[00:01:22] *** Inumedia has quit IRC
[00:01:53] *** Inumedia has joined #chromium-os
[00:04:50] <satorux_> this happens so often
[00:07:40] *** Solet has quit IRC
[00:08:15] *** TW1920 has quit IRC
[00:11:03] <msb__> kliegs, satorux_: pushed
[00:11:12] <kliegs> msb__: thanks
[00:11:29] <kliegs> satorux_: i haven't been able to do a revernt in chromiumos-overlays without needing to rebase :(
[00:11:29] *** semenzato has joined #chromium-os
[00:11:29] <satorux_> thanks!
[00:11:29] *** ChanServ sets mode: +v semenzato
[00:11:43] <kliegs> nearing end of day for me - need to get to gym son
[00:11:48] <kliegs> s/son/soon
[00:13:50] <crosbot> tree became 'Tree is open (reverted http://gerrit.chromium.org/gerrit/#change,6861)'
[00:13:51] <satorux_> heh, sheriffing from gym
[00:18:22] *** lipsinV2 has joined #chromium-os
[00:19:33] <kliegs> satorux_: i can see it now.  today's WOD - 3 sets, AMRAP, 15xkettle balls, 5xcode reviews, 10xback squat
[00:25:37] *** sadrul has quit IRC
[00:27:09] <satorux_> kliegs: 2 x reverts not included. have good evening.
[00:27:16] *** rbyers_ has quit IRC
[00:27:25] *** D|sT has quit IRC
[00:27:32] <kliegs> :) take care.
[00:27:42] <kliegs> oops. i'm the one leaving.. bye
[00:36:12] *** benchan has quit IRC
[00:51:22] *** [S]olet has joined #chromium-os
[00:53:16] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86 pineview full" from 8982204dee44ece1cef2178747a27f08beb374cf: Taylor Hutt <thutt at chromium dot org>, bradnelson at google dot com <bradnelson at google dot com@fcba33aa-ac0c-11dd-b9e7-8d5594d729c2>)'
[00:53:21] <satorux_> looking
[00:54:42] <satorux_> seems to be a timeout issue
[00:54:59] <satorux_> 62 packages pending to be built
[00:55:09] <satorux_> duration: 8 hrs, 0 secs
[00:56:05] <crosbot> tree became 'Tree is closed (x86 pineview full failed. sheriffs looking)'
[00:59:11] <msb__> satorux_: earlier in the log is a curl error for the damageproto: curl: (22) The requested URL returned error: 404
[00:59:24] <satorux_> msb__: good point
[01:00:03] <satorux_> msb__: does this mean it's a network issue?
[01:01:22] <msb__> satorux_: I see Completed installing damageproto-1.2.1 into /build/x86-pineview/tmp/portage/x11-proto/damageproto-1.2.1/image/
[01:01:42] <msb__> and then below that: Still running: job damageproto-1.2.1
[01:01:51] <satorux_> msb__: weird...
[01:01:55] <msb__> if its Completed, why is it still running?
[01:02:20] <satorux_> msb__: maybe an parallel ebuild issue
[01:02:45] <satorux_> davidjames: yt? any ideas on this failure?
[01:05:44] <satorux_> seemed like a deadlock or something
[01:05:51] <msb__> davidjames says to re-open and open a closer
[01:05:55] *** Adys has quit IRC
[01:08:06] <satorux_> msb__: are you filing a bug?
[01:08:25] <davidjames> satorux_: Yeah looks like some issue that causes a hang, just not sure what caused the hang with the logs there
[01:09:15] <crosbot> tree became 'Tree is open (filing a bug for the x86 pineview full hang)'
[01:11:18] <msb__> satorux_: 19801
[01:11:24] <satorux_> msb__: thanks
[01:12:05] <crosbot> tree became 'Tree is open (the x86 pineview full hang bug filed: crosbug.com/19801)'
[01:15:44] *** saintlou is now known as saintlou|away
[01:15:46] *** saintlou|away is now known as saintlou
[01:18:47] <sergiu> hmm I need some help with some inline assembly, anyone with experience that could help me out? :)
[01:19:12] <ellyjones> I might be able to
[01:19:39] <sergiu> lemme post the code to pastebin
[01:19:58] <sergiu> or do we have something similar internally (not that it's anything secret :) )
[01:20:19] *** Adys has joined #chromium-os
[01:20:30] *** srao has quit IRC
[01:21:29] <sergiu> ellyjones, http://pastebin.com/AF9tShE5
[01:21:38] <sergiu> I've left the C code I'm translating as well
[01:21:56] <sergiu> I'm not sure how exactly I should address the output registers
[01:22:13] <sergiu> (this is my first time doing something like that, so I may be horribly wrong :) )
[01:22:56] <ellyjones> did you look at http://gcc.gnu.org/onlinedocs/gcc/Extended-Asm.html ?
