=== horms [~horms@YahooBB219184134048.bbtec.net] has joined #ubuntu-kernel === horms [~horms@YahooBB219184134048.bbtec.net] has left #ubuntu-kernel [] [12:37] fabbione: those video msleep patches have been pushed upstream [01:18] fabbione: Not massively urgent, but it would be good to get them tested soon [01:18] It can wait a while === zul [~chuck@CPE0006258ec6c2-CM000a73655d0e.cpe.net.cable.rogers.com] has joined #ubuntu-kernel [01:59] any reason to keep linux-source-2.6.11 around in breezy? [02:03] i dont see why not imho === netjoined: irc.freenode.net -> tolkien.freenode.net === TheMuso [~luke@dsl-202-173-132-131.nsw.westnet.com.au] has joined #ubuntu-kernel === mdz [~mdz@ca-studio-bsr1o-251.vnnyca.adelphia.net] has joined #ubuntu-kernel === horms [~horms@vagw.valinux.co.jp] has joined #ubuntu-kernel [03:12] hmmm...there is usb suspend/resume config option [05:22] fabbione: you around? [05:56] hey lamont [05:56] howdy [05:56] how goes it? === horms [~horms@vagw.valinux.co.jp] has joined #ubuntu-kernel [06:12] morning [06:17] we can upload the kernel today! [06:17] amazing :) [06:18] lamont: did you try to build on hppa? [06:18] it might need some config tuning [06:19] fabbione: not yet, but it should be happy [06:19] hmm i am not sure about OCFS2 [06:19] the hppa buildds have been busy.... [06:20] you want a test build? [06:20] is not a big deal [06:20] well if you are happy with a possible failure, it's ok with me :) [06:20] OCFS2 did fail on ia64 [06:20] up to you [06:20] this week I'm not overly concerned with a failure, although it would be nice if it just worked,. [06:21] how long does it take to build? [06:21] linux-source-2.6.12: 03:07:47 (12 entries, sigma 01:42:01) [06:22] or better.. will you stay awake long enough to see the results? [06:22] but gcc-4.0 is ahead of it.. :-( [06:22] hmm ok [06:23] last build was just under 7 hours... we're 4 hours into that [06:23] so it's more one of I'll be awake to see the results... [06:23] just upload without worrying about hppa - if it dies, we'll fix it in the next round [06:23] i guess you won't last 6 hours :) [06:23] ok [06:23] I plan to be awake again in about 7 hours. [06:23] also because i have another set of patches to push pretty soon in .12 [06:23] so, no. no plans to last 6 hours. [06:24] eheheh [06:24] ok [06:24] it occurs to me that doing a test build in hoary probably doesn't really cut it. [06:24] hence the gcc-4.0 is first [06:25] well i am pretty sure i did catch all the dpkg-ar fuckage === lamont does a mass-giveback on debian/{hppa,ia64}, just because he can [06:25] well yeah we can upload 2 kernels in a raw :) [06:25] hahha [06:25] 149 packages [06:25] i will soon need to do that on sparc here [06:26] many of which have been superseded, I expect [06:26] i am letting the buildd to do as much as it can out of main [06:26] before giving back half of it [06:26] I abused my local mirror into having source bolted on the side of the archive, so that I could build all of the cxxlibs, modulo missing build-deps. [06:27] I'd like to get main/cxxapps building this week, will deal with getting a real buildd running sometime after June 1. :-) [06:27] May 23 22:27:32 buildd: breezy: total 251 packages to build. :-( [06:27] ahha only 251? [06:28] i have like 2250 in the queue :) [06:28] Total 52 package(s) in state Building. [06:28] Total 44 package(s) in state Dep-Wait. [06:28] Total 11 package(s) in state Failed. [06:28] Total 795 package(s) in state Installed. [06:28] Total 251 package(s) in state Needs-Build. [06:28] Total 12 package(s) in state Uploaded. [06:28] Total 1165 package(s) [06:28] it's not a full mirror [06:28] oh that's only main [06:28] i figured that banning a few things from auto build is good [06:28] and doubtlessly has several hoary packages in there - I'm going to need to fetch the w-b output from p.u.c sometime and see what the actual archive is missing, and force that through [06:29] and i just build them manually in parallel [06:29] that's most of main (minus kde), and all of cxxlibs.txt's packages [06:29] that's not too bad === lamont has 2 