[00:02] sconklin, I can add my system details to the bug. [00:02] sconklin, if needed [00:03] jsalisbury: I don't think we need them but thanks [00:04] sconklin, ok, thanks === sconklin is now known as sconklin-gone === lucent_ is now known as lucent [07:56] is this the right channel to ask about DMA? =) [08:30] * apw yawns [09:09] * _ruben agrees with apw === smb` is now known as smb === ogra_ is now known as ogra === diwic is now known as diwic_afk [14:04] how to backport patches to my kernel? [14:08] do i need to recompile the kernel if I backport some patches? [14:09] apw, linux-backports-modules-wireless-2.6.35-lucid-generic [14:16] fairuz, after applying them to the source, normally yes you need to recompile [14:17] apw: so I just have to apply the patch normally, and recompile the kernel? [14:18] fairuz, thats what i do === sconklin-gone is now known as sconklin [14:18] apw: thanks. I give it a try [14:23] sorry, one more easy question, so to backport what exactly i should do? my idea is to modify the file in question and add the differences? [14:24] (I'm new to the patch and git thing) [14:25] i presume that patch is not able to apply the change you have [14:29] hmm I confuse between a backport and an upgrade === herton is now known as herton_lunch === cking_ is now known as cking [14:42] diwic, something has come up for me on my multitouch stuff [14:42] I don't think I'll be able to help out with the firmware packaging :( [14:44] cnd, today, or ever? [14:44] diwic, well, for a few days at least [14:44] at least [14:44] someone is leaving our team [14:44] so a pile of tasks got shifted on to everyone [14:45] cnd, ok :-( I'll remind you in a week or two then [14:45] and those tasks are a little late already :) [14:45] diwic, you can prepare the upload and tim can push it to the archive [14:56] apw, bug 662288 was brought to my attention [14:56] Launchpad bug 662288 in linux "rt3090: freeze on module rt2800pci unload" [Undecided,Confirmed] https://launchpad.net/bugs/662288 [14:56] there's a fix in linux-next, but it won't make .38 [14:56] I was asked if we could pull it into natty [14:57] http://git.kernel.org/?p=linux/kernel/git/next/linux-next.git;a=commit;h=7f6e144fb99a4a70d3c5ad5f074204c5b89a6f65 [14:57] sforshee, as its for unload its not so important i assume, but sure if it fixes bugs we can take things like that [14:57] apw, I think it causes a hang when shutting down [14:58] sforshee, ok, thats deffo worth a poke. i'll look it over and see shortly [14:58] thanks apw [15:20] sconklin, would you check the moderation queue for ubuntu-kernel-sru@lists.ubuntu.com ? I've sent 2 emails that have not appeared. [15:20] tgardner: not seeing any pending requests [15:21] sconklin, hmm [15:21] This has been happening because I subscribed people and sometimes they send from their alias. As I see them I moderate and whitelist the new address. But I'm not seeing yorus [15:24] sconklin, my mail server says they are going out: Mar 2 19:06:16 mail postfix/smtp[20232]: 6FFCF2A4074: to=, relay=lists.ubuntu.com[91.189.94.204], delay=4, status=sent (250 OK id=1PuysF-0005fa-LX) [15:27] tgardner: don't know what to say - I've gotten moderation requests for two other people on the list and fixed them. Haven't seen yours. I'll have to open a ticket [15:28] sconklin, well, I sent my email to the k-t list with a Cc to the sru list [15:28] (just now) [15:32] tgardner: Re: [kernel-SRU] linux-lts-backport-maverick archs? ??? [15:33] I think what happened is that there's an option turned on in the mailing list to eliminate duplicate emails to members, and it's probably only sending to one of the lists for you. I'll turn that off, as it's counterintuitive [15:33] done === herton_lunch is now known as herton [15:46] brb, need to make some more coffee [15:46] * smb orders one [15:51] * JFo fedexes :) [15:52] JFo, Many thanks. :) [15:52] my pleasure [16:10] ooh I hate computers right now... they keep doing what I tell them to. [16:11] they are supposed to know what I want :) [16:11] * JFo keeps closing pages before he is done with them === mdeslaur_ is now known as mdeslaur === alex_jon1 is now known as alex_joni === jussi01_ is now known as jussi [17:31] <-lunch bbiab === yofel_ is now known as yofel === sforshee is now known as sforshee-lunch [18:06] hi.. i want to begin fixing bugs for kernels. can somebody suggest me easy bugs for the start?.... :) [18:15] JFo: hi.. i want to begin fixing bugs for kernels. can somebody suggest me easy bugs for the start?.... :) [18:16] hi.. i