-edge [18:57] bah [18:57] What about 34.36 What does that mean exactly. If I look at update manager I see 2 different ones: 34 and 36 [18:58] 4.8.0-34.36~16.04.1 (dsc) [18:59] that is the what we call the ABI number [19:00] apw: If I click on changelogs in update manager I see both 34 and 36 described as _4.8.0-36.36~16.04.1/changelog was not found on this server. [19:00] apw: Actually no. 34 is 34.36 [19:01] I could try updating the kernel to 34 (34.36); If it works with the sound hardware I could conceivably recompile that kernel. [19:02] apw: Why doesn't the command download what I am running and downloads a later kernel with that assumption. [19:03] it defines an upload variant within the 4.8.0 series and increments over time [19:03] green_, it seems to assume you would only want to edit the latest code and move forward [19:04] green_, how about this "dget -d https://launchpad.net/ubuntu/+archive/primary/+files/linux-hwe-edge_4.8.0-28.30~16.04.1.dsc" [19:05] green_, i have gone to the source publication history for that package and found the specific .dsc for that version === JanC is now known as Guest50386 === JanC_ is now known as JanC [20:43] is it possible that a Core2 Q8400 would be better served by not using zram swap? This machine is generally swapped to death everytime I go away for a while and come back [20:52] i cant really imagine a scenario where zram wouldnt be a good idea (unless you only have 128M on a 200MHz single core or so) [20:52] but easy to test i guess ... just remove the zram stuff and see