/srv/irclogs.ubuntu.com/2012/12/13/#ubuntu-kernel.txt

=== chiluk is now known as chiluk_away
infinityThis is disconcerting.  Has anyone else noticed a tmpfs leak on raring?01:00
infinity(base)adconrad@cthulhu:~$ du -h /var/lib/schroot/union/overlay01:02
infinity0/var/lib/schroot/union/overlay01:02
infinity(base)adconrad@cthulhu:~$ ls -l /var/lib/schroot/union/overlay01:02
infinitytotal 001:02
infinity(base)adconrad@cthulhu:~$ df -h /var/lib/schroot/union/overlay01:02
infinityFilesystem      Size  Used Avail Use% Mounted on01:02
infinityschroot          12G  792M   11G   7% /var/lib/schroot/union/overlay01:02
infinityHrm, Google points me at http://www.gossamer-threads.com/lists/linux/kernel/164326601:07
=== chiluk_away is now known as chiluk
=== bantu_ is now known as bantu
=== chiluk is now known as chiluk_away
=== chiluk_away is now known as chiluk
=== kentb__ is now known as kentb
=== chiluk is now known as chiluk_away
=== dupondje_ is now known as dupondje
=== smb` is now known as smb
=== henrix_ is now known as henrix
* henrix curses at the ppc builders09:52
ckingppc backlog?09:53
henrixyep09:57
henrixhuge!09:57
henrixi've uploaded the quantal kernel tuesday evening and its still on the queue :)09:58
ckingneed more ppc builders...10:12
* henrix nods10:14
Laneydoes the nexus 7 kernel follow the usual kernel team process for submitting changes?10:20
ckingi believe so10:21
* Laney nods10:21
apwLaney, yes it would indeed10:25
Laneyok cool10:26
Laneyincoming </worms voice>10:26
=== amitk is now known as amitk-afk
henrixapw: we have a few Hiper-V bugs to be verified in P. do you have any chance to take a look at them?10:56
apwhenrix, i probabally won't this week on a fire drill right now.  that said i would ask utlemming (who will be about in about an hour) as he i think said they had tested the hyper-v stuff yesterday; so it may already be done10:57
henrixapw: ok, cool. i'll ask him about that. thanks10:58
StFSHi. I was hoping that someone could maybe help me a bit. I just reported a bug on my wireless NIC, it sometimes (not always) doesn't come back after suspend/resume. I reported the bug when everything was ok (the wireless NIC was up and running) but now I have the problem again but I'm connected to a wired connection.11:09
StFSI'm just wondering whether there is anything I should do now to add more information to the bug? Any tools or commands I should run or anything like that?11:09
StFSthe bug in question is this one: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/108985611:10
ubot2`Launchpad bug 1089856 in linux (Ubuntu) "Wireless network doesn't resume after suspend/resume (sometimes)" [Undecided,Confirmed]11:10
apwStFS, is this a PCI device?  if so an lspci and a dmesg from 'now' as it just happened cannot hurt11:11
apwthe dmesg is the key one i guess, it may have some hints as to the failure during resume11:11
StFSapw: hmm... sorry, dmesg from 'now'? Is there some option to dmesg I don't now about or do you just mean now as in "since the problem is showing now"?11:15
apwi mean the latter11:15
StFSok :) will do that11:16
StFSdo you know the lspci format that is used with apport? I want to run it the same way11:23
StFSprobably just lspci -v11:24
StFSwell that looks a bit suspicious: 03:00.0 Network controller: Intel Corporation Centrino Ultimate-N 6300 (rev ff) (prog-if ff) !!! Unknown header type 7f11:27
apwStFS, it may be -vvnn actually11:31
StFSapw: thanks, I'll update it11:31
=== amitk-afk is now known as amitk
=== yofel_ is now known as yofel
=== henrix is now known as henrix_
=== henrix_ is now known as henrix
jgdxI might be asking in the wrong place, but why is the virtualbox linux kernel driver available to the 3.6.7 kernel but not to the 3.7.0 (rc8) kernel? 13:33
* henrix -> lunch13:37
=== chiluk_away is now known as chiluk
* smb gets more coffee while email client struggles14:06
amitksmb: while email client struggles to get coffee? ;)14:07
smbamitk, nah, it struggles to handle herton 's mails14:08
smartboyhwsmb, my gmail went down because of that:P14:08
hertonhaha, sorry, doing some spamming today :)14:08
amitksmb: aah, that. Good point, I meant to ask herton if there is a reason why he doesn't thread his emails14:08
smartboyhw241...........14:09