machines chunking along at it... one that runs at about 20-25% of the other. [06:30] several of the failed are 'needs hppa love' [06:30] isn't T-bone helping with hppa at all? [06:30] haven't seen t-bone recently [06:30] neither did i [06:30] i think i have seen only a few packages that needs real sparc love [06:32] for a kernel debug image you would have to fuck around with kernel-package wouldnt you? [06:33] zul: why? [06:33] i was just looking at the roadmap agaim [06:37] i was thinking of having a dbg directory in debian/config and something like linux-image-2.6.12-x-dbg-686 or whatever [06:38] zul: we will do something like config/i386/686-dbg or 386-dbg [06:38] that works.. [06:38] it will be a normal flavour with all the possible debugging options turned on [06:39] i was going through the debug options and have a list kind of [06:39] we might want to consider to add some kgdb patch to it, or whatever [06:39] yeah [06:45] lol....i like instruction number 2 http://www.ready.gov/nuclear_visual.html [06:47] Service Unavailable [06:48] stupid americans [06:48] except lamont [06:48] for nuclear blast... [06:48] instruction #2...consider if you can get out of the area [06:49] yea, stupid americans [06:49] and everyone else... [06:49] heh [06:49] oh crap im in soo much trouble now :) [06:49] they write those for the lowest common denominator [06:50] i know but its funny [06:50] we still chuckle about the instructions on a catch-allive mouse trap we bought a while back... [06:50] disposal instructions began "go to a place where mice are needed" [06:50] ahahahha [06:51] lamont: why they'd want to preserve the health of the lowest common denominator, instead of letting nature take its course... [06:51] yeah [06:51] dilinger: they want to give him enough confusion to have him outside debating when the wave hits... who said anything about preserving him [06:52] http://zulinux.homelinux.net/readygov_nuclear.pdf [06:53] well. sleepybye time [06:53] night lamont [06:53] that stuff is sick [06:54] a nuclear explosion done at ground zero can mechanically destroy everything in a few miles radius [06:54] without taking into account the heating generated by the explosion that would melt down basically everything [06:54] the fallout will do the rest [06:56] this without taking into account the other 2 options of making an underground explosion (more mechanical destruction, less fallout) or in air detonation (less mechanical - much more fallout) [06:56] of course you have this all planned out dont you :) [06:57] no, but the picture gives the feeling that walking a few yards away from the explosion is enough :) [06:57] if you are a cockroach [06:58] also.. everybody had a nuclear radiation shield handy :) [06:58] heh [06:58] i always have one in my wallet [06:58] just in case :) [06:58] dont leave home without it :) [06:58] DON'T PUT CONDOM IN THE WALLET! they get ruined by the temperature! [06:59] PUT YOUR NUCLEAR RADIATION SHIELD INSTEAD! [06:59] heh [06:59] it's done in pure anti scratch metal [06:59] my nuclear condom for those special occasions [07:00] i found this super hot girl.. almost radioactive... [07:01] heh...i think im going to bed....night dude...i expect kernel built by the time i get up :) [07:04] good night [07:05] i am going to upload 1.2 today [07:05] pl [07:05] we need OCFS2 out for testing [07:05] ok even === JaneW [~JaneW@wbs-146-146-76.telkomadsl.co.za] has joined #ubuntu-kernel [08:37] hey JaneW [08:37] JaneW: i was just waiting for you :) [08:37] we need a release name :) [08:39] hi fabbione [08:39] cool [08:39] brb [08:39] ok [08:40] fabbione: Merry Macadamia [08:40] ? [08:47] works for me :) [08:47] done :) === fabbione does some baz dance [09:33] fabbione: Does this one have the acpi crack, or is that next time around? [09:34] next round... [09:34] i am already merging them === ..