/srv/irclogs.ubuntu.com/2017/04/06/#ubuntu-kernel.txt

=== JanC is now known as Guest52949
=== JanC_ is now known as JanC
=== ivan_ is now known as ivan
ivanwhy does https://insights.ubuntu.com/2017/04/05/ubuntu-on-aws-gets-serious-performance-boost-with-aws-tuned-kernel/ mention "Disabled CONFIG_NO_HZ_FULL to eliminate deadlocks on some instance types" when it's already N / unset in ubuntu-xenial.git:master?05:11
mvohey sforshee - apw send me your way. I was looking at the testsuite failure of snapd with the latest zesty kernel from proposed. we have some tests that excercise the seccomp filtering. it looks like with the kernel from -proposed we don't get messages like http://paste.ubuntu.com/24327651/ anymore in dmesg which makes the tests fail13:36
mvosforshee: let me know if I can do anything to help with that issue13:37
sforsheemvo: yeah I've been looking, unfortunately the snapd tests have been broken for so long I can't tell when this first started happening :-(13:37
sforsheemvo: do you know if this happens with the kernel in -release?13:38
sforsheemvo: and what's the easiest way to reproduce the problem?13:38
mvosforshee: its fine with the kernel in release13:41
mvosforshee: its broken with the kernel in -proposed13:41
mvosforshee: easiest (for me, ymmv) is git clone https://github.com/snapcore/snapd ; cd snapd/cmd; ./autogen.sh; make check13:44
mvosforshee: let me know if that does not work for you, you may need to do "sudo apt build-dep snapd" first13:45
mvosforshee: but the C parts are light on dependencies so it may work without13:45
mvosforshee: there are three tests failing with the new kernel, they all do dmesg|tail -n1|grep "audit"13:45
sforsheemvo: thanks, I'll see what I can find13:47
jdstrandmvo: curious if ratelimiting is getting in the way13:47
jdstrandmvo: note my (now ancient) mknod PR had a bunch of testsuite improvements in this area13:47
jdstrandmvo: are we at a point where we can consider that PR (and the other seccomp arg filtering PRs) again?13:48
jdstrandmvo: afaik, we still need this: https://github.com/snapcore/snapd/pull/281013:48
mvojdstrand: re rate-limiting> I think not, I don't see any audit message at all in dmesg13:48
mvojdstrand: the mknod is close, we still need #3027 first but that one is close, it looks there there is agreement13:49
jdstrandmvo: well, that could be a sympton. the kernel is funny like that sometimes13:49
jdstrandmvo: ok, cool. I'll keep an eye on 3027. will that be the pr instead of 2810?13:50
mvojdstrand: yes, that is the new direction(the second new direction actually)13:51
mvojdstrand: but close finally :)13:51
jdstrandmvo: great to hear :) thanks! :)13:52
=== Trevinho_ is now known as Trevinho
ad-n770do you have any estimated date on when the 4.10 kernel will land into Ubuntu 16.04 repositories as linux-image-generic-hwe-16.04-edge ?16:00
infinityad-n770: Two days ago.16:08
ad-n770oh, it's there16:11
ad-n770going to try it16:11
ad-n770we had participated in the tracking and fixing of an issue affecting drm/radeon that causes a system lockup under certain loads16:58
ad-n770the fix is at https://cgit.freedesktop.org/~agd5f/linux/commit/?h=amd-staging-4.9&id=03df14330dd6a6cfa91964131e001c1ca58a380f16:58
ad-n770what should be the procedure to request it to be backported into Ubuntu 16.04 4.4 kernels ?16:59
tjaaltonclone ubuntu-xenial.git, backport the patch, send the patch to the list. and file a bug for it, reference it in the patch17:01
tjaaltonor, ask mrcooper to send it to stable@vger17:02
tjaaltontakes a bit longer to reach 4.4 though17:02
ad-n770thanks17:35
ad-n770tjaalton: could please be more specific in which list we should send the patch ?17:37
tjaaltonkernel-team@17:39
tjaaltonhttps://wiki.ubuntu.com/Kernel/Dev/KernelPatches17:41
ad-n770thanks19:14
=== Serge is now known as hallyn

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