/srv/irclogs.ubuntu.com/2006/06/14/#ubuntu-kernel.txt

=== ivoks [n=ivoks@ubuntu/member/ivoks] has joined #ubuntu-kernel
=== zul [n=chuck@CPE0006258ec6c1-CM000a73655d0e.cpe.net.cable.rogers.com] has joined #ubuntu-kernel
BenCzul: ping01:34
zulyeah01:46
zulwhat did i do wrong? :)01:46
BenCrofl01:47
BenCzul: in your patches, ditch the debian/changelog parts01:47
zulok01:47
BenCI manually add those, and I like to keep the patch seperate from the changelog entry01:47
BenCso it's easier to git-revert, without killing changelog history01:47
=== Face1 [n=jacob@studiocity-motorola-bsr1-70-36-195-70.vnnyca.adelphia.net] has joined #ubuntu-kernel
=== Keybuk [n=scott@syndicate.netsplit.com] has joined #ubuntu-kernel
=== zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-kernel
zuli so hate freenode03:23
rohyay!03:54
rohhttp://www.kernel.org/git/?p=linux/kernel/git/davej/cpufreq.git;a=blobdiff;h=cfc4276e670e3eac2ccc2f1c0a81f3258ff8a586;hp=28cc5d524afcd18a2dce1ee4f5ccb3d2c737faf1;hb=f081a529f808ed450c22553de7b3275e0ffde9a0;f=arch/i386/kernel/cpu/cpufreq/speedstep-smi.c fixes my bug it seems03:54
zulroh: already applied to the dapper kernel03:55
rohzul i didnt check that.. i'm currently clicking my way through the git and am impressed.. what would we do without git..  besides inventing it03:56
rohi really have to learn how to use that03:57
rohzul will there be new kernelpackages for dapper also or do i need to built my own?03:59
zulyes there will be updates dont know when yet04:00
rohzul can you point me directly to the docs how to exactly rebuild the kernelpackages the way buddha^Wthe allknowing thing wants?04:33
rohmeaning: i'm currently developing a driver and don't want to sacrifice madwifi for example04:34
=== nigel_c_ [i=nigel@203.171.70.205] has joined #ubuntu-kernel
=== dilinger [n=dilinger@207.210.101.209] has joined #ubuntu-kernel
BenCall the ppl who build custom kernels from our source are gonna love me now05:44
ajmitchwhat did you break? :)05:44
BenChehe05:44
BenCI made it easier to change the ABI, so people don't have to ask a whole lot of questions05:45
BenCpretty much needed it so I can do daily builds05:46
BenCthe method I am using uses the git hash to create a special abi on the fly05:46
ajmitchyou plan to break ABI that often for edgy?05:46
BenCkeeps ppl from getting stock kernels confused with daily builds, plus it makes it easy to know where in the git tree the build was from05:46
infinityBenC: Spiffy.05:47
BenCno, just I don't want to guarantee any ABI syncronizing for daily builds05:47
ajmitchsounds useful then05:47
BenCinfinity: yo!05:47
BenCinfinity: .43 is awaiting fabbione's blessing for a sparc patch, and is build tested05:47
infinityOh, hey, look, 24.42 actually built everywhere.05:47
BenCinfinity: what!?05:47
BenCuse 25.43 for l-r-m05:48
infinityWas it not meant to? :)05:48
BenCI thought someone told me it didn't build every :/05:48
infinityOh, wait.  I don't see a PPC build.05:48
=== infinity looks.
