/srv/irclogs.ubuntu.com/2006/08/15/#ubuntu-kernel.txt

TheMusoBenC: Thanks. Seems like someone has already taken your fixes upstream.12:17
BenCTheMuso: which ones?12:17
TheMusoThe module loading, and the semaphore fixes.12:19
TheMusoAs well as the ppp serial header adjustment.12:19
TheMusoppc 12:19
BenCexcellent12:20
BenCthat was quick :)12:20
TheMusoDaniel drake from Gentoo must be following it, as he is the one who committed the fixes.12:22
TheMusoI don't know if you are also aware, but there was a commit to CVS recently that I submitted to add a keyboard command for laptop keyboards. I can send you a patch if you'd like.12:24
BenCyes, please do12:28
TheMusosent. Thanks again.12:34
TheMusoBenC: I would like to build a test kernel from git to test the speakup stuff. Is there a way that I can build only one kernel flavour, i.e just i686?12:42
BenCTheMuso: See the CategoryKernel wiki in the topic, there's a link on that page to a "howto build" wiki page12:42
TheMusoOk thanks.12:43
BenCTheMuso: or you can wait one day, and the daily builds will be up (URL is in the topic)12:43
TheMusoYeah I saw that too. Cheers.12:43
=== pschulz01 [n=paul@eth6067.sa.adsl.internode.on.net] has joined #ubuntu-kernel
=== anibal [i=ams@ns1.mssinc.biz] has joined #ubuntu-kernel
BenCmdz: ping01:58
mdzBenC: pong01:58
BenCmdz: email with benchmarks sent01:58
BenCmdz: basic idea is, there's little to no difference01:59
BenCsame results for amd64 kernels01:59
BenCmdz: the best numbers to look at are the Ratio's to the no_load02:00
BenCthe Time columns is also important02:01
BenCI'm still running a UP amd64-generic vs amd64-k8 test, but I don't expect it to be any different02:02
mdzBenC: that looks pretty compelling to me02:02
BenCmdz: just say the word02:03
mdzBenC: we should discuss on ubuntu-devel; mind if I forward your email?02:03
BenCsure02:03
pschulz01Greetings.. I am going to be 'lurking' here.. but is there anything that I should be aware of (other than what is in the channel subject line?)02:10
pschulz01the plan is to learn and start contributing where I can.. I have various embedded archtectures I can test on (if relevent) and an older (non-intel) Mac Mini.02:13
mdzBenC: do you have any pointers to documentation which explains what the figures mean?02:19
BenCpshel: welcome, nothing really to explain really :)02:19
mdzpschulz01: there's some information at https://wiki.ubuntu.com/KernelTeam, though the time-related stuff is out of date02:19
BenCmdz: contest.kolivas.org02:20
mdzthanks02:21
pschulz01mdz: Thanks.02:24
pschulz01BenC: Is ther much that needs doing?02:24
BenCpshulz: bug maint is the main thing we need, but we only have light documentation on the wiki for kernel bug triaging02:25
BenCsorry, I keep misspelling your name02:25
pschulz01BenC: How many packages does the team/you look after?02:25
BenCwell, kernel is mainly linux-source-2.6.17 (edgy) 2.6.15 (dapper) 2.6.12 (breezy) and 2.6.10 (hoary)02:26
pschulz01BenC: that's OK :-)02:26
BenCwe also maintain linux-restricted-modules and linux-meta02:26
pschulz01Can I put that on the Wikipage :-)02:27
BenCand to some degree, other packages like initramfs-tools, kernel-package and kernel-wedge02:27
mdzBenC: you want to read over this before I send it, or just get it when it hits the list?02:27
BenCmdz: you can just send it...doubt there's much I need to read over02:27
BenCpschulz01: just keep it simple...the less we have to maintain the info the better02:28
pschulz01BenC: I need somewhere to keep notes :-)02:30
BenCkylem_: ping02:40
pschulz01BenC: Have a look at https://wiki.ubuntu.com/KernelTeam02:50
pschulz01BenC: Is there a lauchpad page?02:51
mdzpschulz01: launchpad.net/people/ubuntu-kernel-team I believe02:51
pschulz01BenC: launchpad02:51
BenCmake sure you get the right one02:51
BenCthere are two kernel teams, one has about 6 members, and the other is empty02:51
BenCchoose the one that has members02:51
BenCI'm waiting for the launchpad folks to implement team merging02:52
BenCpschulz01: you can kill the news portion. Looks great though, thanks!02:53
pschulz01BenC: I need an administrator to add me.. 02:53
BenCI'll get the request02:54
pschulz01Done..02:54
BenCmdz: nice email, thanks02:59
kylem_BenC, pong.03:09
pschulz01BenC: The Launchpad team is listed as 'restricted', as opposed to 'moderated'.. so it doesn't look like you'll even get my request to join.03:12
pschulz01BenC: Some final changes to https://wiki.ubuntu.com/KernelTeam otherwise will switch back to lurk mode.03:18
pschulz01My shipit CD's just arrived..03:23
pschulz0170 CD ROMS - 50 PC; 5 Mac/PowerPC; 15 AMD/64Bit + 40 Stickers.03:24
=== pschulz01 apologises.. related.. but wrong channel.