[01:25:53] <sergiu> I've taken a look at diffrent resources
[01:28:02] <sergiu> one thing I'm not sure is how it identifies output registers
[01:29:44] <sergiu> there are examples such as this one :
[01:29:47] <sergiu> asm ("movl %1, %%eax;
[01:29:47] <sergiu> 	     "movl %%eax, %0;"
[01:29:47] <sergiu> 		:"=r"(y)	/* y is output operand */
[01:29:47] <sergiu> 		:"r"(x)		/* x is input operand */
[01:29:47] <sergiu> 		:"%eax");	/* %eax is clobbered register */
[01:32:10] <sergiu> apparently %0 and %1 also represent the order in the output/input list
[01:54:47] *** saintlou is now known as saintlou|away
[01:57:34] *** sadrul has joined #chromium-os
[01:57:35] *** ChanServ sets mode: +v sadrul
[01:58:32] <satorux_> i'm about to go home
[01:59:21] <zbehan> the PFQ is about to die horribly on nearly all smoke tests dead
[01:59:47] <msb__> satorux_: me too
[01:59:48] <satorux_> does that mean I'm going home just about the right time?
[02:00:07] <satorux_> hmm
[02:00:15] <ellyjones> what happened?
[02:01:13] <ellyjones> wow, it looks like chrome is _totally_ toasted
[02:02:02] <zbehan> actually, it seems rsync onto the VM failed, so the tests are missing some files
[02:02:19] <zbehan> hard to tell in this goulash though
[02:02:23] *** behdad has joined #chromium-os
[02:05:54] <satorux_> is it similar to regressions in the last week? there were some network library related regressions
[02:06:10] <satorux_> these should now be fixed
[02:07:17] <satorux_> but stevenjb was checking in several patches to clean it up, which might be risky
[02:07:41] <satorux_> I need to run to catch a bus, but fwiw, http://codereview.chromium.org/7765009/ is the change I had in mind
[02:08:55] <stevenjb> I'm here, I'll take a look, make sure my commits today aren't causing any problems.
[02:09:36] *** behdad has quit IRC
[02:09:51] *** behdad has joined #chromium-os
[02:11:02] *** reveman has quit IRC
[02:12:01] *** Styx has quit IRC
[02:12:47] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot_master" on "TOT Pre-Flight Queue" from ace4f6d1cf25889621a95bb8cffe224b1b83eeb6: _third_party_ at chromium dot org)'
[02:13:15] *** oc80z has quit IRC
[02:13:15] *** oc80z has joined #chromium-os
[02:18:22] *** D|sT has joined #chromium-os
[02:20:29] *** kinaba has quit IRC
[02:31:07] *** behdad has quit IRC
[02:36:08] *** srao has joined #chromium-os
[02:37:08] <srao> anybody ever seen this while trying to run an autotest with "--build" : dep-pyauto_dep.tar.bz2 could not be fetched from any of the repos ['autoserv://']
[02:37:54] *** saintlou|away is now known as saintlou
[02:46:28] *** kochi1 has joined #chromium-os
[02:49:16] <zbehan> srao: i think you have to run with --use_emerged if you see that
[02:50:01] <gauravsh> and if you are working on a specific test, you should emerge it before running it via run_remote_tests.sh
[02:50:59] *** wfrichar has quit IRC
[02:51:03] <srao> zbehan and gauravsh: thanks, --use_emerged got me going
[02:51:47] <zbehan> srao: running without --use_emerged is a shortcut for tests without deps external to autotest.git tree, but with the introduction of pyauto heavy use, that list of tests was reduced significantly
[02:51:58] *** saintlou has quit IRC
[02:53:26] <srao> ok, I was told to use --build in order to run a shortened version of power_LoadTest.... I ended up starting it without --build, then editing the control file on the target then using --use_emerged
[02:53:41] <srao> i guess I also need to go learn up about pyauto
[02:54:06] <nirnimesh> srao: http://dev.chromium.org/developers/testing/pyauto is a good place to start
[02:54:17] *** kochi1 has quit IRC
[02:54:52] <srao> nirnimesh: cool thanks
[02:55:19] <stevenjb> davidjames or sosa - any idea what is up with VMTest in the PFQ?
