[00:00:16] *** kevers has quit IRC
[00:00:29] <davidjames> ferringb: Nope, the builder that worked was using portage-2.1.10.11, not the -r1 version
[00:00:32] <vapier> did we enable -ffunction-sections somewhere recently ?
[00:00:40] <raymes> nope
[00:00:51] <raymes> well, not globally
[00:01:42] *** darinski__ has quit IRC
[00:01:46] <Keybuk> has tegra2-kaen-canary got the updated portage?
[00:01:53] * ferringb sighs
[00:01:57] <Keybuk> no
[00:02:03] <ferringb> preparing a patch to pull it
[00:02:05] <Keybuk> Portage 2.1.10.11 (!/usr/local/portage/chromiumos/profiles/default/linux/amd64/10.0, gcc-4.4.6, glibc-2.11.1-r2, 2.6.32-33-generic x86_64)
[00:02:14] <Keybuk> ^ kaen-canary
[00:02:15] <Keybuk> compared to:
[00:02:19] *** rmcgrath has joined #chromium-os
[00:02:19] *** ChanServ sets mode: +v rmcgrath
[00:02:20] <Keybuk> Portage 2.1.10.11-r1 (!/usr/local/portage/chromiumos/profiles/default/linux/amd64/10.0, gcc-4.4.6, glibc-2.11.1-r2, 2.6.32-25-generic x86_64)
[00:02:24] <Keybuk> ^ failing arm-generic/tec.
[00:03:40] *** saintlou has joined #chromium-os
[00:04:25] *** saintlou has quit IRC
[00:04:32] *** saintlou has joined #chromium-os
[00:08:50] <ferringb> vapier: anything?
[00:09:22] <vapier> what was the previous portage ?  sys-apps/portage-2.1.10.11 ?  i'm getting split .text with that one too
[00:09:36] <Keybuk> vapier: how are you getting the split text?
[00:09:47] <Keybuk> I'm not getting that my end with either
[00:09:47] <ferringb> vapier: 2.1.10.11, yes
[00:09:53] *** behdad has joined #chromium-os
[00:09:54] *** ChanServ sets mode: +v behdad
[00:10:06] <vapier> armv7a-cros-linux-gnueabi-gcc.real (gcc-4.6.0_cos_gg_177914) 4.6.x-google 20110718 (prerelease)
[00:10:36] <ferringb> vapier: relevant changes are the prepstrip imports, 8ebca724748c0c2046e3120b0f5a813fc4298671..86e3ed5b8d1efdd439415cb3dec2905e5de058dc, approximately
[00:10:43] <Keybuk> vapier: ah, my gcc is older
[00:11:01] <vapier> ferringb: python.o is different before prepstrip gets a chance to run
[00:11:05] <raymes> what's your gcc?
[00:11:14] <Keybuk> 53174
[00:11:18] <vapier> cross-armv7a-cros-linux-gnueabi/gcc-4.6.0-r3
[00:11:34] <Keybuk> 4.4.3-r9
[00:11:48] <vapier> running `armv7a-cros-linux-gnueabi-gcc` by hand produces a python.o with .text.startup too
[00:11:53] <raymes> hmm...i don't know why your gcc hasn't been updated
[00:11:53] <ferringb> vapier: sweet
[00:12:02] <Keybuk> raymes: the failing buildds have 4.4.6
[00:12:11] <raymes> you mean 4.6?
[00:12:30] *** akrpic77 has quit IRC
[00:12:44] <Keybuk> Portage 2.1.10.11-r1 (!/usr/local/portage/chromiumos/profiles/default/linux/amd64/10.0, gcc-4.4.6, glibc-2.11.1-r2, 2.6.32-25-generic x86_64)
[00:12:49] <Keybuk> no, 4.4.6
[00:12:57] <rmcgrath> vapier: do you have whatever unstripped binary eu-strip mangles that you can hand me so i can debug elfutils?
[00:13:41] <vapier> rmcgrath: eu-strip isn't being used here that i can see
[00:13:49] <vapier> i haven't committed any changes to enable it
[00:13:59] <vapier> prepstrip is still running bfd strip
[00:14:11] <vapier> but in this case, i can compile python.c into python.o and dump_syms dies on it
[00:14:20] *** BThompson has quit IRC
[00:14:47] <vapier> really no idea what "dump_syms" does ...
[00:15:01] <raymes> yeh, i also tried with FEATURES=nostrip and i see the same
[00:15:29] <Keybuk> I'd like to get the portage change reverted asap
[00:15:44] <Keybuk> on the basis that buildds are failing once that lands, and those that haven't updated haven't
[00:15:54] <Keybuk> and then we can debug further with a green tree
[00:15:56] <rmcgrath> vapier: oh, puneet poked me to say that the eu-strip changes were what had broken things
[00:15:57] <Keybuk> rather than keeping the tree red
[00:16:00] *** akrpic77 has joined #chromium-os
[00:16:12] <Keybuk> ferringb: do you have a revert patch?
[00:16:31] <vapier> rmcgrath: prepstrip landed to support eu-strip if enabled, but it's not enabled
[00:17:01] <raymes> yeh - i'm taking a look at the state of a bot now, but if it's not that change i'm stumped
[00:17:20] <ferringb> katier: for prepstrip bits, but note the FEATURES=nostrip and seeing the same
[00:17:26] <ferringb> s/katier/keybuk/
[00:17:32] <ferringb> nothing to revert, going by this
[00:17:42] <Keybuk> ferringb: I'd like to try the revert anyway, please
[00:17:54] <davidjames> Gerrit has a handy revert button, does it work in this case?
[00:17:56] <raymes> revert could cause problems?
[00:18:10] <Keybuk> raymes: the tree is read and every arm build is failing, we already have problems
[00:18:16] <vapier> raymes: `armv7a-cros-linux-gnueabi-gcc -S -o - -x c - <<<"main(){}" | grep text`
[00:18:20] *** akrpic77 has quit IRC
[00:18:24] <raymes> yeh, but no point reverting if wont fix it
[00:18:34] <vapier> raymes: `armv7a-cros-linux-gnueabi-gcc -S -o - -x c - <<<"main(){}" -O2 | grep text`
[00:18:38] <Keybuk> raymes: reverting will tell us whether this might fix it
[00:18:46] <Keybuk> and it's one less variable out of the tree
[00:18:46] <vapier> why is gcc changing it from .text to .section .text.startup ?
[00:18:58] <rmcgrath> vapier: ah, ok.  well, i'll go back to claiming that eu-strip is perfect, then. ;-)
[00:19:11] <vapier> rmcgrath: we've got no data to indicate otherwise ;)
[00:19:22] <ferringb> http://gerrit.chromium.org/gerrit/#change,10445 <-- reversion
[00:19:47] <ferringb> ...with a lovely msg on my part, sigh.
[00:19:58] <davidjames> ferringb: +2, although commit msg could be better :)
[00:20:20] <ferringb> intention being to just shove it in and see if things work, or to actually do a temp pull in of it
[00:20:47] <davidjames> ferringb: Looks like it's already pushed, see http://gerrit.chromium.org/gerrit/#change,10446 :)
[00:21:00] *** elly has quit IRC
[00:21:32] <ferringb> yeah, just noticed
[00:21:54] <vapier> hmm seems to be a gcc-4.6.x optimization ... mainline does it too
[00:22:04] <Keybuk> tegra2-seaboard-full just failed oto
[00:22:21] <Keybuk> same failure
[00:22:33] <Keybuk> Portage 2.1.10.11-r1 (!/usr/local/portage/chromiumos/profiles/default/linux/amd64/10.0, gcc-4.4.6, glibc-2.11.1-r2, 2.6.32-25-generic x86_64)
[00:22:42] <davidjames> Keybuk: Just kicked 3 arm builders, they should downgrade to old portage automatically
[00:23:00] <davidjames> Keybuk: Current arm canaries are doomed to fail though, might want to cycle those before they fail
[00:23:17] *** cros_ has quit IRC
[00:23:24] <Keybuk> and previous successful build was previous portage
[00:24:07] <raymes> yeh
[00:24:18] <raymes> ok, i agree that it must be somehow related to the portage upgrade
[00:24:33] *** jennb has quit IRC
[00:24:41] <raymes> either directly or indirectly
[00:25:43] <crosbot> tree became 'Tree is closed (caused by http://gerrit.chromium.org/gerrit/#change,10255; revert in, waiting for portage downgrade on affected builders)'
[00:28:55] *** behdad has quit IRC
[00:31:22] *** kevers has joined #chromium-os
[00:31:23] *** ChanServ sets mode: +v kevers
[00:31:28] *** rbyers has quit IRC
[00:34:41] *** rbyers has joined #chromium-os
[00:34:41] *** ChanServ sets mode: +v rbyers
[00:34:54] <raymes> ok
[00:35:09] <vapier> this dump_syms tool makes me cry
[00:35:21] <raymes> i think the portage change is somehow causing generate_breakpad_symbols to run dump_syms on the python.o when it didn't before
[00:35:34] <raymes> dump_syms on the python.o always failed
[00:36:00] <ferringb> different automake-wrappers are in use
[00:36:03] <raymes> but i just ran generate_breakpad_symbols in my older chroot and it didn't run on that file
[00:36:09] <ferringb> v4 vs v5
[00:36:30] <vapier> v5 was in use for x86/amd64, but not arm ... that was v4
[00:36:35] <vapier> but python doesn't use automake
[00:36:40] <ferringb> yeah
[00:36:43] <ferringb> just going through the delta atm
[00:37:00] <vapier> the automake-wrapper change only affected arm
[00:38:38] *** behdad has joined #chromium-os
[00:38:39] *** ChanServ sets mode: +v behdad
[00:39:26] *** posciak has quit IRC
[00:39:49] *** behdad has quit IRC
[00:39:57] *** behdad has joined #chromium-os
[00:39:58] *** ChanServ sets mode: +v behdad
[00:40:27] *** lipsinV2 has joined #chromium-os
[00:40:33] <vapier> so all dump_syms does for Linux is split out the debug info ?