want to begin fixing bugs for kernels. can somebody suggest me easy bugs for the start?.... :) === JayFo is now known as JFo [18:18] aakshay___: seems to be issues with freenode right now, give it a few minutes [18:18] charlie-tca: okiez.. :) === tg is now known as Guest66130 [18:42] * tgardner --> lunch === cmagina is now known as cmagina-lunch [18:55] well, that was interesting [18:59] is it done yet? === sforshee-lunch is now known as sforshee [19:01] charlie-tca, I think that last one was my provider [19:01] my entire connection dropped [19:01] Mine dropped twice during all those spits [19:02] hmmm [19:02] I think we are as good as we will get for today [19:02] aakshay___, were you looking for me earlier? [19:03] * JFo reads back [19:03] ah, interested in looking at kernel bugs :) [19:03] excellent! [19:03] * JFo digs up some links [19:03] JFo: hi JFo ..... yes i am intrested in fixing kernel bugs [19:03] but dont know where to start with.. :p [19:04] aakshay___, our kernel wiki pages are helpful [19:04] https://wiki.ubuntu.com/Kernel/BugTriage [19:04] is the first place you should go [19:04] from the https://wiki.ubuntu.com/Kernel page there are a number of other pages of interest [19:04] JFo: i went through all the bug triaging process.. [19:04] ah, great [19:05] but i not read debugging yet.. do i need to or can i start? [19:05] it would probably be good to read, but you can start without it for now [19:06] yiipii.... :).. so an easy bug for me? [19:06] that is a good question. Unfortunately, I don't have any easy bugs. All of them require some basic triage to a certain extent [19:07] then there is a more advanced triage that I do [19:07] my best advice to you is to go through the bugs and see which ones you are comfortable looking at [19:07] I am here if you have questions :) [19:08] okiez.. can you please tell me exact bugs page link.. [19:08] sure [19:08] * JFo grabs it [19:08] :) [19:08] this is the entire list of linux package bugs: https://bugs.edge.launchpad.net/ubuntu/+source/linux/+bugs [19:08] thanks..:) [19:08] my pleasure [19:09] let me check if any suits me.. hehe... :) [19:09] sounds good, ping me here if you need any advice or help :) [19:09] sure....thanks... :).. === bjf__ is now known as bjf === bjf is now known as bjf[afk] === cmagina-lunch is now known as cmagina [19:38] JFo: i found this bug "289584: networking disabled after resume from sleep" ... how will it be to start working with it? [19:41] JFo: i need your help.. i dont know how to start with it.. where to get the modules? [19:42] ok, one sec.. [19:42] yes... [19:42] I would start with basic triage, requesting what basic logging we require === _LibertyZero is now known as LibertyZero [19:46] okiez.. so do i need to check it in upstream mainline? [19:47] I'd ask them to check, yes [19:49] JFo: okiez... but sorry to ask can you please give me further directions.. :) [19:50] ok [19:51] what kind of advice do you need? :) [19:51] so on this bug... [19:51] JFo: yes [19:51] I would ask for apport-collect data [19:52] JFo: that as i understand , i need to take the driver files and check them or so? [19:52] so the requests for information text is here: https://wiki.ubuntu.com/Kernel/BugTriage/Responses [19:53] ok.. let me read it [19:53] :) [19:53] ok :) [19:55] sorry about that, was a bit distracted by another conversation going on at the same time. [19:55] no issues... :D [19:57] :) [20:08] JFo: ya.. downloading the image.. and it seems to take long time. so i will come back to you after completing with basic triaging.. :) [20:08] not sure what you mean by downloading the image aakshay___ [20:10] maybe I am misunderstanding what it is you are wanting to do [20:10] JFo: its mention to check with the latest development release [20:10] right, we ask the reporter to check, as they are the ones encountering the issue [20:10] and the URL to ISO CD images is given [20:10] right, we provide that to them so they know where to look for it [20:10] so i moved to it and it ask to download one as per the system arch [20:11] yes.. so i am downloading an image for X86... :) [20:11] but why are you not asking the reporter of the bug to do that? [20:11] I guess that is where you are losing me :) [20:12] JFo: :D.. we can ask the reporter !!!.. okiez... [20:12] bjf[afk] I have a lucid stable branch almost ready. I wanted to make sure you weren't working on it before I create the tracking bug. [20:12] yep, that is preferred as we want to find out if they still have the issue they reported in the