* smartboyhw is ready to mark all the emails read:P14:09
hertonamitk, the patches added to the queue are not threaded, but the review patches are14:09
smbamitk, They appear to be threaded. Though that is relative to how far tb got aware of them before imapfilter rips em away14:10
amitkherton: ok. Just trying to minimize the number of Ctrl-r or Ctrl-d i have to hit in mutt14:11
hertonok, all 241 patches are sent now. I expect to not have this much on next releases (I've caught up now with what was pending)14:16
ckingmailbox overflow15:10
* ogasawara back in 2015:19
=== chiluk is now known as chiluk_away
keesapw, ogasawara: raring c28f49fd37b32bbc88b86a6ff01aeb28948c7aaa disabled SLUB_DEBUG which kills off /proc/slabinfo that is needed by tools like stabtop.15:58
kees(this is a regression in raring)15:58
apwkees, ok15:58
kees(credit to sbeattie for seeing the problem)15:59
* amitk likes stabtop as the name of a Linux tool15:59
keesapw: was SLUB_DEBUG removed for a specific reason? (and can it be put back?)15:59
apwkees, i cannot recall, either we didn't think it was useful and was costly, or as likely right now, i may have zapped a couple wrong in an update16:00
apwwill check it out 16:00
keesthanks!16:01
ckingit was kinda useful, but a little bit costly for small allocations16:08
hggdhsmb: So Xen 4.2 is out. xen-hypervisor-amd64 was yesterday (at least for a while) a purely virtual package with no installation candidates. How is this package updated to the current Xen version?16:08
ckinghttp://kernel.ubuntu.com/~cking/slub-debug/16:08
keescking: what about slab?16:09
keesI have no strong opinion about slabinfo, but some people live by it.16:09
ckingwell if it's useful, then it's probably well worth the teeny weeny overhead16:10
argescking, your performance study is pretty awesome. I would have guessed slub_debug would be more expensive than that16:12
smbhggdh, I'd need to check where that came from exactly but there is xen-system-(i386|amd64). Though it should probably be noted that right now upgrading would not remove the 4.1 utils and hypervisor but install the 4.2 in parallel16:12
argescking, but i agree slub_debug is useful to have esp. when we get those really painful kernel memory bugs16:12
ckingarges, yep, we should therefore keep it in16:13
hggdhsmb: in my case it does not matter, since it I am installing Xen on a fresh system. For the xen-hypervisor-amd64, this is the package you suggest to install on the wiki16:14
hggdhsmb: also, the Grub name for Xen changed on 4.2 :-)16:15
smbhggdh, So maybe the first one we may have messed up right now. You can use xen-system-amd64. The grub thing someone else is to be blamed but I guess you may update the wiki page ;)16:17
=== chiluk_away is now known as chiluk
hggdhsmb: I am going to update it, just wanted to be sure about it16:22
smbhggdh, Well both would gonna be variables based on release. The xen-system-* meta packages should be working for Quantal onwards. And I believe it was Q as well that renamed the whole of grub entries. 16:32
smbAbout the xen-hypervisor-<arch> thing... not sure it seems to appear in the provides section of the xen-hypervisor-4.x-<arch> packages and still exist in raring's xen-4.2...16:33
=== josepht_ is now known as josepht
smbhggdh, So xen-hypervisor-amd64 seems like it would work today. Did not say "yes" to install it into my xen domU I used to check though... ;(16:44
smbI meant ;)16:44
hggdhsmb: ok, so the xen-hypervisor-amd64 error I got yesterday was probably due to the packages not being completely up-to-date16:47
smbhggdh, I would assume. It was just released or moved out of proposed then16:48
hggdhyeah.16:48
henrixinfinity: any idea about when we'll have a free slot with the ppc builders?17:13
sbeattiekees, apw: actually hggdh deserves the credit for catching the SLUB_DEBUG change with QRT17:15
infinityhenrix: Define "free slot".17:17
bjfinfinity, we want some ppc love for a kernel build17:18
henrixinfinity: some time where our kernels will be allowed to build :)17:18
infinityThe quantal SRU?17:18
henrixyep17:18
bjfyes17:18
bjfwe had a regression and had to respin17:19
infinityHow's 4 minutes sound?17:19
bjfwfm17:19
vanhoofbjf: does that pull in the fix for that bug that multiplied a few times henrix sorted out?17:19