[00:41:07] <raymes> i think it does it in a special format
[00:41:12] <raymes> minidump?
[00:41:26] <raymes> find "/build/tegra2_kaen/usr/lib/debug/" -name \*.debug | grep "python\.o"
[00:41:58] <vapier> seems like we already go through the effort in portage to handle split debug files ... would be nice if dump_syms didn't duplicate that ...
[00:42:03] <raymes> cros_generate_breakpad_syms is running over everything in the /usr/lib/debug directory afaict
[00:42:32] <vapier> oh really
[00:42:35] <raymes> yeh
[00:42:46] <raymes> check your /build/tegra2.../usr/lib/debug
[00:42:50] <raymes> see what's there
[00:45:51] <vapier> ok, now it makes sense
[00:45:53] <vapier> how about this:
[00:46:40] *** rbyers has quit IRC
[00:46:47] <vapier> dump_syms has always been broken when the ELF lacks .text.  new portage lands a new prepstrip which is more proactive at splitting out .debug files (from .o where it didn't before).  new gcc-4.6 automatically puts things into .text.startup at higher optimization levels where older versions always did .text.
[00:47:27] <vapier> so yes, new portage CL is the trigger that makes it all fall apart, but ultimately dump_syms is to blame
[00:47:31] <vapier> sound reasonable to people ?
[00:47:59] * ferringb nods dumbly
[00:48:52] <ferringb> vapier: question is, mangle dump_syms, or punt the prepstrip bits
[00:48:57] <ferringb> ...then mangle dump_syms
[00:49:33] *** behdad has quit IRC
[00:49:55] <vapier> lemme look at prepstrip ... might be a one line comment
[00:50:06] <ferringb> rather just pull it tbh
[00:50:14] <ferringb> got enough patches sitting in there as is
[00:50:25] <vapier> i want it :p
[00:50:28] *** rbyers has joined #chromium-os
[00:50:29] *** ChanServ sets mode: +v rbyers
[00:50:56] <ferringb> you've got a bit... but I do need to get that version rolled out
[00:51:30] <raymes> why do we need debug symbols for .o files?
[00:51:48] <raymes> (btw, agree with your analysis vapier)
[00:52:12] <raymes> but it seems like the simple solution is to nuke those
[00:52:24] <ferringb> don't, realistically.  not unless gcc/the manager can integrate it back in
[00:52:34] <vapier> ferringb: i think you want this
[00:52:36] <vapier> +               : process_elf "${x}" ${SAFE_STRIP_FLAGS}
[00:52:41] <vapier> two byte change!
[00:52:56] <davidjames> vapier: And what does that change? :)
[00:53:05] <davidjames> Looks like it's ignoring errors :)
[00:53:19] <vapier> think of it as a comment char
[00:53:43] <ferringb> vapier: that would gut stripping, no?
[00:53:52] <ferringb> (depends on which of the process_elf invocations you're targetting also)
[00:53:57] <vapier> ferringb: for .o files, yes
[00:54:47] <ferringb> according to the comments, it already ignores .o
[00:55:09] <ferringb> and as said, I've no idea which process_elf invocation you're targetting here; executable/shared object, or relocatable
[00:56:25] <vapier> ferringb: http://wh0rd.org/strip.patch
[00:56:34] <vapier> (i'm testing that now)
[00:57:37] <vapier> ferringb: technically comment is outdated
[00:57:55] <vapier> i like having split debug for .o files, but i guess it just makes tools angry
[00:59:20] <ferringb> vapier: unless I'm missing something, only real usage for .o is to integrate it into another final object
[00:59:29] <ferringb> gcc smart enough to grab the debug symbols if they're split out for that case?
[00:59:48] <vapier> ferringb: consider crt*.o from glibc which has the ELF entry point
[00:59:56] <vapier> it gets installed stripped normally w/out debug info
[01:00:02] <ferringb> aware of the file, but not the specifics of how it's used
[01:00:12] <vapier> linking it into the final ELF, you get debug info for everything else but the crt*.o
[01:00:20] <vapier> with my change, you get debug info for that too
[01:00:34] <vapier> gdb doesn't mind (and i don't), but i've seen people get confused in the past
[01:01:07] * ferringb will take your word for it
[01:01:18] <ferringb> mostly in a "fix it fix it fix it fix it" mentality atm ;)
[01:01:22] *** behdad has joined #chromium-os
[01:01:22] *** ChanServ sets mode: +v behdad
[01:01:26] <ferringb> ...fix it fix fix it.  etc. ;)
[01:01:28] <vapier> i've seen gdb tests fail too because they don't expect the first insn to have source/line info
[01:01:33] <vapier> but i like single stepping from ELF entry
[01:01:35] <vapier> maybe i'm odd
[01:02:01] <vapier> take that patch i posted above ... prepstrip will no longer post debug info for .o files
[01:02:21] <vapier> i'm not sure if this is something i want to re-disable ... i'm inclined to tackle broken packages now that i know how the stuff works
[01:02:29] <vapier> (re-disable upstream that is)
[01:03:19] <ferringb> vapier: can you think of any other dump_syms style uses for upstream gentoo?
[01:03:28] <ferringb> mainly thinking about the fact those patches are live in the tree
[01:03:30] <vapier> i'll file a bug for dump_syms breaking on -ffunction-sections
[01:03:47] <vapier> ferringb: i don't understand the question
[01:04:33] <ferringb> nvm, just being paranoid per the norm
[01:07:29] <ferringb> http://gerrit.chromium.org/gerrit/#change,10451
[01:07:53] <vapier> http://crosbug.com/21932
[01:09:09] <ferringb> updated w/ the issue #
[01:10:44] <ferringb> vapier: presume that patch will be pulled once dump_syms is sorted...?
[01:11:10] <vapier> ferringb: that's the plan unless something in upstream Gentoo indicates i want to abandon this completely
[01:11:17] <vapier> but i'm hoping that isn't the case
[01:12:11] <ferringb> 'k.  going to skip bulk and just patch the bugger in.
[01:12:49] <vapier> i gotta jet or smriti is going to hit me
[01:14:37] <ferringb> vapier: that portage version isn't going to reland till tomorrow
[01:14:43] * ferringb isn't going to risk explosion at 8pm PST
[01:16:02] *** rbyers has quit IRC
[01:23:16] *** Sergiu has quit IRC
[01:28:04] *** rbyers has joined #chromium-os
[01:28:04] *** ChanServ sets mode: +v rbyers
[01:28:22] *** gauravsh has quit IRC
[01:39:06] *** behdad has quit IRC
[01:45:25] <davidjames> arm-generic-full should report back in ~15 minutes, so we should be able to reopen the tree then if it's successful
[01:50:01] <davidjames> Keybuk: arm-generic-full is past the problematic point, time to open?
[01:54:18] * ferringb sees no reason to leave it closed
[01:55:06] <crosbot> tree became 'Tree is open (arm_generic_full builder has gotten past the fail point)'
[01:56:55] *** darinski_ has quit IRC
[02:01:16] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "arm generic PFQ" from 1c44d0892b1d03a2edc585a9bc2f8d54a289f218: Mandeep Singh Baines <msb at chromium dot org>)'
[02:03:47] <sosa> msb:  missing ~arm for keywords in wayland ebuild?
[02:04:56] <marcheu> yeah
[02:05:23] <davidjames> msb_: Also, I think your ebuild doesn't have a repo :)
[02:05:52] <crosbot> tree became 'Tree is closed (arm-generic-pfq wayland build error)'
[02:05:55] <Keybuk> ERROR: Cannot find commit id for /home/chrome-bot/trunk/src/overlays/overlay-x86-wayland/x11-base/wayland/wayland-9999.ebuild
[02:06:05] <davidjames> keybuk: I think msb forgot to add wayland to the manifest?
[02:06:16] <Keybuk> davidjames: I think he did too
[02:06:34] <Keybuk> I have ping'd ... I'm timing out in 30s :p
[02:07:42] <srao> Keybuk: I think msb, puneetster, anush are discussing it
[02:08:56] <crosbot> tree became 'Tree is open (reverted msb's change)'
[02:14:37] *** cooled_ has quit IRC
[02:16:52] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "tegra2 PFQ" from 1c44d0892b1d03a2edc585a9bc2f8d54a289f218: Mandeep Singh Baines <msb at chromium dot org>)'
[02:24:16] <crosbot> tree became 'Tree is open (msb's change hit another pfq)'
[02:31:18] *** wfrichar has quit IRC
[02:32:25] *** behdad has joined #chromium-os
[02:32:25] *** ChanServ sets mode: +v behdad
[02:41:10] <davidjames> Is Chromeos-chrome prebuilt supposed to be 900MB? tegra2 PFQ is scheduled to take >= 4 hours to download it :(
[02:42:39] <srao> I think the chrome binary with debugging symbols is very large
[02:42:56] *** wfrichar has joined #chromium-os
[02:42:56] *** ChanServ sets mode: +v wfrichar
[02:43:17] <davidjames> Seems faster to build from scratch than download when it's this big!