mdzBenC: np.  I just got a panic when unplugging a slightly flaky USB device, and got an oops in khubd.  worth a bug report?03:36
=== mdz [n=mdz@studiocity-motorola-bsr1-70-36-194-85.vnnyca.adelphia.net] has joined #ubuntu-kernel
crimsunhmm. the old ndiswrapper-1.1 source package in main generates a ndiswrapper-utils-1.1 binary package that apparently doesn't work with Edgy's kernel.06:42
BenCedgy's ndiswrapper kernel module has been updates...maybe the userspace tools need to be as well06:55
BenCcrimsun: ^^06:55
=== johanbr [n=j@d154-20-189-105.bchsia.telus.net] has joined #ubuntu-kernel
=== chmj [n=chmj@196.44.1.98] has joined #ubuntu-kernel
=== Lure [n=lure@ubuntu/member/lure] has joined #ubuntu-kernel
=== nigel_c [i=nigel@nigel.suspend2.net] has joined #ubuntu-kernel
=== doko_ [n=doko@dslb-088-073-110-181.pools.arcor-ip.net] has joined #ubuntu-kernel
=== lloydinho [n=andreas@rosinante.egmont-kol.dk] has joined #ubuntu-kernel
=== pschulz01 [n=paul@202.174.42.5] has joined #ubuntu-kernel
=== nigel_c [i=nigel@nigel.suspend2.net] has joined #ubuntu-kernel
=== nigel_c [i=nigel@nigel.suspend2.net] has joined #ubuntu-kernel
=== rodarvus [n=rodarvus@ubuntu/member/rodarvus] has joined #ubuntu-kernel
=== rodarvus [n=rodarvus@ubuntu/member/rodarvus] has joined #ubuntu-kernel
=== zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-kernel
=== lfittl [n=lfittl@85-125-147-132.dynamic.xdsl-line.inode.at] has joined #ubuntu-kernel
=== pschulz01 [n=paul@202.174.42.5] has joined #ubuntu-kernel
=== kbyrd [n=Miranda@mailout1.vmware.com] has joined #ubuntu-kernel
=== jwest- [n=jwest@83.110.124.122] has joined #ubuntu-kernel
=== Keybuk [n=scott@quest.netsplit.com] has joined #ubuntu-kernel
BenCmdz: I think given the lack of any real problems with the proposal, I'm going to move forward with the image conversions07:02
BenCI'm not going to upload till I get to germany07:02
BenCso if there's any problems, we can schedule a dev session for it :)07:03
mdzBenC: sounds good07:04
mdzBenC: so what are your thoughts on the naming?07:04
BenCmdz: -386 will remain as an alternate kernels07:04
BenC-generic will become the standard kernel07:04
mdzand -generic will be fairly close to what -686 is now? (SMP-enabled)07:05
BenCbasically it will be the same as the k7/686 kernel, but targeted at 586 and above (386 is targeted at 486 and above)07:05
BenCamd64 will have -generic (which is just a rename of the -amd64-generic kernel) and remove the -xeon and -k8 versions07:06
BenC-amd64-server will be renamed to -server too07:06
BenCso we're killing three kernels07:07
BenCI'm also considering droping the non-smp kernels on ia64, since I doubt there's many non-smp ia64's around07:07
BenCI've never heard of any at least07:07
thomi'm pretty sure the workstation ia64s were uniproc07:08
thom(or the low end ones anyway.) http://www.ia64-linux.org/machines/zx2000.html07:09
=== ivoks [n=ivoks@lns02-0048.dsl.iskon.hr] has joined #ubuntu-kernel
mdzsounds good07:10
BenCthom: it may just have to run an smp kernel if they want to use it07:11
BenCwe force ppc64 users to run an smp kernel07:11
thomfair enough.07:12
BenCI never hear anything from ia64 users anyway :)07:12
BenCprobably will change with the new Itanium2 chips, bu then they are multi-core anyway07:12
BenCmdz: lrm has a "base" kernel for each arch...are we saying we want -generic to be base for i386 now?07:15
mdzBenC: what does that mean?07:15
BenCthe standard kernel used for installs and most running systems07:15