BenCyeah, that's the one, ABI failure05:48
BenC25.43 is locked waiting for a sparc patch, and fabbione promises to upload at 11am his time05:49
BenCit was build/boot tested on i386, ppc and amd6405:49
infinityOkay, so the PPC ABI thing is sorted?05:49
infinityThat's quite the ABI change...05:49
infinityWhy did you pull that serial stuff out of vmlinux?05:49
infinityWon't that bugger early output on serial terminals?05:50
infinityOr should openfirmware provice console/serial emulation until the driver kicks in?05:50
infinity(serial booting on my machine, like just about anything else on an OldWorld, doesn't work for beans, so I don't actually know how OF deals with it)05:51
fabbionemorning06:02
fabbioneBenC, infinity: yes max 11am UTC06:02
fabbioneBenC: so i pull and another ABI bump???06:02
fabbioneinfinity: ok i have the stuff from BenC06:10
fabbioneBenC: good night dude06:10
BenCinfinity: because of 8250 is loaded, PMACZILOG cannot work, so I had to take 8250 out as a module06:23
fabbioneBenC: i got all your stuff..06:24
fabbionei think at least ;)06:24
BenCgood deal06:24
BenCyeah, it's pushed, and should be all ready for you06:24
fabbionelinux-source-2.6.15 (2.6.15-25.43) dapper-security; urgency=low06:24
fabbione  * bcm43xx dma mask handling (> 1GB memory)06:24
fabbione  - Add powerpc IOMMU handlers for dma masks06:24
fabbione  - Sync updates to amd64 IOMMU to handle dma masks06:24
fabbione  * Sync ide changes for better flash driver handling (flash isn't removable).06:24
fabbione -- Ben Collins <bcollins@ubuntu.com>  Tue, 13 Jun 2006 08:16:30 -040006:24
fabbionethat's all i have here...06:24
fabbionecan you confirm?06:24
fabbionei still have the powerpc.ignore file06:25
fabbionethat can go away06:25
fabbionebut i can do it06:25
fabbioneBenC: how do i git tag the release?06:28
fabbioneBenC: or are you going to do it later?06:28
BenCjust push your final, and I'll retag itr06:28
fabbioneok works for me06:28
fabbioneBenC: are you also syncing the changes from .15 into .17?06:49
BenCyep07:16
BenCI pull from dapper to edgy to make sure we keep everything in sync07:16
BenCit's nice07:16
BenCdpkg-deb: building package `linux-headers-2.6.17-2-f23859-386' in `../linux-headers-2.6.17-2-f23859-386_2.6.17-2.2_i386.deb'.07:18
BenCsweetness07:18
=== maswan [i=maswan@kennedy.acc.umu.se] has joined #ubuntu-kernel
fabbionewth..07:26
fabbioneBenC: btw can you pull from my edgy tree?07:27
fabbioneit's a simple change in the control files for the cluster suite07:27
BenCok07:34
BenClet me push later so you can merge and repush07:34
fabbioneok sure07:44
fabbioneBenC: [    0.829976]  BUG: soft lockup detected on CPU#2!08:16
fabbionei get tons of these for different CPU with .17-1.108:16
fabbione(sparc)08:16
=== Lure [n=lure@ubuntu/member/lure] has joined #ubuntu-kernel
=== ivoks [n=ivoks@ubuntu/member/ivoks] has joined #ubuntu-kernel
=== chmj [n=chmj@196.44.1.98] has joined #ubuntu-kernel
=== JaneW [n=JaneW@196.36.161.235] has joined #ubuntu-kernel
=== nigel_c [i=nigel@203.171.70.205] has left #ubuntu-kernel []
=== doko_ [n=doko@dslb-088-073-102-116.pools.arcor-ip.net] has joined #ubuntu-kernel
=== nigel_c [i=nigel@203.171.70.205] has joined #ubuntu-kernel
=== zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-kernel
zulhola01:56
=== chmj [n=chmj@196.44.1.98] has joined #ubuntu-kernel
=== ivoks [n=ivoks@ubuntu/member/ivoks] has joined #ubuntu-kernel
=== JaneW [n=JaneW@196.36.161.235] has left #ubuntu-kernel ["Bye"]
=== Lure [n=lure@ubuntu/member/lure] has joined #ubuntu-kernel
mjg59BenC: If we're going to shift to libata pata drivers, we probably want to do it soon02:49
mjg59Some of them are going to need testing02:49
=== Keybuk [n=scott@quest.netsplit.com] has joined #ubuntu-kernel
=== tuxmaniac [n=aanjhan@60.254.67.17] has joined #ubuntu-kernel
=== mgalvin [n=mgalvin@ubuntu/member/mgalvin] has joined #ubuntu-kernel