[02:55:53] <davidjames> stevenjb: Looks like someone committed a change that broke it, referencing pyauto
[02:57:49] <nirnimesh> session_manager not running at the beginning of the test. chrome could not launch
[02:58:27] <davidjames> nirnimesh: It should probably define pyauto earlier right?
[02:58:42] <davidjames> nirnimesh: So that pyauto variable is at least 'None' and it can check whether it is set :)
[02:59:51] <nirnimesh> Yeah, I agree that the final error is cryptic and needs fixing. but you can also see "^session_manager$ is not running at start of test"
[03:02:58] <davidjames> It is probably a change to chromiumos-overlay between 3:41pm and 4:20pm
[03:03:14] <davidjames> It can't be a workon repo, because in that case internal builders would work
[03:05:07] <davidjames> stevenjb: Buildbot seems to have handled Chrome upgrade OK and passed the first time after Chrome upgrade
[03:05:19] <sosa> only 2 fall under your theory though ... your chroot commit and ellyjones commit for updating libnss
[03:05:42] *** kochi1 has joined #chromium-os
[03:05:59] <stevenjb> davidjames: Great, cheers.
[03:06:53] <davidjames> sosa: Yeah could be the libnss change or flakiness after Chrome upgrade
[03:07:33] <davidjames> sosa: I'm suspecting not Chrome, because TOT Chrome PFQ started failing too around the same time
[03:08:19] <davidjames> sosa: So I'm suspecting ellyjones' change, any chance we could revert that to see if it helps?
[03:09:27] <davidjames> sosa: Could also be flameeyes change to large number of pkgs
[03:09:48] <davidjames> sosa: Oh never mind, it broke before that
[03:09:57] <Flameeyes> davidjames: what's the failure like? afaict my change should be transparent given that the ebuilds simply moved
[03:10:11] <Flameeyes> (if the ebuilds changed from those I committed to portage-stable, it wouldn't have rebased/cherry-picked)
[03:10:42] <davidjames> Flameeyes: Yeah actually it failed before your change went in
[03:10:53] <Flameeyes> okay, phew :)
[03:11:00] <Flameeyes> well, not phew in general, but at least for me ;)
[03:11:22] <Flameeyes> [especially since I'm already/still building to test the next five changes]
[03:14:02] *** chocobo__ has quit IRC
[03:14:57] <davidjames> sosa, stevenjb: chromeos-login unit tests are also failing with nss errors, that makes me suspect ellyjones' change more
[03:15:16] <davidjames> sosa, stevenjb: In particular they're failing with a nss-related segfault, and ellyjones' change touches nss
[03:15:38] <stevenjb> Hm, seems like a reasonable theory to me.
[03:15:56] <stevenjb> But I'm not honestly super familiar with those tests.
[03:20:54] *** grundler has quit IRC
[03:27:59] <stevenjb> davidjames: I see the UnitTest failure in x86-generic-full 252, but not in other buildbots with VMTest failures...
[03:28:21] *** dennisjeffrey has quit IRC
[03:36:05] *** frankf has joined #chromium-os
[03:36:06] *** ChanServ sets mode: +v frankf
[03:36:31] <davidjames> stevenjb: Yeah the preflight bots don't run unit tests
[03:36:56] <stevenjb> oh, ok.
[03:37:03] <davidjames> stevenjb: (Well, they run unit tests, but only for changed packages and chromeos-login was not directly changed)
[03:37:41] <stevenjb> gotha,
[03:38:29] <davidjames> stevenjb: OK just went ahead and reverted it
[03:39:27] <stevenjb> I think that's the right call.
[03:40:44] <stevenjb> I cross referenced a couple of the failing builders and it's somewhere between kschimpf@ : 0ef2a77e and zbehan@ : e6b0c698
[03:40:56] *** Adys has quit IRC
[03:41:08] <davidjames> stevenjb: Yeah I also reproduced the crash locally just cherry-picking ellyjones' change
[03:41:13] <stevenjb> The libnss change is by far the most suspicious :)
[03:41:13] <davidjames> stevenjb: Without the change it works fine
[03:41:17] <stevenjb> Ah, even better.
