=== asac_ is now known as asac === mkrufky is now known as mkrufky-gone [05:53] hi ScottK-laptop. okay to start out you'll need to clone the intrepid git tree [05:53] you can visit http://kernel.ubuntu.com/git for a listing of all the git trees available [05:53] more or less git-clone git://kernel.ubuntu.com/ubuntu/ubuntu-intrepid.git [05:53] Man, now you want me to use git too. [05:53] Doing [05:54] hehe [05:54] i was very resistant of using git myself too at first, but i've come to be at peace with it [05:54] so while that's cloning, click on the ubuntu/ubuntu-intrepid.git link [05:54] I need to learn it. [05:54] can try to explain how the tree is organized [05:55] OK [05:55] so if you click the "tree" link, you'll see an ubuntu/ subdirectory [05:55] all the out of mainline kernel modules are kept there [05:55] inside there there is a Kconfig for everything that is included in there, it ends up sourcing the Kconfig for subdirectories and more modules [05:56] so you would likely make subdirectory for your module, and then end up modifying Kconfig to source the Kconfig of your module [05:56] also, there is a Makefile that drops it down into the subdirectories as well when they are built [05:56] you'll have to add in a line for that too [05:56] Their patch includes kconfig changes [05:57] well it include a Kconfig for the module itself [05:57] the top level Kconfig is just going to call the Kconfig from this module [05:57] Yes, and they have the b/security/Kconfig change to add themselves. [05:58] well they are assuming they end up "in tree" [05:58] based through the security directory [05:58] instead you are in the ubuntu directory [05:58] so you'll be patching the Kconfig in the ubuntu directory instead [05:58] Right. [05:58] So we'd need to modify their patch a little. [05:58] yeah [05:59] just where you are sourcing and what not [05:59] once you've got all those files in place, you need to modify the kernel configs [05:59] in debian/config/(arch)/config [05:59] to add in that extra Kconfig option that they are referring to [05:59] CONFIG_SECURITY_DAZUKO [06:00] the syntax should hopefully be pretty apparent in there [06:00] hopefully. It's 1AM here, so nothing is really apparent at the moment. [06:00] well tomorrow then :) [06:01] let me give you the last two steps and you cna let it sink in and ask questions/what not [06:01] https://help.ubuntu.com/community/Kernel/Compile#Build%20the%20kernel%20(when%20source%20is%20from%20git%20repository,%20or%20from%20apt-get%20source) [06:01] that shows you how to build a kernel from the git tree [06:01] you dont want to build all the different variants, just the generic for testing [06:01] so CONCURRENCY_LEVEL=2 AUTOBUILD=1 NOEXTRAS=1 fakeroot debian/rules binary-generic is probably what you want to do [06:02] when you're sure that everything is functional, you can make a git patch out of everything and submit it to the mailing list or publish a git tree. ping me or someone in here when you are ready to do that so that you can get your git config all ready with commit name and how to use the templates [06:03] OK. [06:03] I'll probably be asking for help several times again between now and then. [06:04] yeah the first time is pretty confusing when you dont understand syntax and all, but hopefully it should end up more apparent [06:05] So I'm reading some more and this is getting a bit scary. [06:06] http://dazuko.org/tgen.shtml#LSM indicates the whole thing may not work so well with Apparmor. [06:06] yuck :) [06:08] I'm guessing we don't have an RSBAC enbaled kernel http://www.rsbac.org/ and aren't likely to get one. [06:11] Which would leave syscall hooking and I'm guessing that won't be popular. [06:11] especially since they dont have snapshots for 2.6.26 [06:15] Yep. [06:15] http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=423281 is also relevant. [06:17] I think I'm going to go to bed and then ponder tomorrow if I really want to open this can of worms or not. Here I was just hoping I could wave a little DKMS magic over it and it would be easy. [06:17] superm1: Thanks for the help. [06:17] ScottK-laptop, sorry if it won't work here, no DKMS magic will help it either :) [06:18] best of luck sorting out though [06:19] Thanks. === emgent is now known as emgent` === emgent` is now known as emgent === mkrufky-gone is now known as mkrufky [17:19] alex_joni: hi, i tried also all_generic_ide + irqpoll, but.. no way to make it working :/ [17:20] seems that ubuntu-server 7.10 and all ubuntu 8.04 does not support all_generic_ide somehow :( hmm so bad :( [17:20] *do not [17:21] all_generic_ide is a very needed parameter for OLD PCs to make them working [17:21] i've personally used all_generic_ide in 8.04 without trouble [17:22] superm1 it worked for me only with xubuntu 7.10, how is it possible? [17:23] (and ubuntu 7.10 too, if i dont mistake. but any server version did) [17:23] *did not [17:23] huh? [17:23] just tried with ubuntu-server hardy [17:23] does not work, i get several I/O errors about my hdd [17:23] well are you sure that's really the root cause of your troubles? [17:24] are you thrown to a functional initramfs prompt? You can check what modules ended up being used [17:24] whether it's ata_piix or what not [17:25] superm1 sorry i forgot to mention that: i refer to livecds :/ sorry, i totally forgot to say that [17:25] yeah in grub it works [17:26] soreiser, well in both cases the kernel parameters are passed the same. you are adding it to the live cd prompt then? [17:26] yeah superm1 [17:27] something like: "..... all_generic_ide -- " (also tried without the '--'" [17:27] soreiser, well that's awfully odd then if its not being passed on properly only from a live cd [17:28] so superm1 what could i do? is it going to be fixed for the next release? [17:28] soreiser, i've used all_generic_ide in both a live cd and a scripted install [17:28] so i'm not sure where your issue is [17:29] superm1 we could talk in italian if is it possible, as i've seen your italian in your personal info [17:29] *yuo're [17:29] **you're [17:30] soreiser, sorry i don't speak italian [17:30] superm1 ah sorry i thought you were :) your is an italian name eheh :) [17:30] i am, but i dont speak it unfortunately [17:31] soreiser, i'd recommend trying to get a bug report together with all of this data you've gathered to get some more eyes on the problem [17:31] superm1 well, however, simply that parameter does not work me with livecds (except xubuntu/ubuntu 7.10) [17:32] superm1 yeah i've already made a bug report, even if were not so sure it is a bug [17:32] soreiser, hopefully something will stick out [17:32] *work for me [17:32] soreiser, sorry i can't be of much more help :) [17:32] well ok superm1 thanks anyway for your support, good ;) === alex_joni_ is now known as alex_joni [17:34] superm1 should i maybe (re :))turn to #ubuntu-dev to report it is not a bug issue then? [17:34] soreiser, well there very well can still be a bug here, your hardware is different than mine, perhaps it's just been a fluke that it's worked in the past for you [17:34] soreiser, i say let the bug triage folks handle it on the report [17:35] that's the best workflow to record all of the empirical data [17:39] well ok superm1 i'll try to get some workaround from the discussion in the bug report, hoping it will be FIXED soon ;) [19:08] Could a member of bug-control please check on bug #258754? i think it could be set to "triaged". [20:05] hm, I shall try again: does anyone know why CONFIG_MD_RAID10 is not selected in hardy's kernel image? [21:37] <_ruben> chris_goe: because it is? [21:37] <_ruben> $ grep RAID10 /boot/config-2.6.24-19-server [21:37] <_ruben> CONFIG_MD_RAID10=m