fabbioneBenC: please pull from both my dapper and edgy tree.03:30
fabbioneBenC: dapper is uploaded and building. Please no more uploads till sparc is released 03:30
fabbioneBenC: otherwise we look fine.03:30
=== mkrufky [n=mk@68.160.103.77] has joined #ubuntu-kernel
=== Lure [n=lure@ubuntu/member/lure] has joined #ubuntu-kernel
zulBenC: #45679 ill update the driver tonight if thats ok?05:10
BenCwhich driver?05:11
zularcmsr05:11
BenCsure thing05:12
BenC    [UBUNTU:arcmsr]  Update to latest patches in -mm tree.05:13
BenCthat's the last place I got patches from05:13
BenCmake sure it's the same set of stuff05:13
zulsure..i was going to get it from the vendor apparently it works according to the bug reports05:13
BenC1.20.00.1305:14
BenCit says it was fixed in 1.20.0X.12, and we have the above version05:15
BenCthat's a little odd05:15
zulyeah..05:15
zulill look into it tonight then05:15
BenC1.20.00.12    9/30/2005       Erich Chen     bug fix with 64bit platform's ccbs using if over 4G system memory05:16
BenCthat's in our ChangeLog.arcmsr05:16
zuljust a sec..05:16
BenCDocumentation/scsi/ChangeLog.arcmsr05:16
BenCand that was from the patches from -mm05:16
zulof course the vendor website has to be down05:18
=== TheMuso_ [n=luke@dsl-202-173-132-131.nsw.westnet.com.au] has joined #ubuntu-kernel
=== JaneW [n=JaneW@dsl-146-167-152.telkomadsl.co.za] has joined #ubuntu-kernel
zulBenC: thats in BenC yeah thats just weird05:25
zulwhat the hell?05:26
ajmitchblame the keyboard gremlins05:26
zuli blame the irq between my brain and keyboard05:27
zulBenC: are you going to be in paris?05:32
BenCyep05:32
zulcoool..05:32
zulhave fun i wont be05:33
=== Lure [n=lure@ubuntu/member/lure] has joined #ubuntu-kernel
=== ivoks [n=ivoks@ubuntu/member/ivoks] has joined #ubuntu-kernel
zulwhen are the security updates going to show up?06:53
=== mgalvin [n=mgalvin@ubuntu/member/mgalvin] has joined #ubuntu-kernel
=== piKant [i=piKant@i577A1E22.versanet.de] has joined #ubuntu-kernel
mkrufkyi spoke to the DViCO systems engineer....   Bug #12162 is bogus. the "fix" from that bug causes bug #33096, which is valid.  The fix would be to revert ubuntu git commit 3a580f39df8937f54976b2e7ce2ef54294da06f9 .....  I have added the appropriate comments to those bugs on launchpad.net07:53
=== piKant [i=piKant@i577A1E22.versanet.de] has left #ubuntu-kernel [">]
mkrufkyBenC: should I send you a patch to revert that commit?  ...or are the launchpad.net comments enough?07:56
BenCmkrifky: Yes, please07:56
BenCmkrufky even07:56
mkrufky:-)07:56
mkrufkyBenC: oh, man.... i just finished cloning ubuntu-2.6.git , and it seems that the bogus commit got automagically reverted as a result of a pull from upstream08:17
mkrufkyhowever, ubuntu-dapper.git still has the bogus patch08:17
mkrufkyi guess i'll have to re-clone from the dapper tree08:18
=== zul [n=chuck@ubuntu/member/zul] has joined #ubuntu-kernel
fabbioneBenC: please hold any dapper uploads please :)08:28
BenCfabbione: wont be doing dapper again till after paris08:32
fabbioneBenC: kthx :)08:33
dilingermeh, i wanted to go to paris :(08:38
zuljoin the club 08:39
zulwhats with these ubuntu conferences when the city speak french?08:39
fabbionezul: hopefully the last one :)08:42
zulhehe08:42
zulill go to the next one08:43
zulonly if they dont speak french ;)08:43
dilingeri was expecting work to send me to this one08:45
dilingerbut things at work aren't going so well08:45
dilingerso.. *shrug*, doesn't look like it's going to happen08:45
mkrufkyBenC: ok, i uploaded the fix to launchpad bug #3309608:47
=== mgalvin [n=mgalvin@ubuntu/member/mgalvin] has joined #ubuntu-kernel
zulfabbione: where can i get the sparc iso?09:15
zullater09:43
=== kimo [n=ahmed@196.202.31.57] has joined #ubuntu-kernel
kimosorry for being n00b, but when it says '2.6.17-1.1' is released, it means released in the git tree only, right ? (no debs?)09:55