[03:41:44] <stevenjb> Pretty easy call then.
[03:43:22] *** stevenjb has quit IRC
[03:55:10] *** Adys has joined #chromium-os
[04:02:38] <crosbot> tree became 'Tree is closed (Should cycle green with nss revert, waiting for preflight -> http://gerrit.chromium.org/gerrit/6877)'
[04:03:17] <davidjames> crosbot: sheriffs?
[04:03:17] <crosbot> davidjames: sheriffs: satorux, msb, falken, kliegs, kochi
[04:03:51] <davidjames> satorux, msb, falken, kliegs, kochi: Hopefully my revert fixed the tree, bots are running now
[04:03:55] <davidjames> See  http://gerrit.chromium.org/gerrit/6877
[04:12:33] <kochi> davidjames: thanks, i will see.
[04:12:57] <davidjames> kochi: Thanks, if any builder that runs tests cycles green then probably the issue is fixed
[04:13:27] <davidjames> kochi: (i.e. any x86-* bot)
[04:13:40] <kochi> non PFQ bots?
[04:13:52] <davidjames> kochi: PFQ bots do run VM tests
[04:14:06] <davidjames> kochi: So that should be enough to verify the fix
[04:14:11] <kochi> i see, okay
[04:14:44] <davidjames> (I confirmed the revert fixes unit tests, but did not test VM tests, so waiting for bots on those)
[04:16:03] *** elly has quit IRC
[04:16:18] *** elly has joined #chromium-os
[04:29:23] <kochi> okay, TOT PFQ #5072 looks it is going to be the first one to complete VMTest after davidjames's change.
[04:33:26] *** Sarten-X has quit IRC
[04:41:01] <zbehan> kochi: x86 PFQ is green
[04:41:55] <davidjames> zbehan: Yeah probably time to open, note though that canaries will fail soon so we should probably stop them before they fail
[04:43:20] <kochi> zbehan: I'm wondering why some VMTests already have gone green before catching the davidjames's 19:09:03 change.
[04:43:39] <zbehan> kochi: which?
[04:43:45] <kochi> e.g. one at 19:08:07 TOT PFQ
[04:44:26] <davidjames> kochi: I committed my change earlier than that on TOT I believe
[04:44:29] <kochi> That build (#5071) is before davidjames's change, right?
[04:44:35] <davidjames> kochi: CommitDate: Mon Aug 29 18:37:37 2011 -0700
[04:44:57] <kochi> oh, the leftmost column has some delay, then.
[04:44:58] <davidjames> kochi: I also had changes on branches, maybe you are looking at the branch version of my change
[04:45:20] <davidjames> kochi: (Same change had to be reverted on R13 and R14)
[04:45:21] <kochi> I see
[04:45:24] <zbehan> kochi: that one did have it
[04:45:34] <zbehan> kochi: (it's on the commit list)
[04:45:48] <kochi> ok, then I'll open the tree
[04:45:54] <davidjames> kochi: Heading home now, but suggest opening the tree
[04:46:11] <zbehan> i can watch out and reopen if i see other builders failing with the same
[04:46:45] <kochi> opened.
[04:46:53] <davidjames> zbehan, kochi: Thanks, see you tomorrow
[04:46:54] <crosbot> tree became 'Tree is open'
[04:47:12] <kochi> davidjames: have a good night!
[04:48:44] <zbehan> davidjames: thanks for tracking this down!
[04:50:17] *** Sarten-X has joined #chromium-os
[05:58:45] *** kinaba has joined #chromium-os
[06:14:29] *** powderluv has quit IRC
[06:31:24] *** nirnimesh_ has quit IRC
[06:40:24] *** powderluv has joined #chromium-os
[06:40:24] *** ChanServ sets mode: +v powderluv
[07:06:55] *** deshantm is now known as deshantm_away
[07:10:03] *** stalled has quit IRC
[07:16:10] *** powderluv has quit IRC
[07:19:39] *** Adys has quit IRC
[07:19:49] *** Adys has joined #chromium-os
[07:20:09] <kochi> [Chrome OS] autotest login trouble
[07:20:43] <kochi> oops
[07:23:06] *** vmil86 has joined #chromium-os
[07:43:19] 
[07:44:20] <kochi> mm, this one should be the one davidjames's cl should fix it.