* vanhoof thinks it was henrix at least17:19
vanhoof-ETOOMUCHMAIL17:19
bjfvanhoof, yes 17:20
vanhoofsweetness17:20
bjfvanhoof, the gfx driver we all love so much17:20
vanhoofthe bane of my existence17:20
henrixinfinity: thanks!17:36
infinityhenrix: No problem.17:39
infinityhenrix: I'll check back later today and promote/override.17:39
henrixinfinity: cool, thanks a lot17:40
infinity(Looks like it'll take adare ~6h to build that)17:40
infinityThat new hardware can't come a moment too soon.17:40
keesapw: btw, who should I re-send this to? http://git.kernel.org/?p=linux/kernel/git/kees/linux.git;a=commitdiff;h=ab2ad33a9510fd3588fd431ceff5891992e5999e18:10
apwi thought that went in18:11
keesoh? maybe I missed it. let me check linux-next18:11
apwif you poke me on monday i'll see what i can make happen, slammed right now18:11
keescool, thanks18:12
keesapw: ah-ha, yes, it is. nm!18:13
=== chiluk is now known as chiluk_away
=== chiluk_away is now known as chiluk
=== chiluk is now known as chiluk_away
=== chiluk_away is now known as chiluk
infinityhenrix: Is there a linux-lts-quantal set being uploaded to match this quantal emergency SRU?18:48
henrixinfinity: yes, i was just waiting for the bot to generate the tracking bug #18:48
infinityOh, right, it doesn't do that until they've all built. :/18:49
infinityThat still feels like a misfeature to me. :P18:49
henrixinfinity: yeah. anyway, the -lts doesn't require ppc builders :)18:49
infinityMaybe we can discuss said misfeature after Christmas, but for now, can you manually prod it to create the bug just for lts-q?18:50
infinity(It can build in parallel with the PPC one, and they can all go in together was my motivation).18:50
henrixherton: ^18:51
henrixinfinity: i guess we could create it manually, yes...18:51
infinityWell, it's either that, or we get a bit of a turnaround delay, I'm not picky either way, I suppose.18:51
infinityOh, the regression is only in -proposed, I guess this isn't as urgent as I initially thought.18:52
henrixyeah, there's that as well18:53
infinitySo, I'm going to pull these from quantal-proposed and precise-proposed.18:54
infinityThe regression was only in i915, right, so ARM kernels are unaffected?18:54
henrixcorrect18:54
hertonthe bot waits only to ensure at least master builds before someone rebases to a derivative or backport18:55
infinityIs anyone looking at the new bugs referenced here?18:56
infinityThe ftrace stuff?18:56
hertonbut if that is not too much desired, could again be changed. Just got more noticed now with this ppc long wait18:56
infinityOr is this the same bug?18:56
henrixftrace?18:56
infinityAhh, yeah.  It's all the i915 nug.18:56
infinitys/nug/bug/18:56
henrixherton: yeah, i guess that's the correct behaviour. maybe i should just reuse the old tracking bug18:57
infinityWas just reading hggdh's comments, s'all.18:57
henrixah, ok. that's what's shown on the stacktrace. yep, that's the i915 thing18:57
hertonhenrix, you can also open a new one, and close the old one and the new one which the bot opens as duplicates18:58
hertonhenrix, probably on the old one you will have to reset all the tasks, but any way is fine18:59
henrixherton: ah, right. i could use the script to do that.18:59
apwhenrix, henrix, how close to having -21 (quantal) in the -proposed pocket are we ?19:00
bjfhenrix, we really want that lts-quantal19:01
bjfapw, we're just waiting on ppc :-P19:01
bjfapw, and a week of testing19:01
apwbjf, i just care about it being in -proposed19:02
apwright now the source is out of sync with published reality19:02
bjfapw, then it's just the ppc issue19:02
henrixbjf: ack, i'll have it uploaded today 19:02
bjfapw, we've been waiting since tue. to get it built19:02
infinitySomeone should have mentioned it earlier. :/19:03
apwwhen we going to get some more horses in that queue19:03
bjfinfinity, so we should just nag you constantly?19:03
infinitybjf: Sometimes. :P19:03
infinityAnyhow, failed SRUs removed from -proposed.19:03
bjfsforshee, do you have the link to the amd64+mac iso images. i just seem to be going around and around in circles trying to find it19:05
infinitybjf: For which release?19:06
bjfinfinity, quantal19:06
sforsheebjf, I noticed recently that they seem to be impossible to find on ubuntu.com19:06