crimsun_it means it was uploaded.  [https://lists.ubuntu.com/archives/edgy-changes/2006-June/000040.html] 09:57
kimowould this new version, include fixes commited to the offcial dapper kernel ?09:59
kimolike the recent fixes to my sky2 driver ?09:59
rohoh.. edgy already started?10:00
kimoroh: yeah :) ubuntu+110:01
crimsun_kimo: see the sky2-related commits at http://www.kernel.org/git/?p=linux/kernel/git/bcollins/ubuntu-2.6.git;a=shortlog10:01
rohthen i hope soon somebody takes the time to make rootfs via uuid working10:02
kimocrimsun_: thanks, great :) I am still not sure if there is a 'nightly build' deb package I can install to get those fixes 'now' ?? (or should I compile from a git tree) ?10:04
crimsun_kimo: I presume the URL mentioned in [https://lists.ubuntu.com/archives/ubuntu-devel-announce/2006-June/000150.html]  will be populated10:05
kimoohh cool :) bcollins makes a daily build system, which is unfortunately empty ;) http://people.ubuntu.com/~bcollins/kernels-daily/10:08
kimook I found the deb in upper dir http://people.ubuntu.com/~bcollins/10:09
kimodownloading, thanks ;)10:11
=== ivoks [n=ivoks@ubuntu/member/ivoks] has joined #ubuntu-kernel
rohmmmh.. should i build from linux-source-2.6.15 or from the sourcepackage for linux-image-2.6.15-23-386 ?10:49
crimsun_the latter.10:50
rohtnx10:51
rohwhich way of packaging is the recommended then? via make-kpkg or dpkg-buildpackage?10:53
crimsun_either, depending on whether you want to use the infrastructure. If you're not sure, choose the latter.10:54
mkrufkyroh: are you rebuilding your kernel right now just for the sake of the dvb development?10:55
mkrufkyroh: that hg repo i told you about is backwards compatible with older kernels10:56
mkrufkyso... if that's what you're doing.... you can save yourself some time10:56
rohmkrufky no.. mostly because speedstep does not work with dapper release on my tp-x2010:56
mkrufkyah, okay10:56
rohand i want piix ide fixed in the kernel since its the rootfs, and the initrd sometimes loads the sii-ide before piix and my root is  hde then ;)10:57
mkrufkyok -- valid reasons  ;-)10:59
kimoroh: u might want a newer prebuilt from http://people.ubuntu.com/~bcollins/ ?11:02
rohkimo if there is enough source with it to build other modules later which need a working makefrile from a kerneltree11:03
=== ivoks [n=ivoks@lns02-0909.dsl.iskon.hr] has joined #ubuntu-kernel
BenCroh: speedstep will be fixed for you in the next kernel update for dapper11:27
rohBenC is there a package which gives me a /lib/modules/`uname -r`/source symlink to working makefiles and headers for building external drivers which depend on headers of existing drivers?11:38
BenClol11:38
BenCapt-get install linux-headers-`uname -r`11:39
rohi know that package.. but it does not contain tda1004x.h for example, which i need for my driver since the attach function for this demod is declared there11:42
rohhm.. but... forget it.. i propably just compile the whole dvb stuff from hg, so i can use the header from there...11:43
BenCroh: linux-source-2.6.1511:44
rohBenC do you know when the new packages get out?11:45
BenCthe dapper 2.6.15 update, or edgy 2.6.17?11:47
BenCwell, either will be available within a day or two11:47
rohdapper.. for now i'll propably stay with dapper, even if i'd like the new kernel.. its a 600mhz tp-x21.. no fun with a few hundred updates every week11:47
mkrufkyroh: seriously -- dont worry about the dvb frontend sources in the ubuntu kernel... you REALLY need to use the hg stuff11:52
mkrufkyroh: you should work within the hg v4l-dvb tree on linuxtv.org11:52
mkrufkyotherwise you're going to have a real pain once you're ready to merge your new driver11:53
mkrufky... (or I will)11:53
mkrufkyroh: i'm leaving now... if you need help later with the hg stuff, i'll be back in #linuxtv in a few hours11:58
mkrufkybye all11:58

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