[02:43:19] <srao>  808M 2011-09-29 18:37 chrome.debug
[02:43:29] <srao> for a arm debug binary
[02:43:37] <srao> yeah sounds like a network problem
[02:44:11] <srao> should be able to download 900MB faster than that....64kb/sec
[02:44:27] *** cooled has joined #chromium-os
[02:46:35] <davidjames> srao: It takes me 50 secs to download same file using gsutil, from the same builder (18MB/s)
[02:47:35] <davidjames> srao: And 3+ hrs using curl
[02:50:17] <ferringb> er
[02:50:52] <ferringb> davidjames: this via emerge, or via the scripts?
[02:51:31] <ferringb> davidjames: alternatively, snag a tcpdump trace of it while you're in the builder
[02:53:48] <davidjames> ferringb: I just ran curl directly and could reproduce
[02:53:58] <davidjames> ferringb: gsutil downloads 18MB/s, curl downloads at 64kb/s
[02:56:03] <ferringb> got a tcpdump trace of it?
[02:56:07] * ferringb suspects user-agent
[02:56:33] <ferringb> which would be rather funny, actually
[02:58:04] <ferringb> davidjames: add a /gsutil # to the useragent for curl
[02:58:06] <ferringb> quick test
[02:58:33] *** behdad has quit IRC
[02:58:37] <srao> that would be amusing... and somewhat unfortunate
[02:59:34] <ferringb> well, frankly... sans some tcp flag or something, boto should be slower
[02:59:37] <ferringb> python implementation vs c
[02:59:56] <ferringb> c should win just on turn around time for responses, as daft as that may sound
[03:00:12] <ferringb> (max speed mind you)
[03:01:22] <ferringb> davidjames: suspect this will be one of the more entertaining speedups you find...
[03:05:38] <srao> well IMHO it would be a foolish and easily defeatable way of trying to penalize external users
[03:05:43] <zbehan> doesn't gsutil do some more crazy stuff like parallel downloads, overriding potential throttling?
[03:06:30] * ferringb points at awget
[03:06:37] <ferringb> possible, going through it's implementation right now
[03:06:39] <zbehan> it's true that for this kind of speed increase, it'd have to be 300 parallel streams
[03:08:05] <davidjames> Normally gsutil is slower than curl, they just changed something on the backend today tnough
[03:09:47] <davidjames> Just downloaded chrome on behalf of the builder to see if it runs faster now :)
[03:20:51] *** yusukes_ has joined #chromium-os
[03:25:08] *** rbyers has quit IRC
[03:26:39] *** cooled_ has joined #chromium-os
[03:27:10] *** cooled has quit IRC
[03:32:09] <Keybuk> davidjames: all tegra2 canaries are going green
[03:37:18] *** jrbarnette has quit IRC
[03:39:54] *** kevers has left #chromium-os
[03:41:14] *** yusukes has quit IRC
[03:51:35] *** wfrichar has quit IRC
[03:54:27] *** sadrul has joined #chromium-os
[03:54:27] *** ChanServ sets mode: +v sadrul
[04:08:13] *** D|sT has joined #chromium-os
[04:21:25] *** elly has joined #chromium-os
[04:32:51] *** Stepan has joined #chromium-os
[04:33:20] <Stepan> re
[04:33:28] *** rbyers has joined #chromium-os
[04:33:28] *** ChanServ sets mode: +v rbyers
[04:34:34] *** rbyers has quit IRC
[04:40:28] *** jimhebert has quit IRC
[04:55:00] *** powderluv has quit IRC
[05:04:40] *** rbyers has joined #chromium-os
[05:04:41] *** ChanServ sets mode: +v rbyers
[05:19:19] *** saintlou has quit IRC
[05:49:36] *** rbyers has quit IRC
[05:53:00] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86-alex-binary" from bb3a446122e0baeab71e0ee1c566620bb6705f14: Hung-Te Lin <hungte at chromium dot org>)'
[06:16:01] *** rbyers has joined #chromium-os
[06:27:32] *** TW1920 has joined #chromium-os
[06:44:01] *** petermayo has quit IRC
[06:44:16] *** petermayo has joined #chromium-os
[06:44:17] *** ChanServ sets mode: +v petermayo
[07:09:01] *** rbyers has quit IRC
[07:34:50] *** saggu has joined #chromium-os
[07:45:18] *** saintlou has joined #chromium-os
[07:46:55] *** greyhat has joined #chromium-os
[07:48:27] *** Malmis_ has quit IRC
[07:49:21] *** saintlou has quit IRC
[07:49:27] *** Malmis has joined #chromium-os
[07:50:16] <crosbot> tree became 'Tree is open (x86-alex-binary -> sig 11)'
[07:52:03] <crosbot> tree became 'Tree is open (x86-alex-binary -> crosbug.com/21629)'
[07:52:56] *** silverroots has joined #chromium-os
[08:29:24] *** GodoPPL has quit IRC
[08:40:36] *** Malmis has quit IRC
[08:40:46] *** Malmis has joined #chromium-os
[08:48:19] *** petermayo has quit IRC
[08:49:11] *** akrpic77 has joined #chromium-os
[09:02:42] *** cooled_ has quit IRC
[09:02:54] *** cooled has joined #chromium-os
[09:20:13] *** akrpic77 has quit IRC
[09:22:41] *** akrpic77 has joined #chromium-os
[10:13:49] *** behdad has joined #chromium-os
[10:17:33] *** patcito has quit IRC
[10:21:34] *** JakeSays has quit IRC
[10:23:04] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot_master" on "TOT Pre-Flight Queue" from 4fa58297637ee1d2b11f1e96868c31d525936a69: Hung-Te Lin <hungte at chromium dot org>)'
[10:26:54] <crosbot> tree became 'Tree is closed (crosbug.com/21800)'
[10:28:26] <crosbot> tree became 'Tree is open (crosbug.com/21800)'
[10:34:06] *** njw` has joined #chromium-os
[10:34:07] *** ChanServ sets mode: +v njw`
[10:34:20] *** njw has quit IRC
[10:38:55] *** greyhat has quit IRC
[10:40:33] *** JakeSays has joined #chromium-os
[10:42:22] *** njw`` has joined #chromium-os
[10:42:24] *** ChanServ sets mode: +v njw``
[10:42:26] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot_master" on "x86 generic PFQ" from 29a149c43719c3e5c67eb1424c15fc194c406cc1: Luigi Semenzato <semenzato at chromium dot org>)'
[10:42:40] *** njw` has quit IRC
[10:44:42] *** greyhat has joined #chromium-os
[10:51:02] *** yusukes_ has quit IRC
[10:57:32] *** saggu has quit IRC
[11:13:55] *** elly has quit IRC
[11:17:16] *** saggu has joined #chromium-os
[11:20:34] *** elly has joined #chromium-os
[11:25:34] *** behdad has quit IRC
[11:30:43] <crosbot> tree became 'Tree is open (crosbug.com/21800)'
[11:34:19] *** falken has quit IRC
[11:40:18] *** vmil86 has joined #chromium-os
[11:40:36] *** njw``` has joined #chromium-os
[11:40:37] *** ChanServ sets mode: +v njw```
[11:40:56] *** njw`` has quit IRC
[11:51:07] *** njw``` has quit IRC
[11:51:12] *** njw``` has joined #chromium-os
[11:51:12] *** ChanServ sets mode: +v njw```
[12:13:16] *** akrpic77_ has joined #chromium-os
[12:13:19] <akrpic77_> hello.
[12:18:41] <akrpic77_> after patching, emerging, recompiling... nvidia-drivers 260.19.36, during boot i can see it erroring out with "unknown symbol __raw_spin_lock_init", any clues?
[12:55:13] *** greyhat has quit IRC
[12:56:29] <saggu> hi... want to configure the rootfs.. can anyone please tell me the way and where can i locate the source code for rootfs..?
[12:57:15] *** wbednarski has joined #chromium-os
[13:00:24] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86-alex-binary" from b7984b9553bbe5c0f9234bbf6f31fcbcac90918a: Sonny Rao <sonnyrao at chromium dot org>)'
[13:08:20] *** kochi has left #chromium-os
[13:15:25] *** mazda has left #chromium-os
[13:24:46] *** jochen___ has joined #chromium-os
[13:24:46] *** ChanServ sets mode: +v jochen___
[13:25:05] *** greyhat has joined #chromium-os
[13:26:02] *** jochen__ has quit IRC
[13:26:05] *** jochen___ is now known as jochen__
[13:27:23] *** akrpic77 has quit IRC
[13:30:48] *** akrpic77 has joined #chromium-os
[13:30:51] *** hashimoto has left #chromium-os
[14:00:15] <saggu>  hi... want to configure the rootfs.. can anyone please tell me the way and where can i locate the source code for rootfs..?
[14:06:18] *** Sergiu has joined #chromium-os
[14:06:19] *** Sergiu has joined #chromium-os
[14:24:11] *** erikat has joined #chromium-os
[14:30:07] *** silverroots has quit IRC
[14:36:33] *** nexusz99 has joined #chromium-os
[14:41:33] *** akrpic77 has quit IRC
[14:42:11] <akrpic77_> where is the connection between board overlay and kernel configuration ?
[14:42:20] *** akrpic77 has joined #chromium-os
[14:47:47] *** ukai has quit IRC
[14:47:49] <akrpic77_> akrpic77_: in make.conf in board overylay directory
[14:48:55] *** katier has quit IRC
[14:49:11] *** yoshiki has joined #chromium-os
[14:49:36] *** katier has joined #chromium-os
[14:49:36] *** ChanServ sets mode: +v katier
[14:58:01] <cwolfe> akrpic77_: think most of it is the CHROMEOS_KERNEL_SPLITCONFIG variable. That gets picked up by kernel/files/chromeos/scripts/prepareconfig, which spits out the appropriate kernel .config
[14:58:57] <akrpic77_> cwolfe: tnx, I already answered to myself.