BenCI think linux-meta makes the distinction as well07:15
BenClrm uses it for udeb creation07:15
mdzI don't see why lrm should care07:15
BenCit doesn't create udebs for all kernels07:16
mdzif it's for udebs, then it should be the kernel we use on the alternate CD07:16
mdzwhich will continue to be -38607:16
BenCbecause the liveCD doesn't need udeb's, correct?07:16
mdzcorrect07:16
BenCok07:16
mdzlinux-meta is trickier, since we use that on both07:16
BenClinux-meta is going to be tricky because it's gaining a lot of rename cruft :)07:18
maswanBenC: Hey, a question, I got buried behind real work and stuff, did you get any time for a -dbg or similar kernel package with vmunix?07:19
BenCmaswan: we've had linux-image-debug-* for several kernel versions now07:19
BenCwell, several revisions of the edgy kernel07:20
maswanBenC: ah, awesome. we're just getting around to upgrading to dapper. :)07:20
maswanBenC: thanks and keep up the good work07:21
BenCmaswan: np07:21
mdzBenC: nvidia-kernel-common recommends nvidia-kernel-source | nvidia-kernel.  should the kernel provide nvidia-kernel or should we do away with that business entirely?07:37
=== allee [n=ach@allee.exgal.mpe.mpg.de] has joined #ubuntu-kernel
BenCdo we even need that package?07:39
BenCI guess we do07:40
BenCI'd say remove it07:40
BenCthat package is just a dep of the lrm stuff, so it gets pulled in when needed07:40
BenCI mean remove the recommends07:40
zulBenC: i have an 2.6.17ish xen kind of working07:53
BenCthat sounds vaguely cool :)07:54
BenCmdz: we only support single release upgrades, right (like you can't upgrade directly from breezy to edgy)?07:54
zul2.6.18 xen was cherry picked and back ported now just to cherry pick edgy07:54
kylem_BenC, you pinged last night?07:57
BenCkylem_: I forget why07:58
kylem_k :)07:58
zulgrrr...suse sucks08:09
mdzBenC: correct08:09
infinityBenC: That said, I entertain bug reports about skipped-release upgrades, if the upgrade path doesn't take me more than 10 seconds to divine and hack into a maintainer script.08:10
infinity(With most packages, we get that for free anyway, since Debian's release cycle is slower, so they handle most upgrade paths)08:11
BenCjust wondering how long I should keep linux-meta upgrade meta-packages08:15
BenClike the ones we had in dapper for -686-smp, and the ones that have been there for who knows how long for power[34]  kernels08:15
BenCI think I'm going to kill off the power cruft08:22
BenCleave the {686,k7}-smp for one more release08:22
zulso what is the end result going to be?08:23
zulso i can reflect it with the xen bits08:24
=== jwest- [n=west@83.110.124.122] has joined #ubuntu-kernel
=== jwest- [n=west@83.110.124.122] has joined #ubuntu-kernel
BenCzul: i386 => generic (standard kernel), 386, server, server-bigiron08:57
BenCamd64 => generic (standard kernel), server08:58
BenChppa and ia64 are getting their -smp versions cut (basically it will be SMP only for those arches, but no biggie)08:58
zulok cool..08:59
zulill make my changes as well08:59
BenCmdz: hey remind me to give you back that euro power adapter you lent me in paris09:01
mdzBenC: ah, so that's where it went09:02
=== jwest- [n=west@83.110.124.122] has joined #ubuntu-kernel
=== lfittl [n=lfittl@85-125-147-132.dynamic.xdsl-line.inode.at] has joined #ubuntu-kernel
=== BenC [n=bcollins@debian/developer/bcollins] has joined #ubuntu-kernel

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