infinityhttp://releases.ubuntu.com/12.10/19:06
bjfinfinity, thanks. for some reason my fingers keep typing cdimage.ubuntu.com19:07
sforsheebjf, if you want to do an efi install the standard images should work fine too19:07
infinitybjf: Yeah, for released images, there's a weird split, where supported stuff lands on releases.u.c, and everything else on cdimage.u.c/releases/19:07
infinity(Because lots of people mirror releases.u.c and don't want 500GB of crap)19:08
bjfsforshee, i'm actually not sure how i'm installing on this air. i have an OSX partion and refit19:08
sforsheebjf, actually I think the standard image also work for a csm install as long as you're using a usb stick. I want to say the +mac images are only needed for actual CDs.19:08
bjfsforshee, i can certainly try both to see19:09
sforsheebjf, for an air an efi install works fine, and subjectively I think the performance is a little better19:09
bjfsforshee, how can i tell ?19:09
mjg59The +mac images are only needed for CSM installs on old Macs19:20
sforsheebjf, sorry, I got a phone call19:27
sforsheedid you get what you need?19:28
bjfsforshee, think so19:38
sforsheebjf, let me know if you have any more questions19:39
bjfsforshee, well, refit doesn't seem to like the non mac iso19:47
sforsheebjf, are you using dd or usb-creator-gtk? I have better luck with the latter19:50
bjfsforshee, dd on precise19:51
bjfsforshee, i'll try usb-c-g19:51
sforsheebjf, that should give you better results19:51
bjfsforshee, not so much20:06
sforsheebjf, really20:09
sforsheeI haven't used refit in a while20:09
sforsheemaybe there's something there I forgot about20:09
sforsheebut what should work is to hold alt while you boot the macbook and select the usb stick from the apple boot picker20:10
bjfsforshee, so i hold down the alt key and I am presented with 5 icons. 3 disk and 2 usb. the 3 disk icons are labeld 'rEFIt', 'Windows', 'Recovery HE' the two USB icons are 'Windows' and 'EFI Boot'20:12
sforsheebjf, you should obviously pick one of the usb options, I'd suggest going with 'EFI Boot'20:13
sforsheethat should get you to the installer, and after you install refit *should* see the efi executable for grub and offer it to you as an option20:14
bjfsforshee, so yes, that does seem to work20:18
sforsheebjf, good. Was the problem with refit that it didn't display the USB boot options or that it failed to boot them?20:19
bjfsforshee, given all the odd icons refit was showing and no labels it was hard for me to decide what to select20:20
bjfsforshee, i'm probably responsible for configuring it to not display labels20:21
sforsheebjf, heh20:21
bjfbecause that just wouldn't be as cool, you know20:21
bjfthat does seem to have borked refit though20:24
sforsheed'oh20:24
sforsheeI wonder why20:24
bjfsforshee, so my boot options are now "rEFIt", "Windows", "Recovery HD" (no usb stick in it and just holding down the alt key)20:26
sforsheeslangasek, are you guys planning to integrate all the iso image changes and apple boot picker integration that mjg59 figured out? That would sure make life easier for folks installing ubuntu on macs.20:26
bjfsforshee, "rEFIt" and "Windows" lead to unbootable grub menu20:26
bjfsforshee, so i can't seem to boot to my good OSX either20:26
sforsheebjf, that's really odd and unfortunate20:27
sforsheeI don't know what the installer did to bork refit20:28
sforsheebjf, so I suppose your options are to boot the live image and try to fix it or to boot to the recovery partition and start from scratch20:29
bjfsforshee, yup20:29
sforsheebjf, refit is stored on the hfs+ partition, and I didn't think our installer touched anything there20:31
sforsheebjf, the refit option takes you straight to grub without any intermediate menu? I can't figure out why that would happen.20:34
infinityhenrix: Thanks for getting on top of the lts-q uploads.20:40
infinityhenrix: I'll promote the whole mess this evening when everything's done.20:40
infinityhenrix: (don't forget linux-signed)20:40
bjfsforshee, i'm mistaken, the refit got me to three other icons one of which lets me boot my osx install, so that's still there20:42
sforsheebjf, and the other two get you to a broken grub?20:43
sforsheefwiw if you were booting ubuntu with CSM before then the 'Windows' option is probably just a vestige of that install20:43