[14:59:08] <cwolfe> alright :)
[14:59:18] <akrpic77_> hmm.
[15:01:46] <akrpic77_> i got another problem. building x86-fruitloop with latest nvidia (285, same problem on 260), dmesg reports disabling lock debugging due to kernel taint. next line is nvidia: Unknown symbol __raw_spin_lock_init, looks like this symbol is debug only. what's wrong?
[15:02:32] <akrpic77_> could be that during emerge of nvidia-drivers, kernel is configured to debug locks.
[15:05:32] *** saggu has quit IRC
[15:16:10] *** ukai has joined #chromium-os
[15:16:11] *** ChanServ sets mode: +v ukai
[15:19:06] <maruel> orry I blew up gclient for ~45 seconds
[15:19:09] <maruel> sorry*
[15:23:19] *** dkz has joined #chromium-os
[15:24:07] *** rochberg has joined #chromium-os
[15:24:08] *** ChanServ sets mode: +v rochberg
[15:24:20] <rochberg> Investigating
[15:27:32] <dkz> anyone can help me to enter correctly on chroot with cros_sdk? http://paste.pocoo.org/show/496063/ -> I'm getting this error. I tried to mount tmpfs on /dev/shm and /run/shm on /etc/fstab but same issue, not sure if it's the problem. I'm on debian amd64
[15:28:46] *** ]R[ has joined #chromium-os
[15:32:25] *** sadrul has quit IRC
[15:36:41] *** rbyers has joined #chromium-os
[15:41:46] *** dkz has quit IRC
[15:44:08] *** jochen___ has joined #chromium-os
[15:44:08] *** ChanServ sets mode: +v jochen___
[15:44:33] <akrpic77_> emerging kernel.... where does that place it's output ? what's KBUILD_OUT value ?
[15:45:15] *** jochen__ has quit IRC
[15:45:18] *** jochen___ is now known as jochen__
[16:05:57] *** akrpic77_ has quit IRC
[16:16:48] *** cros_ has joined #chromium-os
[16:19:26] *** rbyers has quit IRC
[16:22:13] <ellyjones> crosbot: tree?
[16:22:13] <crosbot> ellyjones: tree: Tree is closed (Automatic: "cbuildbot" on "x86-alex-binary" from b7984b9553bbe5c0f9234bbf6f31fcbcac90918a: Sonny Rao <sonnyrao at chromium dot org>)
[16:22:24] <ellyjones> anyone debugging this?
[16:23:18] <ellyjones> woah, ctest.py sadness or something
[16:23:24] <ellyjones> crosbot: sheriffs?
[16:23:25] <crosbot> ellyjones: sheriffs: jrbarnette, dianders, rochberg
[16:24:51] <ellyjones> davidjames: do you know this?
[16:28:08] <cros_> If one wants to build for an ARM tegra2 target is there anyway to get the required open GL es libraries? I see in the opengles ebuild there is a tegra-local-bins use flag that is set by default but I don't see those binaries anywhere. It seems I can't access the git server to pull it myself either.
[16:28:57] <rochberg> x86 alex binary appears to be failing on crostestutils.generate_test_payloads.payload_generation_exception.PayloadGenerationException: Failed to generate a required update.
[16:29:54] *** rbyers has joined #chromium-os
[16:29:54] *** ChanServ sets mode: +v rbyers
[16:31:57] *** rharrison_chrome has joined #chromium-os
[16:31:57] *** ChanServ sets mode: +v rharrison_chrome
[16:37:31] *** gut4 has joined #chromium-os
[16:42:28] *** kevers has joined #chromium-os
[16:42:28] *** ChanServ sets mode: +v kevers
[16:51:56] *** greyhat has quit IRC
[16:57:56] <ellyjones> rochberg: want to update the tree status?
[16:58:15] <crosbot> tree became 'Tree is closed:  image building failing on at least x86-alex-binary and x86-zgb-binary'
[17:04:01] *** greyhat has joined #chromium-os
[17:05:51] *** wfrichar has joined #chromium-os
[17:05:52] *** ChanServ sets mode: +v wfrichar
[17:10:46] <crosbot> tree became 'Tree is closed:  http://crosbug.com/21945'
[17:15:49] *** gut4 has quit IRC
[17:22:39] *** gut4 has joined #chromium-os
[17:23:31] <ellyjones> rochberg: tree staying closed?
[17:28:25] *** greyhat has quit IRC
[17:28:51] <rochberg> We're not running tests on those platforms, so I'm not going to open
[17:31:05] <ellyjones> ok
[17:36:56] *** petermayo has joined #chromium-os
[17:36:56] *** ChanServ sets mode: +v petermayo
[17:44:09] *** rbyers has quit IRC
[17:44:29] *** rbyers has joined #chromium-os
[17:44:29] *** ChanServ sets mode: +v rbyers
[17:45:16] *** rbyers has quit IRC
[17:47:09] *** gut4 has quit IRC
[17:48:38] *** gut4 has joined #chromium-os
[17:49:37] *** seumas has quit IRC
[17:50:22] *** cros_ has quit IRC
[17:55:24] *** cros_ has joined #chromium-os
[17:56:05] *** GodoPPL has joined #chromium-os
[17:57:07] *** cgreco has joined #chromium-os
[18:04:26] *** seumas has joined #chromium-os
[18:04:26] *** ChanServ sets mode: +v seumas
[18:13:13] *** superm1 has left #chromium-os
[18:20:08] *** jrbarnette has joined #chromium-os
[18:22:12] *** jrbarnette_ has joined #chromium-os
[18:24:30] *** dkz has joined #chromium-os
[18:24:42] *** jrbarnette has quit IRC
[18:24:43] *** jrbarnette_ is now known as jrbarnette
[18:25:45] *** jrbarnette_ has joined #chromium-os
[18:28:13] *** saintlou has joined #chromium-os
[18:28:14] *** ChanServ sets mode: +v saintlou
[18:29:25] *** jrbarnette has quit IRC
[18:29:25] *** jrbarnette_ is now known as jrbarnette
[18:32:21] <jrbarnette> sheriff on duty here
[18:33:03] <jrbarnette> rochberg: tree is closed; any idea whether the problem will repeat if we re-open?
[18:34:21] <dkz> I built and signed a kernel image and I put this image on usb with: dd if=kern.bin of=/dev/sdb4 (I want to boot from USB stick). When I reprioritze the partitions for booting my kernel first I'm getting the error -> cgpt add: GptSanityCheck() returned 2: GPT_ERROR_INVALID_HEADERS. What I'm doing wrong?
[18:40:16] *** jochen___ has joined #chromium-os
[18:40:17] *** ChanServ sets mode: +v jochen___
[18:41:02] <jrbarnette> OK.  looks like the failure *has* repeated at least once on alex-binary
[18:41:17] *** jochen__ has quit IRC
[18:41:22] *** jochen___ is now known as jochen__
[18:42:25] *** BThompson has joined #chromium-os
[18:42:29] *** ChanServ sets mode: +v BThompson
[18:42:54] *** rharrison_chrome has quit IRC
[18:46:43] *** seumas has quit IRC
[18:49:20] *** seumas has joined #chromium-os
[18:49:20] *** ChanServ sets mode: +v seumas
[18:49:34] *** gauravsh has joined #chromium-os
[18:49:34] *** ChanServ sets mode: +v gauravsh
[18:50:52] <jrbarnette> there are failures in other builders that don't match the alex failure
[18:51:34] *** dianders_ has joined #chromium-os
[18:54:47] <dianders_> Hello.  Coming online for sheriffing.  On the bus now.  Trying to come up to speed...
[18:58:30] <davidjames> rochberg: It looks like you diagnosed issue 21945 correctly
[18:59:31] <dianders_> davidjames: Thoughts about what to do for 21945, then?
[18:59:37] <jrbarnette> davidjame:  what's the right handling for the issue?
[18:59:52] <jrbarnette> davidjames: what's the right handling for the issue?
[19:00:04] <davidjames> dianders_, jrbarnette: Is that bug happening consistently?
[19:00:19] <jrbarnette> twice on alex-binary; i haven't seen it in other logs
[19:00:26] <jrbarnette> i haven't searched all logs, though
[19:00:59] <jrbarnette> multiple choice - a) revert something, b) fix something, c) re-open the tree and pray
[19:01:07] <davidjames> jrbarnette: I'd suggest reopening while we look into a fix
[19:01:29] <davidjames> jrbarnette: As long as PFQ is green :)
[19:01:36] <davidjames> jrbarnette: Oops, PFQ is red, can you look at that? :)
[19:01:41] <jrbarnette> PFQ is RED i'm looking now
[19:02:00] <jrbarnette> can't find a @$!^ error message in the logs
[19:02:15] <davidjames> jrbarnette: supplied_chrome sig 11
[19:02:34] <jrbarnette> oh, is *that* it?  i thought those didn't cause failures?
[19:02:34] <dianders_> Yup.
[19:02:47] <davidjames> jrbarnette: That's mkrebs bug, basically if Chrome exits too fast the crash reporter can't get a stack trace
[19:02:48] <jrbarnette> is that a known chrome crash?