[07:48:40] <kochi> mm, no?
[07:50:30] *** powderluv has joined #chromium-os
[07:50:31] *** ChanServ sets mode: +v powderluv
[08:08:09] <crosbot> tree became 'Tree is open (x86 generic full VMTest failure AU test flaky? -> kochi looking)'
[08:11:59] *** magn3ts has quit IRC
[08:17:52] *** chocobo__ has joined #chromium-os
[08:17:52] *** ChanServ sets mode: +v chocobo__
[08:21:37] *** chocobo__ has quit IRC
[08:26:57] *** sergiu has quit IRC
[08:37:03] *** stalled has joined #chromium-os
[08:43:46] 
[08:45:54] *** zmedico has quit IRC
[08:51:10] *** zmedico has joined #chromium-os
[08:53:21] <kochi> LKGM candidate sync error... seems safe to ignore.
[08:54:04] <crosbot> tree became 'Tree is open (x86 generic full VMTest failure AU test flaky? -> kochi looking)'
[08:58:40] *** chocobo__ has joined #chromium-os
[08:58:41] *** ChanServ sets mode: +v chocobo__
[09:00:09] *** chocobo___ has joined #chromium-os
[09:00:13] *** jujugre has joined #chromium-os
[09:00:17] *** ChanServ sets mode: +v chocobo___
[09:00:24] *** chocobo___ has quit IRC
[09:03:47] *** chocobo__ has quit IRC
[09:12:05] *** gfrog has quit IRC
[09:24:48] *** Styx has joined #chromium-os
[09:25:11] *** gfrog has joined #chromium-os
[09:26:23] *** ferringb has quit IRC
[09:28:26] *** ferringb has joined #chromium-os
[09:32:44] <zbehan> Flameeyes: ping?
[09:32:49] <Flameeyes> pong
[09:33:08] <zbehan> Flameeyes: looking at the dhcp CL, what is the purpose of that? isn't the ebuild properly unmasked now?
[09:33:34] <Flameeyes> zbehan: ebuild was modified in portage-stable to change the keywords, which is messy, most other ebuilds with modified keywords are handled through p.keywords
[09:33:52] <Flameeyes> p.unmask _might_ be unnecessary ? I just happened to have it around because I was working on the same change last night
[09:33:53] <zbehan> Flameeyes: ah, i see
[09:34:02] <Flameeyes> (thus why I fixed the libcap cross-compile thing in main tree ^^;;)
[09:34:42] <zbehan> Flameeyes: sorry :) been trying to pull this into the tree for days now because of linux 3 compatibility
[09:34:52] <Flameeyes> yeah I guessed that ;)
[09:35:09] <Flameeyes> same here fwiw, it's just that it takes .. a while to build everything
[09:35:12] <zbehan> package.unmask is necessary, but i put it into portage-stable/profiles/
[09:35:17] <Flameeyes> I guess the tinderbox host is getting old :(
[09:35:27] <zbehan> (a little more global)
[09:36:48] *** ferringb has quit IRC
[09:36:49] <zbehan> Flameeyes: so i guess landing your CL is worth a followup back into portage-stable, should i send that?
[09:37:07] <Flameeyes> zbehan: http://gerrit.chromium.org/gerrit/#change,6889 together with other ebuilds ;)
[09:38:49] *** ferringb has joined #chromium-os
[09:39:12] <zbehan> Flameeyes: ook, looks you already got it all covered, sorry for the interference then :) does the global package.unmask in portage-stable fall into that category? maybe you want to remove it then
[09:40:07] <Flameeyes> zbehan: that was never discussed so I honestly have no idea, it doesn't really make a difference to me, so until there's need I guess it could stay :) ? now if only build_packages completed with success ... :(
[09:40:35] <zbehan> Flameeyes: what does it fail on?