[topic/#ubuntu-kernel:fabbione] : Ubuntu kernel development discussion | http://www.ubuntulinux.org/wiki/KernelTeam | There are no kernel bugs.. only broken hardware | http://people.u.c/~lamont/Archives/kernel-team@ubuntu.com--2005/ playground: kernel-debian--pre1,3--2.6.11.92 [09:35] mjg59: i had to take this kernel out ASAP [09:35] because i need tests on OCFS2 [09:35] Sure, that's no problem [09:35] and the new version of GFS [09:36] me .. must .. find .. time .. package .. userland [09:36] Hahaha === chmj [~d3vic3@dumbledore.hbd.com] has joined #ubuntu-kernel [09:38] the GFS userland is a royal mess [09:38] the main issue is that there is part of it already packaged [09:38] but kernel and userland need to be in sync [09:38] that makes the other set of packages useless [09:40] HOLY JESUS! [09:41] the acpi patch is HUGE [09:41] Yes [09:41] when you created the diffs, where they on top of what patch set? [09:42] The version in the archive rather than the baz one, I think [09:42] ok, did you apply them as last? [09:42] or first? [09:42] Yes [09:42] Last [09:42] ok [09:43] they both come from the bk acpi tree.. so i can mark them as "external" [09:43] The second one isn't from the bk acpi tree yet [09:43] It's from the mailing list [09:44] ah ok [09:44] well i can still mark it as external [09:44] I need to hack on it a bit more and submit it for that (need to get the _GTF method supported) [09:44] hmm ok [09:44] It makes several more machines work at the moment [09:45] and how many are going to break? [09:45] Haha [09:45] It hasn't broken any of the ones I have access to [09:45] i start to wonder how many laptops do you have... [09:45] 6 with Ubuntu [09:45] my best guess is that you have one even in front of the toilet seat [09:46] morning [09:46] hi chmj [09:46] And another one running the ubuntu kernel [09:46] thats a lot of laptops mjg59 [09:46] chmj: Yup [09:46] I tend to collect them... [09:47] chmj: he's going to build a house of out them when he has enough. [09:47] Mithrandir: I'd love to see that === mjg59 is expecting rather a lot more in the next month or so... [09:48] ehhe [09:48] Mithrandir: ipw2200 1.0.4 is in [09:48] i just uploaded the kernel [09:48] chmj: either that, or he needs a new house for all the laptops. [09:48] fabbione: rock on. [09:48] fabbione: I need to feed you SATA suspend/resume support at some point [09:48] it will take a bit before it's on the mirror [09:48] fabbione: universe for now? [09:48] mjg59: do you have any x40's ? [09:48] Mithrandir: Odd you should say that... [09:48] chmj: Only the one. I'm keeping it. [09:48] mjg59: ok [09:48] Mithrandir: yes [09:49] fabbione: Currently SATA machines will entirely fail [09:49] Mithrandir: the next one will go to main [09:49] fabbione: *bounce* [09:49] :) [09:49] mjg59: SATA is a mess [09:49] mjg59: aah man, I trying to save for one of those, wish I had one :/ [09:49] Mithrandir: i had no time to give the kernel some d-i love [09:49] given that we can push to main and start swithing to it [10:04] I heard that mjg59 was joing to mail me a T42, so I don't have to order one from IBM^WLenovo [10:04] s/joing/going/ [10:04] So, uhh... [10:04] mjg59 : Thanks! [10:05] mjg59: bah 1.2 is FTBFS because dPATCH sucks [10:05] i might as well get your patches in [10:06] now i understand why it was asking about some options I was sure i did pre-configure [10:06] fabbione: what was the decision, if any, on dpatch's removal ? [10:08] chmj: yeah hopefully soon [10:16] fabbione: cdbs and I are mortal enemies though [10:19] chmj: i think i am going to add some sanity checks at build time [10:19] like lsdiff -H * |grep debian [10:30] mjg59: your crack is FTBFS on ia64 [10:31] (at least..) [10:31] the others are still building [10:31] fabbione: Oh, cock. Hang on. [10:32] mjg59: [10:32] fabbione: In drivers/char/agp/hp-agp.c? [10:32] In file included from drivers/firmware/pcdp.c:18: [10:32] drivers/firmware/pcdp.h:48: error: field `addr' has incomplete type [10:32] drivers/firmware/pcdp.c: In function `setup_serial_console': [10:32] drivers/firmware/pcdp.c:27: error: `ACPI_ADR_SPACE_SYSTEM_MEMORY' undeclared (first use in this function) [10:32] drivers/firmware/pcdp.c:27: error: (Each undeclared identifier is reported only once [10:32] drivers/firmware/pcdp.c:27: error: for each function it appears in.) [10:32] Waah [10:32] mjg59: ok.. no panic.. i will upload 1.3 without the patches [10:32] Damnit. Ok, I'll take a look [10:32] so we can workout them properly [10:32] well if it's not too much work i can wait [10:33] the others need to build anyway === mjg59 had tested x86 [10:33] yeah i am building amd64 and ppc too [10:33] it's probably a missing include or something [10:33] it should be simple to fix [10:33] Yeah, it's a missing acpi include [10:34] Goddamnit. You'd think Intel would actually test on their own hardware. [10:34] struct acpi_generic_address addr; [10:34] include/acpi/actypes.h:#define ACPI_ADR_SPACE_SYSTEM_MEMORY (acpi_adr_space_type) 0 [10:34] let see if that helps :) [10:36] In file included from drivers/firmware/pcdp.c:15: [10:36] include/acpi/actypes.h:95:2: #error ACPI_MACHINE_WIDTH not defined [10:36] include/acpi/actypes.h:203:2: #error unknown ACPI_MACHINE_WIDTH [10:36] no it doesn't [10:38] fabbione: Try just including acpi.h [10:38] Uh, acpi/acpi.h [10:38] #include [10:38] including.... [10:38] Ooh, handy, Jeff's just posted a new libata suspend/resume [10:39] Ok, yeah, linux/acpi.h should do [10:39] oh there [10:39] it builded [10:39] Hurrah [10:39] It's possible that there'll be another couple of them [10:39] hmmm [10:39] the point is.. why did it break on ia64 and not on the others? [10:39] will this fix break other stuff? [10:39] That's an IA64-only driver [10:39] ah ok [10:40] It's likely that they've fixed all x86 stuff [10:41] no wonder :) [10:41] fabbione: And PPC shouldn't have been touched [10:41] (with luck) [10:42] mjg59: it's building :) [10:42] other than the usual random ppc sigkills.... [10:43] Heh [10:44] fabbione: why does breezy's make-kpkg tell me the just unpacked 2.6.11-rc4 tree is not a linux toplevel directory ? [10:45] svenl: i have no idea.. works here [10:47] it tells me a warning about the cramfs patch. [10:47] fabbione: on pristine upstream 2.6.11-rc4 ? [10:47] svenl: no idea. i don't even have 11rc4 here [10:48] err, 2.6.12-rc4 obviously. [10:48] fabbione: should i fill a bug report ? [10:48] svenl: i am not even sure what you are trying to do.. [10:48] can you at least [10:48] 1) show me what you are trying to do [10:48] 2) how [10:48] 3) the full error [10:49] svenl: 2.6.*11*-rc4? [10:50] Oh, soryy [10:50] fabbione: i am trying to compile a kernel without cpufreq to track the sleep bug i and benh have been seeing with the latest ubuntu kernel in breezy. [10:50] fabbione: the bug is : when waking from sleep, the box shuts down. [10:50] sven@tael:~/kernel/linux-2.6.12-rc4$ make-kpkg --revision 1 --append-to-version -sven kernel-image [10:51] You should invoke this command from the top level directory of [10:51] a linux kernel source directory tree, and as far as I can tell, [10:51] the current directory: [10:51] /home/sven/kernel/linux-2.6.12-rc4 [10:51] is not a top level linux kernel source directory. [10:51] (If I am wrong then kernel-packages and the linux kernel [10:51] are so out sync that you'd better get the latest versions [10:51] of the kernel-package package and the Linux sources) [10:51] Please change directory to wherever linux kernel sources [10:51] reside and try again. [10:51] [Bsven@tael:~/kernel/linux-2.6.12-rc4$ ls [10:51] arch CREDITS Documentation fs init kernel MAINTAINERS mm README scripts sound [10:51] COPYING crypto drivers include ipc lib Makefile net REPORTING-BUGS security usr [10:51] There. [10:52] hold on a sec [10:53] # See if we are running in a linux kernel directory [10:53] if ((!(-d "drivers" && -d "kernel" && -d "fs" && -d "include/linux")) [10:53] && (!(-d "dev" && -d "kern" && -d "fs" && -d "i386/include"))){ [10:53] my @other_targets = grep (! m/^modules/, @ARGV); [10:53] if ($#other_targets != -1 || ! -d "include/linux") { [10:53] i just untarred the 2.6.12-rc4 tarball from ftp.kernel.org, copied the default ubuntu .config, did make oldconfig, disabled cpufreq. [10:53] this is the check [10:54] fabbione: can you test it on your box or something ? [10:54] fabbione: what is the : && (!