[19:03:05] *** rginda_home has quit IRC
[19:03:13] <davidjames> jrbarnette: Well, there *is* a chrome crash, we just don't know what the crash is (that's what 'supplied_chrome' means, it's code for 'couldn't get a stack trace')
[19:03:35] <jrbarnette> is there a bug # for the supplied_chrome failure?
[19:03:42] <dianders_> http://code.google.com/p/chromium-os/issues/detail?id=21629 ?
[19:03:46] <davidjames> It's a flaky failure, but worth bumping priority up of the bug because we've seen a lot of supplied_chrome crashes lately (in my opinion)
[19:03:54] <ellyjones> crosbot: tree?
[19:03:54] <crosbot> ellyjones: tree: Tree is closed:  http://crosbug.com/21945
[19:03:57] <dianders_> ...maybe not...
[19:03:58] <ellyjones> davidjames: should we reopen?
[19:04:16] <dianders_> ...just saw that in last night's report
[19:04:32] *** benchan has joined #chromium-os
[19:04:32] *** ChanServ sets mode: +v benchan
[19:04:58] <davidjames> It's been a long time since PFQ has run, I'm wondering if there is a bug in the buildbot triggering
[19:05:14] <dianders_> davidjames: was wondering about that.  It says "building", tho
[19:05:17] <jrbarnette> ok, i'm all for re-opening; any qualms?
[19:05:50] <dianders_> jrbarnette: OK w/ me to reopen...  I believe that http://crosbug.com/21629 is the PFQ problem.
[19:05:59] <dianders_> ...but I don't know why it stopped running.
[19:06:12] <dianders_> davidjames: you want me to try to force a build on PFQ, or are you going to investigate first?
[19:06:35] *** vapier has quit IRC
[19:06:46] <dianders_> ...oh, or someone could press it for me...
[19:07:02] <davidjames> dianders_: I think there's a different bug for supplied_chrome crashes
[19:07:06] <crosbot> tree became 'investigating http://crosbug.com/21945 and others; submit with care'
[19:07:22] <ellyjones> anyone seen this around? rsync: link_stat "/var/log/monitor-ssh-reboots" failed: No such file or directory (2)
[19:07:30] <dianders_> davidjames: Rong thought it was the same
[19:07:55] <dianders_> ...at least, there's a "supplied_chrome" crash referenced in his logs
[19:07:55] <jrbarnette> davidjames: are you working now on 21945?
[19:08:09] *** SpanKY has joined #chromium-os
[19:08:09] *** ChanServ sets mode: +v SpanKY
[19:08:11] *** vapier has joined #chromium-os
[19:08:14] <ellyjones> jrbarnette: usually 'submit with care' is 'Tree is throttled'
[19:08:37] <jrbarnette> mostly i was trying to be ironic :-/
[19:08:41] <ellyjones> oh :P
[19:08:53] <crosbot> tree became 'Tree is open (investigating http://crosbug.com/21945 and others; submit with care)'
[19:08:54] *** powderluv has joined #chromium-os
[19:08:54] *** ChanServ sets mode: +v powderluv
[19:08:55] <ellyjones> carry on, then :)
[19:09:32] *** lipsinV2 has quit IRC
[19:10:48] <davidjames> jrbarnette: Just assigned that one to sosa@
[19:11:55] *** saintlou has quit IRC
[19:13:38] *** rbyers has joined #chromium-os
[19:13:38] *** ChanServ sets mode: +v rbyers
[19:13:51] <davidjames> jrbarnette: So, interesting, Bug 21629 was *originally* unrelated to the PFQ problem, but the last two comments by snanda / rochberg are clearly related :)
[19:13:58] *** saintlou has joined #chromium-os
[19:13:58] *** ChanServ sets mode: +v saintlou
[19:14:33] <davidjames> jrbarnette: Last time I filed a bug for fixing the supplied_chrome issue, http://code.google.com/p/chromium-os/issues/detail?id=19301, it got closed as a duplicate of Bug 6299
[19:14:39] <dianders_> davidjames, jrbarnette: Yah, those were the two comments I ws looking at
[19:15:11] <dianders_> Searching for supplied_chrome gives me: http://crosbug.com/21559
[19:15:52] <dianders_> Just about to go offline for ~10 minutes (bus -> desk).  I'll try to sort out the supplied_chrome stuff when I get in...
[19:15:54] <davidjames> dianders_: Ah, perfect, that's even better... I think that's where we should track the current tree closure
[19:16:17] <dianders_> davidjames: I'll update the bug and keep digging when I get to my desk...
[19:16:32] <dianders_> ...err, update the tree status.
[19:16:44] *** rginda_home has joined #chromium-os
[19:16:52] <rochberg> arching for supplied_chrome gives me: http://crosbug.com/21559
[19:17:03] <crosbot> tree became 'Tree is open (investigating http://crosbug.com/21945, http://crosbug.com/21559 and others; submit with care)'
[19:19:15] <davidjames> rochberg: Yup that's the bug!
[19:21:16] *** dianders_ has quit IRC
[19:24:12] <rochberg> Er, misdirected arrow-up + return
[19:24:16] <dianders> OK, at desk.  ;)  Going to dig through the various supplied_chrome crash and also create a shared log document.
[19:27:01] <jrbarnette> ok, i'm going offline to drive in; back in ~30 min
[19:27:17] <dianders> jrbarnette: See you soon!
[19:28:24] <ellyjones> kernel build just failed on my trybot
[19:28:25] <ellyjones> oO
[19:28:56] <ellyjones> kernel r588, which the pfq has not yet built
[19:30:10] <ellyjones> msb_: r588 breaks chromeos-kernel for me
[19:30:17] <dianders> ellyjones: OK, syncing and will try to reproduce locally to see what happens, too.
[19:30:42] <ellyjones> msb_: http://naya.i.corp.google.com/tries/c1543677-dea9-424c-a24e-8de7ac7f16d3.txt
[19:30:51] <ellyjones> dianders: you can see the build log there, it doesn't look spurious
[19:31:08] <dianders> ellyjones: OK, thanks.  Something easy to revert?
[19:31:36] <ellyjones> I don't know
[19:32:01] <dianders> ellyjones: OK, I'll look...
[19:32:16] <dianders> chromeos-kernel-0.0.1-r588:  * Using kernel config: chromiumos-i386 chromeos-kernel-0.0.1-r588: cat: chromeos/config/i386/i386.common.config: No such file or directory
[19:32:22] *** jrbarnette_ has joined #chromium-os
[19:33:13] <dianders> Can revert MSB change...
[19:33:48] <msb_> ellyjones: taking a look. yeah, just revert
[19:34:04] <ellyjones> ok
[19:34:27] <dianders> ellyjones: you doing the revert, or me?
[19:34:29] <ellyjones> http://gerrit.chromium.org/gerrit/#change,10497
[19:34:30] <ellyjones> LGTY?
[19:34:46] <dianders> ellyjones: LGTM
[19:34:52] *** ChocoboLee has joined #chromium-os
[19:34:53] *** ChanServ sets mode: +v ChocoboLee
[19:35:06] <ellyjones> landed
[19:35:12] <dianders> ellyjones: TY!  :)
[19:35:22] <ellyjones> the pfq is going to fail
[19:35:28] <ellyjones> since it started with msb's change but not the revert
[19:35:52] <dianders> ellyjones: I'll restart it.
[19:36:04] *** jrbarnette has quit IRC
[19:36:14] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "tegra2 PFQ" from 9ab5460945d4bd2c962838f63644ee4b09ea26a1: Ben Chan <benchan at chromium dot org>)'
[19:36:18] <dianders> Ah, there it goes
[19:36:28] <ellyjones> cool
[19:36:37] <dianders> Double checking the failure before restarting
[19:36:38] *** jrbarnette_ has quit IRC
[19:36:43] <dianders> yup
[19:36:51] * ellyjones tempted to 'while true; do try-x86; done' on one of her workstations :P
[19:37:46] <crosbot> tree became 'Tree is open (http://gerrit.chromium.org/gerrit/10497 fixed PFQ)'
[19:38:02] <crosbot> tree became 'Tree is open (http://gerrit.chromium.org/gerrit/10497 fixed PFQ, (investigating http://crosbug.com/21945, http://crosbug.com/21559 and others; submit with care)'
[19:38:17] <crosbot> tree became 'Tree is open (http://gerrit.chromium.org/gerrit/10497 fixed PFQ, investigating http://crosbug.com/21945, http://crosbug.com/21559 and others; submit with care)'
[19:38:31] <ellyjones> victory
[19:38:41] *** rharrison_chrome has joined #chromium-os
[19:38:42] *** ChanServ sets mode: +v rharrison_chrome
[19:38:55] <ellyjones> third trybot attempt for this CL! go
[19:43:39] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "stumpy canary" from None: )'
[19:44:07] <dianders> investigating tree failure...
[19:44:10] <davidjames> dianders: boo, google storage is down again
[19:44:18] <ellyjones> curl: (7) Failed to connect to 2001:4860:4001:6::18: Network is unreachable
[19:44:21] <ellyjones> fail!
[19:44:32] <davidjames> Woah, everything fails now!
[19:44:33] <ellyjones> reachable for me
[19:44:42] <ellyjones> in fact that url is fetchable
[19:45:18] <ellyjones> autotest-tests: ERROR:root:Dependency fakegudev does not exist
[19:45:20] <ellyjones> "???"
[19:45:22] *** grundler has joined #chromium-os
[19:45:23] *** ChanServ sets mode: +v grundler
[19:45:25] <dianders> davidjames: Ick, is that the problem for all of these?