[09:40:53] <Flameeyes> chromeos-chrome-15.0.865.0_rc-r1: scons: *** [scons-out/nacl_irt-x86-32/obj/src/shared/ppapi_proxy/command_buffer/client/gles2_cmd_helper.o] Error 1
[09:41:10] <Flameeyes> it's "retrying later", let's see if it makes a difference
[09:41:20] *** ferringb has quit IRC
[09:41:20] *** ferringb has joined #chromium-os
[09:41:20] *** ChanServ sets mode: +v ferringb
[09:41:26] <zbehan> ah, it doesn't even get that far for me (linux3 compatibility)
[09:41:58] <Flameeyes> strange, I'm running linux3 as well (but this is still just x86-generic, arm-generic is to follow)
[09:43:27] <zbehan> there's some fairly freaky sightings on the topic of linux3 compatibility, some builds only fail on some linux3 machines and not other linux3 machines, like o3d or chrome, or now, nacl, while the failure is quite clearly something along the lines of "linux3: no such key"
[09:43:55] <Flameeyes> heh
[09:44:00] <zbehan> i didn't investigate further and fix a superset instead
[09:44:14] * Flameeyes really could slap Linus around a bit for bumping major without enough time for packagers to fix build systems :(
[09:46:43] <zbehan> Flameeyes: but... he was not comfortable counting so high! ;)
[09:47:18] <Flameeyes>  /var/lib/portage/distfiles-target/chrome-src/src/native_client/toolchain/linux_x86_newlib/bin/../libexec/gcc/x86_64-nacl/4.4.3/cc1plus: /usr/lib/gcc/x86_64-pc-linux-gnu/4.3.4/32/libstdc++.so.6: version `GLIBCXX_3.4.11' not found (required by /var/lib/portage/distfiles-target/chrome-src/src/native_client/toolchain/linux_x86_newlib/bin/../libexec/gcc/x86_64-nacl/4.4.3/cc1plus)
[09:47:18] <Flameeyes> this is my issue here :/
[09:47:30] <Flameeyes> zbehan: yeah well.. lovely how he didn't foresee this problem it seems :|
[09:48:06] <zbehan> ah, yeah, i think i saw that problem before ^
[09:50:22] *** ferringb_ has joined #chromium-os
[09:51:07] *** ferringb_ has joined #chromium-os
[09:51:07] *** ChanServ sets mode: +v ferringb_
[09:51:19] <Flameeyes> zbehan: any solution?
[09:51:34] *** ferringb has quit IRC
[09:51:42] *** ferringb_ is now known as ferringb
[09:52:39] <zbehan> Flameeyes: nope, that is the prebuilt nacl toolchain, i suppose you could try installing libstcd++, but i'm not sure if it fixes the problem
[09:53:10] <Flameeyes> that looks more like nacl built against gcc 4.4.3 when system gcc is still 4.3.4
[09:53:23] *** gfrog has quit IRC
[09:54:41] <zbehan> Flameeyes: well, nacl toolchain built against that, probably, but it matters little, as it's binary and downloaded
[09:55:38] <zbehan> Flameeyes: could try to get the right gcc version emerged on the host, that may be difficult though
[09:56:38] <Flameeyes> it's binary, downloaded, and you can't downgrade libstdc++ versions ;)
[09:56:56] <Flameeyes> so if you link something with gcc X.Y you can only execute it if the system's libstdc++ is from gcc >=X.Y
[09:57:21] <Flameeyes> gcc 4.4.6 is already installed in the system, just not selected
[09:57:21] *** ferringb has quit IRC
[09:57:27] <zbehan> Flameeyes: yup, btw. your chroot has both of them, right? wouldn't selecting gcc-4.4 manually with gcc-config save the day?