(-d "dev" && -d "kern" && -d "fs" && -d "i386/include"))){ [10:54] for ? [10:54] it's like bash [10:55] if [ -d foo ] [10:55] ok. [10:55] but the first group is indeed kernel sources. [10:55] the second line is strange. [10:55] svenl: i am checking.. give me a sec. [10:56] svenl: works here [10:56] That check has been there for ages, and works fine. [10:56] what version of the kernel package do yo heve? [10:57] strange. [10:58] whatever was in breezy yesterday. [10:58] 8.132ubuntu1 [10:58] get ubuntu2 [10:58] i uploaded yesterday [10:59] and please svenl, be always sure to have the latest stuff.. always :) [11:00] I don't get the message on sarge. [11:00] svenl: you need to upgrade [11:00] fabbione: one minute. [11:00] fabbione: are you making fun of me for running breezy ? [11:00] fabbione: bah. [11:01] fabbione: sucks. [11:01] No, he wasn't, but he probably should. :) [11:01] I don't run breezy yet, I just break it merrily. [11:01] And I know I'm not alone. :) [11:01] infinity: i just got into that mess because i was trying to build the ppc64 biarch toolchain in the first place, to build ppc64 kernels. [11:01] infinity: i know i need to reinstall. [11:01] infinity: and it is worse, since i run beta biarch toolchain. [11:02] fabbione : Oh, the amd64 machine is up and running, BTW, should you need a testing bitch. [11:02] so, now apt-get is unhappy because i have libgcc 3.4.3-13ubuntu1 :/ [11:02] infinity: oh yeah... how many boxes do you have around? [11:03] fabbione: mmm, do i really need a separate cramfs patch, or did it make it upstream. [11:03] svenl: the cramfs patch has been there forever. [11:03] fabbione : Locally, only two. The amd64 desktop from hell, and the i386 craptop. [11:03] fabbione : But I do most of my work on a remote PPC machine. [11:03] infinity: hmmmmm [11:04] (And I can get a button bitch to reboot that one if I break it remotely, should that be necessary) [11:04] infinity: you could play a bit with OCFS2 [11:04] even in a single node cluster [11:04] fabbione: in that case, kernel-package should be patched to know about it maybe :) [11:04] http://udu.wiki.ubuntu.com/ClusterFilesystems [11:04] svenl: dude... do you know what cramfs is used for? [11:04] fabbione: sure. [11:04] and in any case you can explain that to Manoj [11:04] fabbione: to read the initrd :) [11:04] since he is upstream [11:05] and the ubuntu patch is there only to deal with the new dpkg that is NOT in Debian yet [11:05] all the other stuff -> Manoj [11:06] mjg59: In file included from drivers/char/agp/hp-agp.c:18: [11:06] fails too [11:06] like you predicted [11:07] fabbione: ok. [11:07] BTW, do you have a packaged git already ? === fabbione shakes Master ACPI mjg59 [11:08] svenl: it's in Debian and Ubuntu already [11:08] cogito [11:09] mjg59: ?????? come on dude... [11:09] WAKE UP [11:09] fabbione: oh. ok. [11:09] thanks. [11:13] include/asm/acpi-ext.h:15: error: parse error before "hp_acpi_csr_space" [11:13] mjg59: i assume that acpi-ext.h is totally borked [11:13] fabbione: acpi-ext.h should include acpi/actypes.h [11:14] yeah it seems to build === fabbione updates the patch again [11:15] Cool [11:15] no [11:15] that's not enough [11:15] Bah [11:16] Same failure? [11:16] no much smaller [11:16] probably it wants acpi/acpi.h [11:16] Yeah [11:16] Weird, the fix in -mm is just actypes.h [11:16] for the asm... [11:16] but the driver still fails to build [11:16] Hngk. [11:16] What's the error? [11:17] CC [M] drivers/char/agp/hp-agp.o [11:17] yeah with acpi/acpi.h goes on [11:17] Ok, cool [11:17] do you still want the error? [11:17] Nah, if it works then that'll do [11:51] mjg59: amd64 is go [11:56] fabbione: Excellent === JaneW [~JaneW@wbs-146-146-76.telkomadsl.co.za] has left #ubuntu-kernel ["Leaving"] === Seveaz [~seveas@seveas.demon.nl] has joined #ubuntu-kernel === doko_ [~doko___@dsl-084-059-053-209.arcor-ip.net] has joined #ubuntu-kernel [02:24] mjg59: kernel uploaded === fabbione does another baz dance === chmj laffs [02:29] lamont: 1.3 is on the way to the buildd's [02:29] you might want to wait to build hppa [02:29] i am tagging it now [02:30] 1.3? what happened to 1.2? [02:30]