[19:46:04] <davidjames> dianders: Yup, I can reproduce from my desktop too, can't download http://commondatastorage.googleapis.com/chromiumos-sdk/cros-sdk-21.10.11.003845.tbz2
[19:46:24] <dianders> davidjames: OK.  Escalation path here?  ...or just wait?
[19:46:30] <ellyjones> davidjames: what IP are you getting for it? I get... actually, I get an ipv4 address for it
[19:46:33] <dianders> davidjames, rochberg: OK to open tree?
[19:46:44] <davidjames> dianders: Trying 2001:4860:4001:6::18...
[19:47:07] *** patcito has joined #chromium-os
[19:47:07] <rochberg> No objections here
[19:47:09] <davidjames> dianders: I'll send you the escalation path over chat :)
[19:47:19] <ellyjones> won't buildboard just fail again?
[19:47:26] <dianders> davidjames: OK, thanks.  Escalating.
[19:47:48] <ellyjones> hah, I can fetch it from everywhere except mtv :P
[19:48:04] <ellyjones> has anyone seen this 'dependency fakegudev does not exist' message before?
[19:48:40] <SpanKY> is there a tag in the tracker for gardening tasks ?
[19:48:52] <ellyjones> Type-Cleanup
[19:49:40] <SpanKY> but what if it's a low priority bug and a cleanup task ;)
[19:49:59] <davidjames> SpanKY: Pri-3 if it's really low priority :)
[19:50:03] <ellyjones> then Type-Cleanup Type-Bug Pri-3 :)
[19:50:04] <dianders> OK, opening tree...
[19:50:08] <ellyjones> I think you're allowed more than one Type
[19:50:14] <SpanKY> tracker doesn't allow multiple Type-
[19:50:17] <ellyjones> oh
[19:50:18] <ellyjones> loss
[19:50:21] <ellyjones> well, just use Type-Bug then
[19:50:48] <crosbot> tree became 'Tree is open (bigstore failure from Google MTV escalated, investigating http://crosbug.com/21945, http://crosbug.com/21559 and others; submit with care)'
[19:51:22] <SpanKY> should add a Gardening tag me thinks
[19:51:28] <ellyjones> go for it
[19:52:27] <SpanKY> i think only admins get that access
[19:53:01] <davidjames> dianders: google storage is back up now I think
[19:53:15] <ellyjones> get what access? anyone can type a tag in
[19:53:23] <dianders> davidjames: yes
[19:53:25] <ellyjones> I just added it to 20183 by typing 'Gardening' into a tag field
[19:54:16] <SpanKY> yes, but it doesn't show up naturally for others
[19:54:21] <SpanKY> a bit harder for it to grow organically
[19:54:35] *** jrbarnette has joined #chromium-os
[19:54:36] *** ChanServ sets mode: +v jrbarnette
[19:54:52] <ellyjones> show up naturally...?
[19:55:04] <SpanKY> autocomplete when you start typing in a tag name
[19:55:10] <ellyjones> oh
[19:55:18] <SpanKY> i'll just post to cros-dev and see if people like the idea or think it's dumb
[19:55:25] <crosbot> tree became 'Tree is open (bigstore seems OK now, investigating http://crosbug.com/21945, http://crosbug.com/21559 and others; submit with care)'
[19:56:55] <jrbarnette> sheriff back on duty
[19:57:04] <ellyjones> hi
[19:59:12] <dianders> jrbarnette: welcome back
[20:02:19] *** wbednarski has quit IRC
[20:02:23] *** rbyers has quit IRC
[20:02:25] *** D|sT has quit IRC
[20:02:41] <dianders> OK, all the canaries are restarted (I think).  What else needs attention?  ...still need to resolve http://crosbug.com/21559
[20:04:41] *** elly has quit IRC
[20:05:02] <jrbarnette> 21559 seems to have two components:
[20:05:12] <jrbarnette> a) the failure to produce a stack trace
[20:05:16] <jrbarnette> b) the underlying problem
[20:05:46] <jrbarnette> for b), could that be http://code.google.com/p/chromium-os/issues/detail?id=20638
[20:06:11] *** rbyers has joined #chromium-os
[20:06:11] *** ChanServ sets mode: +v rbyers
[20:06:11] *** elly has joined #chromium-os
[20:06:32] <jrbarnette> and could 21559 and 21629 be duplicates?
[20:09:01] <jrbarnette> sosa:  online?
[20:09:23] <dianders> jrbarnette: I think davidjames was earlier saying that 21629 was originally a different bug.
[20:09:36] <dianders> ...but that the last two comments in it actually should have been 21559...
[20:10:05] <jrbarnette> ok, i think my current project is to sort these out
[20:10:11] <kliegs> crosbot: sheriffs?
[20:10:11] <crosbot> kliegs: sheriffs: jrbarnette, dianders, rochberg
[20:10:31] <kliegs> ^^^  I've got a pair of CL's to go in that are mutually dependent.  was planning on checking them in now
[20:10:53] <dianders> kliegs: OK, I guess it's as good a time as any...
[20:10:56] <jrbarnette> gerrit references?
[20:10:57] <kliegs> ^^^ I expect I can click 'submit' fast enough that there won't be a build with only one, but there is a small chance it will break the build
[20:11:02] <rochberg> My gardening project is to make it easier to get to the first @@@STEP_FAILURE@@@ (and then hopefully I can bubble that up the stack so we can reduce the number of clicks between a red tree and a cause)
[20:11:14] <cros_> If one wants to build for an ARM tegra2 target is there anyway to get the required open GL es libraries? I see in the opengles ebuild there is a tegra-local-bins use flag that is set by default but I don't see those binaries anywhere. It seems I can't access the git server to pull it myself either.
[20:11:21] <kliegs> http://gerrit.chromium.org/gerrit/10424 and http://gerrit.chromium.org/gerrit/10425
[20:11:32] <ellyjones> rochberg: this weekend I might hack crosbot to find and report those to IRC
[20:11:48] *** rbyers has quit IRC
[20:11:58] <kliegs> ok. pushing them now.
[20:12:00] <rochberg> I have the buildbot source.  How can I run that?  I have memories of people running related infrastructure and having it affect the global buildbot
[20:12:21] <jrbarnette> kliegs: as i understand from sosa, CLs will be pushed in the order you click submit
[20:12:25] <ellyjones> rochberg: run what?
[20:12:39] <rochberg> The buildbot status applicatoin
[20:12:46] <kliegs> jrbarnette: i pushed them in the order that shouldn't cause any redness.
[20:12:46] <ellyjones> oh, I wish I knew
[20:13:06] <kliegs> jrbarnette: although there are a few overlays that might act slightly odd
[20:13:06] <jrbarnette> ok, we'll watch for it
[20:13:40] <kliegs> cros_: I haven't tried doing that but the actual tegra binaries it references are I believe private
[20:14:06] <rochberg> Maybe if I read the docs....  https://sites.google.com/a/chromium.org/dev/developers/testing/chromium-build-infrastructure/getting-the-buildbot-source/running-a-buildbot-instance
[20:14:23] <kliegs> cros_: I believe I've seen people build tegra2 but don't remember exactly what they've done in terms of graphics drivers.
[20:15:13] <cros_> kliegs: Yeah, I would hope the binaries are available at least.
[20:15:18] <aaronp> cros_: It might be possible to use the binaries from http://developer.nvidia.com/beta-forum#/discussion/56 , though you may need to roll back the kernel too.
[20:15:27] <aaronp> There's a beta release in the works, but I don't know when it'll actually be released.
[20:17:01] <kliegs> cros_: the binaries come from nvidia.  I'm not sure if they have people active in this channel or not
[20:17:04] <cros_> kliegs: I had looked at a similar package this morning but I don't know if the graphics drivers are there. There is one image there but not sure what all it contains.
[20:17:28] <kliegs> cros_: there's also a #tegra channel which might be more useful,
[20:17:59] <dianders> rochberg, jrbarnette: What's up with the chrome PFQ failures, anyway?
[20:18:01] <cros_> kliegs: I guess everyone I've seen here working on tegra2 is a googler then.
[20:18:12] <cros_> kliegs: I'll checkout the other channel. Thanks.
[20:18:42] <kliegs> cros_: I've seen others who aren't in the past.   But can't remember there names - you might want to check the irc logs for the channel as I do remember this being discussed before
[20:19:10] <kliegs> sorry I'm not being more helpful - my memory suffers from too much bit rot
[20:19:36] <ellyjones> you need ECC
[20:19:50] <ellyjones> every time you remember a person's name, also remember the xor of some of the letters in it
[20:20:44] <kliegs> This xoring could go wrong.
[20:20:55] <kliegs> I might start referring to you as cmajones
[20:21:05] <ttuttle|work> haha
[20:21:10] <ellyjones> that would be awkward
[20:21:42] <ellyjones> platform_Pkcs11InitOnLogin import error: cannot import name pkcs11.  Skipping platform_Pkcs11InitOnLogin
[20:21:46] <ellyjones> anyone seen that one before?
[20:24:31] <jrbarnette> hmmm...  need a trooper:  x86-alex-canary failed with "no space left on device"
[20:26:52] *** cgreco has left #chromium-os
[20:26:52] <ellyjones> uhoh
[20:26:59] <ellyjones> kliegs:
[20:27:25] <kliegs> ellyjones: you forgot to xor my name
[20:27:34] <ellyjones> chromeosbuild2 is out of disk :P
[20:27:50] <dianders> crosbot: troopers?