sforsheebjf, at least that makes more sense :-)20:44
sforsheeI was really puzzled about how refit got broken20:44
bjfsforshee, yes, but after one boot into osx, the refit is now missing ... still figuring this out20:45
sforsheebjf, do you just have the OSX option instead?20:45
=== BenC- is now known as BenC
=== ehw_ is now known as ehw
bjfsforshee, so i had to reboot into the osx config once, that seems to have sort of restored refit20:47
bjfsforshee, now when i boot and don't hold down alt, i get 3 icons, one (that used to boot ubuntu) and now just gets me to "error: file '/boot/grub/i386-pc/normal.mod' not found. grub rescue>"20:49
bjfsforshee, a second icon which boots into the ubuntu i just installed20:49
bjfsforshee, and a third icon which boots into my OSX20:49
bjfit's odd but useable20:50
sforsheebjf, that's more or less what I expected to happen. The option that doesn't work is likely leftover from a CSM install.20:50
bjfsforshee, i agree though it was confusing20:51
sforsheebjf, I have no idea why you had problems initially. Everything to do with refit is in the hfs+ partition, which shouldn't have been touched by the installer.20:52
slangaseksforshee: yes, that's definitely the goal20:55
sforsheeslangasek, sweet :-)20:56
slangaseksforshee: if you have some time tomorrow afternoon, I think it would help me to pick your brain further so I can better understand exactly what I'm seeing currently... there are a lot of moving parts20:56
sforsheeslangasek, sure. You'd better catch me tomorrow because I'll be on holiday for the following two weeks20:57
slangaseksforshee: righto - sorry, I misunderstood earlier and thought you were /already/ on vacation, or I would've been hunting you down earlier this week :)20:57
sforsheebjf, you might be interested to know that in 3.8 the wifi on the macbook air should work somewhat better in 3.820:58
sforsheebjf, I'm still chasing down some of the problems though. I've found serious issues with background scans this week.20:58
sforsheeslangasek, nope, still working. My schedule is wide open tomorrow so just grab me whenever is convenient for you.20:58
bjfsforshee, at one point i did have raring on this, i'm actually trying to get back to it21:00
sforsheebjf, I put raring on my air this week and it seems to be working fine21:00
bjfsforshee, the daily iso today didn't work for me so i went down the rabbit hole21:00
bjfand got very lost21:00
sforsheeuh oh. I haven't updated in the past couple of days, maybe that's a good thing21:00
bjfi'm hoping it's just the daily image being the issue, couldn't get past the purple Ubuntu screen21:01
bjfbut now i may try it again doing it the same way i just put quantal back on it to see if it will work21:02
sforsheeI'm updating to see if my machine still works okay21:03
sforsheebjf, I'm now fully up-to-date and raring seems to be working fine21:13
bjfi'm creating a new usb stick with today's raring, will give it a spin shortly21:13
bjfsforshee, nope, seems to be the image is borked21:19
sforsheebjf, I'll give it a try, probably tomorrow. I did an upgrade from quantal so I haven't tested our raring images on this machine yet.21:20
bjfsforshee, interesting, if i "try ubuntu" it comes up21:20
bjfbut if i go directly to "install ubuntu" it doesn't work21:21
sforsheebjf, have you tried it on any other machines? Maybe that's just plain broken in general right now.21:23
henrixinfinity: i'll upload the -signed in a minute. i had to have the main pkg built first before uploading21:23
bjfsforshee, no, not tried it anywhere else ... good point21:23
infinityhenrix: The build-deps should be versioned, no?  It'll just dep-wait if they upload out of sync.21:25
infinityhenrix: (Which is what happens in the primary archive)21:25
=== henrix is now known as henrix_
=== henrix_ is now known as henrix
bjfsforshee, the install went fine as did a post-install dist-upgrade21:36
sforsheebjf, you're referring to the raring install on the macbook?21:37
bjfsforshee, yes21:38
bjfsforshee, from a live desktop21:38
sforsheebjf, good deal21:38
=== jussi01 is now known as jussi
=== chiluk is now known as chiluk_away
henrixapw: still around?22:26
wmphello, where i can found package source to linux-3.5.0-1922:27
