/srv/irclogs.ubuntu.com/2012/08/29/#ubuntu-kernel.txt

=== chrisccoulson_ is now known as chrisccoulson
* henrix will be out for a bit14:09
ogra_rtg, i added some info to bug 1042918 ... (now dont ask me where to get wl1271-nvs.bin)14:54
ubot2`Launchpad bug 1042918 in linux-firmware "no wireless networks detected during install of Quantal on Panda board" [Undecided,In progress] https://launchpad.net/bugs/104291814:54
rtgogra_, just about to upload linux-firmware with a fix14:54
ogra_oh, awesome, i wasnt sure we even have that file available somewhere 14:55
ogra_(i dont think it was needed in precise)14:55
rtgogra_, I've been cleaning out unnecessary firmware files and removed that one erroneously because the kernel neglected to declare is usage.14:56
rtgits usage*14:56
apwrtg, that kernel is a lazy sonbitch15:13
* apw may have a theory on davemor2's bug ... am spinning a kernel for him now15:13
henrixapw: not sure if you read my comment (and davemor2's reply) in bug report15:16
henrixapw: it looks like there's a workaround for it15:17
apwhenrix, yes, as a result of looking at what the workaround does15:17
henrixapw: ah, ok15:17
apwhenrix, i think when we drop the packet we don't reset it right and can get into a loop ... forever15:17
henrixapw: could it be related with 8a3f0ede2b3f5477122060af1a816c6bbf09fcd2 ?15:17
apwhenrix, that'll teach me not to update my linus tree before debugging ... that is exactly what my patch does15:19
henrixapw: heh, cool :)15:19
rtghenrix, should we cherry-pick taht one ?15:19
apwdavmor2, so ... if yo uhaven't started testing let me cherrypick this offical patch15:19
henrixrtg: i guess it's better to test it before cherry-pick15:20
* apw is engaged with dave so i'll get the testing confirmed15:20
rtghenrix, frankly, I don't it'll hurt anything15:20
rtgapw, ack15:20
davmor2apw: me stops in his tracks15:21
apwdavmor2, be about 5 mins to get teh updated kernels ...15:22
=== yofel_ is now known as yofel
davmor2apw: right I'm back to a broken system15:40
apwdavmor2, ok the newer dated test kernels in the same place as before15:41
rtgapw, I think you ought to apply them all: git log --reverse -p v3.5..HEAD -- drivers/media/video/uvc15:50
rtgthey are all worthy of stable updates15:50
davmor2apw: sorry mainline or your fixed branch?  /me just remembered that you gave me 2 links15:51
apwdavmor2, on my people.canonical.com15:51
davmor2apw: cool thanks15:51
davmor2apw: the one ending 1621 is install and cheese is running smoothly15:58
apwdavmor2, nice i'll get it applied15:59
davmor2apw: that is amd64 that I tested15:59
apwdavmor2, thats fine, its a clear generic issue any testing is gooed enough for me16:00
apwdavmor2, applied, should be in the next upload16:00
davmor2apw: nice :)  I'll try a fresh install once I know it's in the iso and ensure that the desktop installer continues past the previous stuck bit16:01
apwdavmor2, i suspect it won't make beta unless the stars align, i think we already uploaded for that16:02
rtgapw, I was just think I'm gonna upload this fix since it appears to affect a lot of folks.16:02
apwrtg .... ok16:02
apwrtg, if you are let me know so i can hold lowlatency for it16:03
rtgapw, will do. gimme a bit16:03
davmor2apw: if you let the release team know I'm sure it can get landed and once it is get them to give me a ping, I'm on 60meg fiber with zsync at the ready :)16:03
apwdavmor2, heh ... ok16:04
davmor2now back to breaking developer portal16:05
rtgapw, pushed master and master-next16:28
rtgquick build tests, then uploading16:28
apwrtg, yekk w16:30
apwe16:30
apwrtg, rebasn16:30
apwrtg, rebasing ... (apparently after 20 beers)16:30
rtgapw, I was gonna say, something wrong with your keyboard ?16:30
apwrtg, let me know when your build tests are done and it is in the pipe16:30
dileksrebeering16:31
apwi wish16:31
rtgapw, uploaded after test building i386, amd64, and armhf. I'm reasonably sure the rest will build OK.16:43
apwrtg, ok lowlatency done17:11
* rtg -> lunch17:27
* cking -> EOD17:36
=== bryceh is now known as bryce
rtgapw, given the email problems I saw you mention to IS, I'm wondering if you missed this one: https://lists.ubuntu.com/archives/kernel-team/2012-August/021801.html18:33
apwrtg, update-initramfs detects the kernel and does _not_ set the deferred triggers, it then lets the /etc/kernel.d/*thingy* hooks run things18:43
apw:18:43
=== henrix_ is now known as henrix
rtgapw, why does this make me feel so dense ? don't we want it just freaking create the initrd when called ?18:44
apwrtg its all about cost.  you need to update the initramfs on kernel install, but also if you change initramfs-tools, /bin/sh, upstart, fsck, etc etc18:45
apwdo all of those say "update-initramfs" and that triggers a delayed rebuild at the end of everything18:46
apwthat delayed rebuild can only be for the current kernel18:46
apwfor a new kernel we have to do it there and then and the hooks handle it18:46
rtgapw, its definitely triggering a delayed build (or rebuild), which is why I can't figure out how our postinst works.18:47
apwrtg, on omap4 or on x86, its definatly not delayed there18:47
apw(on x86)18:47
rtgapw, it says it is on x8618:47
apwupdate-initramfs: deferring update (hook will be called later)18:47
rtgtahts the line ^^18:48
apwthat message does not mean it is delayed18:48
apwthat means that its _not_ delayed18:48
apwits just being done at the end of the kernel package install18:48
apwupdate-initramfs: deferring update (trigger activated)18:48
apwthat one means its being proper delayed to the end of the apt-get18:48
apw:q18:51
apwogra_, on your panda with kernel install issues what is in your /etc/kernel-img.conf18:58
apwogra_, as in the past that file has gotten lost and that makes kernel installs go to put19:01
apwogra_, poke me in the morning and we can discuss19:02
* apw wanders to get provisions19:15
=== henrix is now known as henrix_
=== henrix_ is now known as henrix
* rtg -> EOD20:24
bjfjjohansen: next week we are building the last natty kernel \o/ if you have anything that you just have to have in it CVE wise, you better speak up quickly20:41
jjohansenbjf: okay thanks for the heads up20:42
bjfjjohansen: i didn't want to give you too much of a heads up because you might have found something if i did :-)20:44
jjohansenhehe20:44
=== henrix is now known as henrix_

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