places we point them to [20:12] that helps the team determine if a fix exists [20:13] it also allows us to find out the differences between the two [20:13] so we can identify the problem further [20:13] so i will contact him and ask to run the same to check.. [20:13] sounds good :) [20:13] then i need to add comment there?? [20:14] yes, you would take one of the responses I pointed you to and see which one fits the best [20:14] and then provide that to the reporter [20:15] let me see which one can be best [20:15] ok [20:16] JFo: i think this is best "Old With No Activity" [20:16] yep :) [20:16] and then when i will add in comments there.. :) [20:16] that sounds fine :) [20:17] JFo: sorry to say but i think reporter will take lot of time to respond?? .. :( [20:18] perhaps, it is also possible that they will not respond. [20:19] their issue may have been fixed already in a later release [20:19] and they may just not be willing to answer [20:19] once you comment, set the bug to incomplete so that we will have the correct state of that issue [20:19] JFo: ok.... [20:20] JFo: is there any direct way to enter in developing modules for kernel?.. :) [20:20] not sure what you mean [20:21] JFo: i have been reading a lot of stuff on programming linux kernel.. so eager to implement and design something. [20:21] ah [20:22] so you are wondering what the best way to contribute is? [20:22] JFo: like any module of kernel. :).. please suggest something [20:22] yes.. in development of it [20:22] on that I am afraid I would be no help. I am the bug triager :) [20:22] development would be something one of the engineers would be better advising on [20:23] :)... no problem.. so whom can i ask here or elsewhere? [20:23] well... [20:23] let me see. [20:23] please.. :) [20:24] apw, you forgot to mark yourself away. :) [20:24] sconklin, got a minute? [20:24] or are you in the middle of something? [20:24] for you, sure ;) [20:24] heh, thanks [20:24] sconklin, aakshay___ was wondering where to begin on kernel module development [20:25] reading back [20:25] as I am hopelessly lost in bugs, I wondered if you could point in the right way [20:25] thanks [20:26] there was an excellent book out a few years ago, let me see if it's up to date with the latest kernels. One sec [20:26] sconklin: hi.. i am looking for programming the kernel.. can you please suggest me [20:26] sconklin: please :) === zul_ is now known as zul [20:27] aakshay___: If you want to know specifically about how to write drivers, this book is probably the best resource: http://lwn.net/Kernel/LDD3/ [20:27] drivers are equivalent to modules, essentially [20:28] sconklin: thanks.. :) and i have completed half of it.. [20:28] and doing the "Programming with Linux" as well [20:28] so eager to implement now [20:28] :) [20:29] "LDD3 is current as of the 2.6.10 kernel" that's a bit old [20:29] The examples in that are going to give you a different method for building the kernel than the way we typically build Ubuntu kernels, so the build steps may not be what you see on our wiki for how to build kernels [20:29] one good book is also "Writing Linux Device Drivers" (Jerry Cooperstein) [20:29] Krunch: yes.. :) [20:30] yes, but I don't think that the major APIs have changed. Some things like debugfs have changed tho. I wish it were more up to date [20:30] well, it can certainly get you started [20:30] the more I think about it the more I think of that changed [20:30] tgardner: say, are there any sandybridge systems we can do builds on? [20:30] but double check all the details in the source (that applies even with the latest book really) [20:31] oh yeah, the kernel is a constantly moving target [20:31] kees, none of the kernel buildd's are SBs. you might ask VanHoof [20:31] aakshay___: are you interested in developing a new driver, or working on an existing one? [20:31] it is a main reason all of the bugs require constant triage [20:31] tgardner: okay, cool. I was curious about some benchmarks. [20:31] kees, they are all laptops AFAIK [20:31] as i am a begineer.. dont have much idea.. you suggest me.. :) [20:32] vanhoof: if you get some sandybridge systems that I can get access to for some builds, I'd be curious. not a priority at all, just idle curiosity. :) [20:32] tgardner: yeah. just a random bit of curiosity. thanks! [20:32] sconklin: as i am a begineer.. dont have much idea.. you suggest me.. :) [20:32] kees: I'm sure we can sort something out :) === popey_ is now known as popey [20:32] kees: after a desktop I assume? [20:33] kees: the bulk of what we have out there now are mobile, but there are a few desktops scattered around the team [20:33] vanhoof: no real preference, just curious about some the CPU extensions [20:33] *some of the... [20:33] all of them are presently in flux with a pretty awesome S4 issue :) [20:33] vanhoof, you forgot the tag [20:33] :) [20:34] vanhoof: cool, yeah, if one goes idle, keep me in mind. consider me at the end of the line after any real work. :) [20:34] JFo: I sensed that tag :) [20:34] aakshay___: well, the two things that are important are to pick something you have hardware for, and to pick something that interests you. The way that a lot of kernel driver development is done is by sub-groups who work on video4linux, or audio, or any number of other things. === bjf[afk] is now known as bjf [20:35] aakshay___: But to just experiment, if you can get to the point of building your own kernels, and loading and using a very simple 'hello world' driver, then you will have learned a lot [20:35] kees, me too, but I was worried for the others ;) [20:35] hehe [20:36] kees: bjf /may/ have a SNB desktop in hand, I want to say he was doing some builds there [20:36] but i can see if i can sort one out for you once this s4 bug unwinds a bit [20:36] we're nearly there [20:36] sconklin: so i start with the "hello world" driver...?? [20:37] vanhoof: cool; thanks [20:37] aakshay___: if there is one called that. I was using it as a dort of generic name. I can't find my copy of the book, but I think it has an example driver that just loads and unloads and is a basic shell [20:38] kees, i have a sugar-bay SDP here at home, bryce has a variant as well (i delivered it to him) [20:38] aakshay___: but the very first things to do are to use git to fetch the kernel source and get it to build and make sure it runs on your machine. You should do that before you start writing any code at all [20:39] aakshay___: have you ever built a kernel? [20:39] sconklin: no.. :( [20:39] ok, hold on while I find our wiki page about how to build the Ubuntu kernel. You should get that working before you do anything else [20:40] sconklin: its my first time to be here on this channel.. [20:40] ok [20:40] :) [20:41] aakshay___: https://wiki.ubuntu.com/Kernel/BuildYourOwnKernel [20:42] sconklin: for this, first i will download the kernel from "git" [20:42] then build as per instruions in wiki? [20:42] aakshay___: have you read much C so far? [20:42] aakshay___: git is a user application tool used for source code management. You will use it to fetch a copy of the source code [20:43] Krunch: yes.. i have been keep reading more and more of it [20:43] :) [20:43] sconklin: ok.. i will do this. [20:44] a good starting point is to build your own kernel, look at the CONFIG_* options and go see in the code how they are implemented [20:44] and will finish with the LDD3.. [20:44] bjf: ah, cool; I'll check with bryce and see if I can snag it for an afternoon [20:45] that should give you a good general idea of how the kernel works and what the code looks like compared to userland applications [20:45] but then, if you only want to write drivers that's not strictly necessary [20:45] i have a confusion. here "ownKernel" is the one from Git? :) [20:46] aakshay___: own kernel means that you have compiled it yourself from source code instead of installed one that we built for you [20:46] build a kernel, from git or from ubuntu, doesn't matter much for a start [20:47] ok.. so i will first build own kernel.. and then come back with problems if any.. :) [20:48] sconklin: Krunch: JFo: thanks a lot for guidance.. [20:49] aakshay___: I hope it helps ;-) [20:49] thanks.. :) [20:49] and will be back to ask more [20:49] :) [20:50] good bye... :) [20:53] thanks sconklin [21:39] JFo, amber had an easy way to find all the wiki pages under a certain path, do you remember that ? [21:39] * JFo thinks [21:40] let me see if I have it marked... one sec... [21:41] hmmm, I don't see it [21:41] akgraner, do you happen to have that wiki link for all kernel pages? [21:42] bjf i think that is described under macros in the wiki help which you get to from the bottom of an edit page [21:42] apw, looking [21:42] we use something link it in the faq edit page i think [21:43] <> [21:43] something like that in your page would do it [22:41] * jjohansen -> lunch [23:13] apw, is there some way to know where the "Action" wiki pages have been included so one could look at them in context to see if they make sense ? === sconklin is now known as sconklin-gone