=== ppetraki_ is now known as ppetraki
infinitybjf: Was the update on #1088979 a manual nudge of the bot, or did it screw up?22:28
infinitybjf: (It claimed the package was built and published an hour before PPC finished)22:28
bjfinfinity, appears to be a bug22:29
henrixinfinity: yeah, i was discussing this earlier with herton. he (or me) will investigate this later22:30
infinityMmkay.22:30
infinityhenrix: Also, going to dupe 1090126 to 1087221, I assume?22:30
henrixinfinity: ah, right. will do22:32
henrixapw: ping22:34
infinityhenrix: While I'm complaining about the bot, any reason that 1087221 is stuck in Prepare-package?  Perhaps a problem with having reset all the tasks and reused the bug?22:54
infinity(I'm happy to manually mark that Fix Released and do my promotion bits, but if the bot's "stuck", we might want to know)22:54
henrixinfinity: hmm... maybe the bot isn't running? bjf ^^22:55
bjfhenrix, infinity just ran it by hand, it just changed promote-to-proposed state for that bug23:08
* vanhoof waits with anticipation and <3 for bjf ;)23:09
vanhoofnot infinity though23:10
vanhoofhe said he's not buying me a birthday present23:10
bjfhenrix, i'm not sure what is up but git is struggleing to pull your quantal-lbm23:11
bjfhenrix, the problem is on my end somehow23:12
henrixbjf: that's odd, because i've just pulled it into tangerine and it was quick23:12
henrixah23:13
bjfhenrix, the quantal trees look good23:13
henrixbjf: cool. i'll push them and upload the src pkgs23:14
henrixthanks23:14
infinitybjf: Thanks for that.  So we're all set to promote and override quantal and lts-quantal?  Will do.23:24
infinityAnd then I will get back to important holiday tasks like re-watching Arrested Development.23:24
bjfinfinity, we'll have quantal-lbm and precise-lbm pkgs that we're trying to "rush" through23:25
henrixinfinity: btw, just fyi, i'm uploading a few more pkgs at the moment...23:25
henrixinfinity: namely: precise and quantal -lbm and -meta packages23:25
henrixinfinity: these are just minor version bumps btw23:26
bjfheh23:26
infinityOh, for they who shall not be named, yes. :/23:26
bjfyup23:26
infinityWell, I'm promoting what's there now.  When those are ready to go, let me know, and I'll review and push.23:26
bjfack23:27
infinityDo they have their own separate bug?23:27
henrixthey should start building in a few seconds23:27
henrixno, they don't have a bug tracking bug as they are just minor version bumps23:27
henrixor should they...?23:27
infinityI didn't mean a tracking bug, but an SRU bug at all.23:27
infinityThey should be fixing a bug, surely.23:27
henrixah, they do have an sru bug #23:28
bjfinfinity, yes, we are following the rules :-)23:28
henrixinfinity: and another request... could you make sure the -meta packages won't get stuck with the ppc builders? :)23:29
infinityhenrix: Absolutely.23:29
henrixinfinity: cool, thanks 23:30
infinityBumped for precise.  Will do the same for quantal when I see it.23:31
infinityThough, both will get stuck behind the current webkit build.23:32
infinityWhich is in its final link, at least, so probably only an hour or two, tops.23:32
henrixcool, thanks23:32
bjfogasawara, the lbm and meta packages have been uploaded23:32
* henrix -> EOD23:33
infinityI wonder what Laney and micahg did to make webkit take twice as long to build. :/23:33
=== henrix is now known as henrix_
infinityapw: Was there a reason those lbm-meta packages are -virtual instead of -generic, if they're built agains the generic kernel?23:48
infinityapw: Seems a little inconsistent with the naming of all the other lbm metapackages.23:49
infinityapw: Actually, precise is a bit strange there too, since you seem to be building both -generic and -virtual packages from lbm, but then the meta has the virtual->generic cross-depend.23:52
infinityapw: So, neither of these looks quite right to me.23:52
infinityogasawara: ^23:52
infinityYeah, meta is definitely wrong for both of these.  lbm looks fine, though.23:58
bjfinfinity, sigh23:58
bjfinfinity, looking23:59
infinitybjf: So, for Q, meta should just have generic->generic metapackages.23:59
infinitybjf: For P, it needs generic->generic, generic-pae->generic-pae and virtual->virtual.23:59

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