[20:27:50] <crosbot> dianders: troopers: djmm, scottz, bradnelson, maruel (EST), kliegs (EST)
[20:27:55] <kliegs> logging in
[20:28:13] <dianders> kliegs: Thanks!  :)
[20:28:23] <ellyjones> I bet all the space is being taken up by /usr/lib. You should rm -rf /usr/lib. *nods*
[20:28:35] <ttuttle|work> ugh, /usr/lib
[20:28:38] <ttuttle|work> almost as bad as /usr/share
[20:29:12] <kliegs> hmm. machine has 13gb free on root
[20:29:17] <kliegs> and 161gb free on /b
[20:29:24] <kliegs> ellyjones: build2?
[20:29:34] <ellyjones> http://chromeos-botmaster.mtv.corp.google.com:8026/buildslaves/chromeosbuild2 this guy
[20:30:17] *** elly has quit IRC
[20:31:01] <kliegs> hmm.  its got lots of space
[20:31:23] <ellyjones> okay, who made client/cros in the autotest build be a symlink outside of the work tree
[20:31:54] <ellyjones> lrwxrwxrwx   1 ellyjones portage    49 Oct 21 14:00 cros -> /build/tegra2_kaen/usr/local/autotest/client/cros
[20:31:58] <ellyjones> do not want
[20:32:15] <kliegs> did anyone else free up spay on chromeosbuild2?
[20:33:36] <kliegs> s/spay/space
[20:36:24] <ellyjones> ah! ahhaha!
[20:36:31] <kliegs> i'm showing enough space on the device.  the directory it was writing to only need half a gig more and there's more
[20:36:43] <ellyjones> autotest uses the files it installed earlier in /build/$board/usr/local/autotest!
[20:36:58] <ellyjones> except that if you add a new file to autotest and then depend on it, you can't build autotest to install the new file!
[20:37:12] *** rbyers has joined #chromium-os
[20:37:12] *** ChanServ sets mode: +v rbyers
[20:38:29] <ellyjones> that makes little sense, though
[20:41:19] <kliegs> dianders: i'm not seeing it being low enough on space to have triggered those messages
[20:41:38] *** saintlou has quit IRC
[20:42:04] <dianders> kliegs: OK.  We can see if the next build passes?
[20:43:59] <kliegs> dianders: yah.  can you file a bug on this though?
[20:44:12] <dianders> kliegs: OK, no problem.
[20:46:19] *** raymes has quit IRC
[20:49:12] <dianders> OK guys, gotta run to a quick lunch.  Back soon.
[20:49:48] <dianders> kliegs: Bug is filed as http://crosbug.com/21955
[20:49:54] <dianders> ...not sure who should be owner, tho...
[20:50:20] <kliegs> dianders: me either.  i'll update with the df info I found
[20:50:31] <kliegs> make sure chrome-infrastructure-team is on cc
[20:53:00] *** behdad has joined #chromium-os
[20:53:14] *** jimhebert has joined #chromium-os
[20:53:14] *** ChanServ sets mode: +v jimhebert
[20:54:50] <kliegs> dianders: forwarded it on to infrastructure-team
[20:56:30] *** Sergiu has quit IRC
[20:59:53] <ellyjones> sosa: given that http://build.chromium.org/p/chromiumos/builders/x86%20generic%20commit%20queue/builds/12183 is green, why did http://gerrit.chromium.org/gerrit/10301 get 'fails'?
[21:01:58] *** behdad has quit IRC
[21:02:20] *** achuith has quit IRC
[21:02:27] *** achuith has joined #chromium-os
[21:02:57] <SpanKY> ellyjones: it says "apply cleanly" which means it couldn't cherry pick it
[21:03:02] *** achuith has joined #chromium-os
[21:03:02] *** ChanServ sets mode: +v achuith
[21:03:05] <SpanKY> iirc, the bot runs with a bit more pessimistic picking
[21:03:09] <ellyjones> ahh
[21:03:27] <SpanKY> showing the log in the cherry picking fail case is probably weird
[21:04:02] <SpanKY> or maybe not ... the log says: 2011/10/21 11:50:29 - validation_pool.py - INFO    : Change ellyjones:10301 did not apply cleanly.
[21:04:37] <SpanKY> error: Merge requires file-level merging
[21:04:37] <SpanKY> Strategy: resolve failed, try another
[21:04:40] <ellyjones> which is weird because I just said 'repo sync .; git rebase cros/master' and git was like 'ok!'
[21:04:42] *** D|sT has joined #chromium-os
[21:04:53] <ellyjones> so maybe it refuses to do any merging at all
[21:04:55] <SpanKY> yeah, i think repo sync runs with much more free wheeling rebase strategy
[21:05:04] *** rbyers has quit IRC
[21:05:04] <ellyjones> w4
[21:05:06] <ellyjones> er, fail
[21:05:11] <SpanKY> you're a w4
[21:05:37] <ellyjones> sometimes, yes
[21:05:51] <SpanKY> i think the cherry-picking-fail-case should link directly to the commitqueuesync step since that's the interesting bit
[21:06:02] <SpanKY> i'll file a tracker
[21:06:06] <ellyjones> there, manual rebase
[21:06:14] *** rbyers has joined #chromium-os
[21:06:14] *** ChanServ sets mode: +v rbyers
[21:06:24] <SpanKY> maybe even update the CQ doc
[21:09:43] *** seumas has quit IRC
[21:11:03] *** kevers has quit IRC
[21:11:47] <SpanKY> ah the CQ FAQ does mention: The commit queue uses the same cherry-pick mechanism that Gerrit does when it tries to commit your change which is more conservative than traditional git-cherry-pick.
[21:12:29] <kliegs> Who is SpanKY?
[21:12:35] <SpanKY> i am
[21:12:51] <SpanKY> sometimes i'm also vapier
[21:12:58] <kliegs> that makes more sense :)
[21:13:02] <ellyjones> have you considered screen+irssi? :)
[21:13:16] <ellyjones> (btw, kliegs, he's in nerf gun range :P)
[21:13:20] <SpanKY> comes in handy when i want to drum up support for my ideas
[21:13:35] <ellyjones> haha
[21:13:38] <kliegs> ellyjones: i'd recommend ducking then if you're suggesting screen+irssi
[21:13:44] <ellyjones> sockpuppets on chromium-os-dev!
[21:13:51] <ellyjones> kliegs: I don't think he has a nerf gun :P
[21:14:01] <ellyjones> irssi is the correct client, though
[21:14:57] <SpanKY> i'm more of a bx guy
[21:15:11] *** rbyers has quit IRC
[21:15:47] <ellyjones> not cool.
[21:19:15] <SpanKY> bx predates irssi :p
[21:19:18] *** behdad has joined #chromium-os
[21:19:37] <SpanKY> http://crosbug.com/21958 for better msg
[21:22:16] *** seumas has joined #chromium-os
[21:22:16] *** ChanServ sets mode: +v seumas
[21:23:00] *** posciak has joined #chromium-os
[21:23:01] *** ChanServ sets mode: +v posciak
[21:23:36] *** posciak has quit IRC
[21:23:48] <ellyjones> cool
[21:24:48] *** posciak has joined #chromium-os
[21:24:48] *** ChanServ sets mode: +v posciak
[21:31:35] *** posciak has quit IRC
[21:32:07] *** seumas has quit IRC
[21:33:57] *** posciak has joined #chromium-os
[21:33:57] *** ChanServ sets mode: +v posciak
[21:37:40] *** behdad has quit IRC
[21:43:08] *** BThompson has quit IRC
[21:50:32] <ellyjones> sosa: is there a way to tell the pfq to ignore a commitqueue failure? I suffer from <http://build.chromium.org/p/chromiumos/builders/x86%20generic%20commit%20queue/builds/12185/steps/VMTest/logs/stdio>
[21:51:48] <ellyjones> (or tell it to try again or whatever)
[22:05:51] *** kevers has joined #chromium-os
[22:05:52] *** ChanServ sets mode: +v kevers
[22:07:38] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "lumpy-binary" from 7d6ee1550ba11a7c53a48514723f1674aa941140: Mandeep Singh Baines <msb at chromium dot org>)'
[22:08:27] <dianders> looking at closure (BTW: back from lunch but jrbarnette now at lunch)
[22:09:11] *** darinski_ has joined #chromium-os
[22:09:20] <dianders> kernel failure
[22:09:26] <dianders> suspect msb.  ;)  (sorry msb)
[22:09:36] *** akrpic77 has quit IRC
[22:10:11] <crosbot> tree became 'Tree is closed (kernel failure (msb?) => dianders)'
[22:13:20] <cros_> ARM folks, it seems the bkpt instruction is being executed at some points in the code triggering a prefetch abort. Can someone explain why this is being done and what function or macro is called in the code for this?
[22:13:56] <dianders> MSB change reverted
[22:16:09] <dianders> Revert is here: http://gerrit.chromium.org/gerrit/10509
[22:16:27] <ellyjones> lgtm++ :)
[22:16:27] <dianders> ...going to reopen tree on the assumption that this should be fixed.
[22:16:44] *** itsonlyme has joined #chromium-os
[22:17:05] <crosbot> tree became 'Tree is open (still some flaky failures, but generally OK; kernel CL reverted)'
[22:19:21] <dianders> Hmmm, why doesn't my revert show up in the waterfall?  :(  ...would have expected it to.  Let me see if it actually is going to be in the -binary builds.
[22:19:55] <ellyjones> I see it in the waterfall
[22:20:00] <ellyjones> at 13:10
[22:20:26] <kliegs> dianders: the waterfall seems to lag ~5 minutes behind I think
[22:20:49] <dianders> kliegs: Oh, that's unfortunate.  I assume that it will still show up in the builds I kicked off?
