/srv/irclogs.ubuntu.com/2015/09/15/#ubuntu-kernel.txt

mIKEjONE1hey guys01:09
mIKEjONE1is there a way to build the kernel without having to build the whole thing from scratch everytime I make a change?01:09
mIKEjONE1I'm using fakeroot with debian/rules01:09
mIKEjONE1I'd like to have something that's able to only compile things that changed (i.e. the way Linux's Makefile works)01:10
mIKEjONE1it's getting kind of ridiculous, so any help would be greatly appreciated01:32
apwmIKEjONE1, yes after build if you want to change a few files and rebuild you can remove debian/stamp/stamp-build-<flavour> and then re-run fakeroot debian/rules binary-generic08:11
rtghenrix, kamal: I just sent a patch to the k-team list in reference to https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1397976/comments/4 - that commit looks good for stable.13:49
ubot5Ubuntu bug 1397976 in linux-lts-trusty (Ubuntu Trusty) "tty hangup regression in 3.13 kernel (trusty LTS)" [Undecided,In progress]13:49
henrixrtg: yep, looks like it could be applied to 3.13 kernel (all others seem to have it already)13:52
rtghenrix, clean cherry-pick13:52
henrixrtg: cool, thanks.  kamal, i guess you can pick it ;-)13:53
rtghenrix, it might be a bit early for California slackers :)13:54
henrixheh, indeed13:54
apwhenrix, so did we have that in P and fail to apply it to T somehow ?13:54
henrixapw: no, i we don't have it P but i believe it is not applicable to such an old kernel...13:56
* henrix goes check13:56
apwhenrix, ahh ok, it was the regression word that confused me13:56
henrixapw: right, so the issue was introduced by commit eafbe67f8476 ("n_tty: Refactor input_available_p() by call site"), which was cherry-picked into trusty.  so it *may* not even be applicable to stable 3.13, depending on whether or not it contains this commit13:58
henrixkamal: ^13:59
jsalisbury**15:03
jsalisbury** Ubuntu Kernel Team Meeting - Today @ 17:00 UTC - #ubuntu-meeting15:03
jsalisbury**15:03
kamalrtg, apw, henrix: fyi, 3.13-stable does not contain eafbe67 "n_tty: Refactor input_available_p() by call site", so I'll take no action on that topic, barring further discussion.16:22
henrixkamal: ack16:27
jsalisbury##16:55
jsalisbury## Kernel team meeting in 5 minutes16:55
jsalisbury##16:55
bkeroCan I ask that someone install libssl-dev to the kernel build hosts so that we can actually generate kernel packages? http://kernel.ubuntu.com/~kernel-ppa/mainline/v4.3-rc1-unstable/16:56
bkeroIs that off-topic for the meeting?16:57
hallynwhat is all this:20:24
hallynroot     22390     2  0 15:23 ?        00:00:00 [kworker/u16:88]20:24
bkeroa kernel worker process20:26
hallynwhy so many of them?20:28
hallyn(this is appears to be new)20:28
hallynoh, actually they've died down20:29
hallynperhaps a side-effect of resume.  interesting20:29
bkeroIt's a kernel thing20:29
bkerocheck out the kernel docs20:29
rharperis there any place to get previously published but superceeded kernels?  I've got an issue on the latest vivid kernel but works fine on the vivid released kernel;  would like to bisect but without building all of those kernels 20:36
sforsheehallyn: entirely possible it was a result of resume, if that generated a lot of workqueue items. I think the kernel scales the number of worker threads.20:37
sforsheerharper: aren't they all still in the archive?20:39
rharperdid a apt-cache policy linux-image-generic and only showed the latest; so I probably need some apt-fu to find it20:40
sforsheelinux-image-generic is a meta package. Every new kernel version is its own package so that multiple kernels can be installed at the same time.20:43
rharperah, ok20:44
rharpercool; I think I can get them now20:44
rharpersforshee: thanks20:44
sforsheenp20:44

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!