[22:21:11] <dianders> Ah, yes.  It's there.  :)
[22:21:12] <kliegs> dianders: not sure, honestly.  I'm not sure where the actual lag is
[22:21:16] <kliegs> dianders: ok. that's good
[22:21:28] <dianders> From http://git.chromium.org/chromiumos/third_party/kernel
[22:21:28] <dianders> a79c10d..0ddc193  chromeos-2.6.38 -> cros/chromeos-2.6.38
[22:21:39] <dianders> ...and 0ddc1931523a579af6962358a01cbb45d525ae75 is the revert.  :)
[22:27:47] <jrbarnette> sheriff back on duty
[22:28:51] <dianders> jrbarnette: Welcome back.  Gonna grab a quick coffee, then look at this one:
[22:28:52] <dianders> http://build.chromium.org/p/chromium.chromiumos/builders/ChromiumOS%20%28x86%29/builds/542/steps/cbuildbot_chrome/logs/stdio
[22:29:03] <dianders> ...looks like cros_mark_chrome_as_stable bug
[22:29:11] <ellyjones> does anyone know how to cause the commitqueue to retry?
[22:29:29] <dianders> ellyjones: I think you need to remove the commit queue as a reviewer?
[22:29:47] <ellyjones> done, now what?
[22:29:56] <dianders> ...then I _think_ you just wait
[22:30:04] <dianders> ...but sosa would know for sure...
[22:30:19] <ellyjones> sosa didn't answer the last time I asked :)
[22:31:26] <jrbarnette> sosa is at lunch
[22:34:28] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86 pineview full" from 7d6ee1550ba11a7c53a48514723f1674aa941140: Mandeep Singh Baines <msb at chromium dot org>)'
[22:34:51] <sosa> hmm
[22:35:12] <sosa> didn't it tell you in the email?
[22:35:23] *** rbyers has joined #chromium-os
[22:35:23] *** ChanServ sets mode: +v rbyers
[22:36:41] <ellyjones> it told me to sync, rebase, and reupload
[22:36:49] <dianders> pineview full is the same kernel config bug, I think
[22:36:52] <dianders> ...verifying
[22:37:10] <dianders> yup, must have not noticed that needed to be restarted...
[22:37:42] <ellyjones> cool
[22:38:33] <crosbot> tree became 'Tree is open (pineview is another kernel config failure; still some flaky failures, but generally OK)'
[22:40:00] <dianders> ...actually, it looks like the pineview might be a different failure also related to the same CL.  ...so I think we're still OK.  I'll make sure msb checks that one, too.
[22:41:00] *** itsonlyme has quit IRC
[22:41:56] <ellyjones> cool
[22:42:28] *** rbyers has quit IRC
[22:44:00] *** rbyers has joined #chromium-os
[22:44:00] *** ChanServ sets mode: +v rbyers
[22:45:40] <dianders> Just to keep IRC updated: petermayo is working on the chromium.chromiumos failure that I mentioned above (thanks to jrbarnette for finding the right person).
[22:46:46] <dianders> Note that cros_sdk still seems to be failing on recent chromium.chromiumos builds.  I think this is the same bigstore failure that we saw earlier.  Double-checking that failure mode looks the same...
[22:48:23] *** saintlou has joined #chromium-os
[22:48:23] *** ChanServ sets mode: +v saintlou
[22:48:25] <dianders> Nope.  It is the cros_mark_chrome_as_stable bug, again.  Just fails differently than the URL I pointed to above.
[22:48:33] <dianders> ...so I guess petermayo is still on it.  ;)
[22:49:04] <sosa> elly:  looking at 10301 there are two different messages sent to you for the two different cases
[22:50:05] <dianders> ...OK, so back to digging chrome PFQ
[22:50:07] <sosa> ellyjones:  the first time your CL was rejected because the commit queue could not apply your change to TOT cleanly, so it asked you to rebase and reupload. The second one was beacuse of some flaky build issue and told you a different message (if you think this happened in error just remove the Gerrit reviewer).  I believe that's what you wanted
[22:50:20] <ellyjones> yeah, I removed the Gerrit reviewer
[22:50:50] <ellyjones> it has not gotten into the commit queue yet
[22:52:28] <jrbarnette> dianders:  according to sosa, the chrome PFQ depends on the chromium.chromiumos builders
[22:52:32] <jrbarnette> ... more or less
[22:52:49] <dianders> jrbarnette: Fair enough.  ...that would explain why they're not building anymore.
[22:52:56] <dianders> ...but the question is: was the bug fixed?
[22:53:07] <dianders> ...there was a real compile bug in Chrome.
[22:53:21] <dianders> content/renderer/java_bridge_dispatcher.cc:69:76: error: passing NULL to non-pointer argument 4 of ?bool CreateNPVariant(const NPVariant_Param&, NPChannelBase*, NPVariant*, gfx::NativeViewId, const GURL&)? [-Werror=conversion-null]
[22:53:23] <sosa> i got an email this morning saying that they reverted a change committed last night that broke us
[22:53:26] <sosa> i'll forward you the email
[22:53:37] <dianders> sosa: OK, thanks!
[22:54:07] *** D|sT has quit IRC
[22:54:29] <sosa> oh wait it was sent to chromeos-team
[22:57:15] <dianders> sosa: Ah, see it now.  Sorry I missed it earlier.  :(
[22:57:24] <dianders> ...we're seeing a different error, though...
[22:59:49] <dianders> ...though I guess it is related to the "all warnings being treated as errors"...
[23:04:59] <petermayo> dianders:  off to a 50 minute commitment ... back on it again after.
[23:05:15] <dianders> petermayo: OK.
[23:06:07] *** rharrison_chrome has quit IRC
[23:06:29] *** petermayo has quit IRC
[23:09:18] *** petermayo has joined #chromium-os
[23:09:19] *** ChanServ sets mode: +v petermayo
[23:10:28] *** SpanKY has quit IRC
[23:11:37] *** gut4 has quit IRC
[23:12:11] *** rbyers has quit IRC
[23:17:24] *** rbyers has joined #chromium-os
[23:17:31] *** ChanServ sets mode: +v rbyers
[23:17:41] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86-zgb-binary" from 1d2e54a6c24aae92e8da1f5142ccfe68a2e52fe9: Eric M. Blake <eblake at chromium dot org>)'
[23:18:27] <dianders> petermayo: I did do a little digging into the chromium.chromiumos issue (not sure how far you got).  It looks as if the --chrome_version argument to cbuildbot changed.
[23:19:12] <dianders> Looking at build 423, --chrome_version=106189
[23:19:13] <dianders> http://build.chromium.org/p/chromium.chromiumos/builders/ChromiumOS%20%28x86%29/builds/423/steps/cbuildbot_chrome/logs/stdio
[23:19:38] <dianders> ...at 550, --chrome_version=svn://svn-mirror.golo.chromium.org/chrome/trunk@106777
[23:19:40] <dianders> http://build.chromium.org/p/chromium.chromiumos/builders/ChromiumOS%20%28x86%29/builds/550/steps/cbuildbot_chrome/logs/stdio
[23:20:02] <dianders> ...and that's the bit that's confusing things.
[23:20:38] <dianders> It's possible that a quick change to cros_mark_chrome_as_stable would suffice to just have it handle the new format, but you probably understand all that code better than I do...
[23:20:52] <dianders> ...whoops, didn't notice tree closure.  Looking!
[23:21:36] <dianders> Ah, another http://crosbug.com/21945
[23:21:51] <sosa> dianders: sent you a cl to workaround issue for now for chrome builders
[23:22:05] <sosa> dianders: peter's going to try to track down what changed later but it'd be good to get these builders back up
[23:22:08] <dianders> sosa: OK.  1 sec...
[23:23:03] <petermayo> dianders: yeah ... that is where things came off the rails.  ;-(
[23:23:19] <crosbot> tree became 'Tree is closed (zgb is another http://crosbug.com/21945; lumpy a kernel config => dianders)'
[23:25:03] *** SpanKY has joined #chromium-os
[23:25:04] *** ChanServ sets mode: +v SpanKY
[23:25:49] <jrbarnette> ok, i spoke w/ davidjames, and he pushed the cl to fix 21945
[23:27:08] <dianders> jrbarnette: OK, thanks.  The kernel config error must not be related to msb's stuff, since that's been reverted.  Maybe it's something else?  Digging
[23:27:12] *** cros_ has quit IRC
[23:29:14] *** McMAGIC--Copy has quit IRC
[23:30:55] <dianders> Hmmm, sorta surprised that the kernel config error happened in lumpy.  It seems like revert should have fixed it.  :-/
[23:31:07] *** McMAGIC--Copy has joined #chromium-os
[23:32:57] <jrbarnette> do we care about lumpy failures?
[23:33:32] <dianders> Ah, I see.  The lumpy failure is because the build_image emerges an older version that didn't have my revert
[23:33:33] <dianders> Ick
[23:33:39] <dianders> That's pretty weird.  Anyway, it will work next time.
[23:33:53] <dianders> ...kicking the build and re-opening
[23:34:04] <jrbarnette> hip hip hooray!
[23:34:34] <crosbot> tree became 'Tree is open (zgb is another http://crosbug.com/21945; lumpy will work next time when build image picks up kernel 590)'
[23:40:26] *** dkz has quit IRC
[23:46:51] *** ChanServ sets mode: +v vapier
[23:51:49] *** SpanKY has quit IRC
[23:52:37] *** sadrul has joined #chromium-os
[23:52:37] *** ChanServ sets mode: +v sadrul
[23:58:15] *** petermayo has quit IRC





top