[09:57:56] <Flameeyes> re-executing build_packages with gcc 4.4 now
[09:58:28] <zbehan> yeah, a CL is up to fix the unsynced host toolchains, and we already pushed something to fix the new ones
[09:58:41] <zbehan> so on a new chroot, you only get 4.4 now
[09:58:50] <zbehan> (or should, that is)
[10:04:16] <crosbot> tree became 'Tree is open'
[10:10:12] *** gfrog has joined #chromium-os
[10:11:57] *** ferringb has joined #chromium-os
[10:11:57] *** ChanServ sets mode: +v ferringb
[10:22:47] *** TW1920 has joined #chromium-os
[10:31:51] *** patcito has quit IRC
[10:41:05] *** TW1920 has quit IRC
[10:41:26] *** D|sT has quit IRC
[10:42:45] *** TW1920 has joined #chromium-os
[10:43:00] *** D|sT has joined #chromium-os
[10:53:35] *** D|sT has quit IRC
[10:54:13] *** D|sT has joined #chromium-os
[10:59:43] *** razer has joined #chromium-os
[11:00:02] *** razer has quit IRC
[11:00:26] *** razer has joined #chromium-os
[11:01:45] <razer> Which one would be a better choice? The flow OS BY Hexxeh or the Chrome OS mod for DELL by Dell
[11:02:07] <razer> I have a Dell mini
[11:03:22] *** razer has quit IRC
[11:10:24] *** TW1920_ has joined #chromium-os
[11:14:45] *** TW1920 has quit IRC
[11:17:09] *** Hexxeh has quit IRC
[11:17:43] *** Hexxeh has joined #chromium-os
[11:27:18] *** Sarten-X has quit IRC
[11:33:13] *** Sarten-X has joined #chromium-os
[12:01:36] *** joaodasilva has joined #chromium-os
[12:01:37] *** ChanServ sets mode: +v joaodasilva
[12:02:14] *** kochi1 has quit IRC
[12:04:43] *** elly has quit IRC
[12:17:07] *** D|sT has quit IRC
[12:19:08] *** D|sT has joined #chromium-os
[12:21:20] *** octe has joined #chromium-os
[12:26:19] *** elly has joined #chromium-os
[12:36:09] *** quannnum has quit IRC
[12:36:51] *** quannnum has joined #chromium-os
[12:45:25] <BigWookie> definitely the DELL mod
[12:45:31] <BigWookie> flow is very outdated
[13:31:42] *** razer has joined #chromium-os
[13:33:35] *** razer has left #chromium-os
[13:56:48] *** gfrog has quit IRC
[14:04:24] *** gfrog has joined #chromium-os
[14:35:41] *** yoshiki_ has quit IRC
[14:36:07] *** yoshiki_ has joined #chromium-os
[14:57:51] *** eggy has quit IRC
[14:58:35] *** eggy has joined #chromium-os
[14:59:17] *** Sarten-X has quit IRC
[15:09:11] *** Sarten-X has joined #chromium-os
[15:13:45] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "x86-alex-binary" from d5f834b5ae8af41286ab210a73aa3c4f45451356: yurys at chromium dot org <yurys at chromium dot org@268f45cc-cd09-0410-ab3c-d52691b4dbfc>)'
[15:40:30] <BigWookie> :/ I know this is a very basic question, but I tried to include nano in my build, included it in chromiumos-overlay/chromeos-base/chromeos/chromeos-..... and moved the link to a higher revision,then build_packages and then build_image but it isn't included
[15:41:57] <BigWookie> (also included it in chromeos-dev)
[15:54:30] *** jujugre has quit IRC
[16:02:25] *** rbyers_ has joined #chromium-os
[16:02:37] *** unreal has quit IRC
[16:11:32] *** jujugre has joined #chromium-os
[16:18:24] <vapier> kliegs: pong
[16:19:27] <kliegs> vapier: I think we talked in chat about it.  was the make.conf USE flag stuff
[16:19:43] <kliegs> and how make.conf needed USE="${USE} ..."
[16:19:52] <vapier> k k
[16:19:54] *** unreal has joined #chromium-os
[16:20:43] *** vpalatin has quit IRC
[16:21:06] *** cros_ has joined #chromium-os
[16:21:09] *** ukai_ has quit IRC
[16:21:30] *** wfrichar has joined #chromium-os
[16:21:30] *** ChanServ sets mode: +v wfrichar
[16:21:31] *** wfrichar has quit IRC
[16:26:18] <ellyjones> hmm
[16:26:26] <ellyjones> did I break all the things?
[16:26:29] <redpig> last night?
[16:27:05] *** vpalatin has joined #chromium-os
[16:27:05] *** ChanServ sets mode: +v vpalatin
[16:29:15] <kliegs> ellyjones: If I say yes will you fix things?
[16:29:55] <ellyjones> kliegs: I'll fix some of the things
[16:30:03] <ellyjones> although I'm sheriff today so maybe I'd better fix all the things
[16:30:10] *** rbyers_ has quit IRC
[16:30:13] <kliegs> was just about to ask crosbot that :)
[16:30:25] <kliegs> alex binary is the only red?
[16:30:27] <ellyjones> crosbot: sheriffs?
[16:30:27] <crosbot> ellyjones: sheriffs: wfrichar, rhashimoto, ellyjones, kochi
[16:30:54] <ellyjones> x86-alex-binary is sick, as are a couple of random other bots
[16:31:55] <ellyjones> 08/30 06:08:48 ERROR|base_packa:0461| packages.checksum could not be fetched from any of the repos ['autoserv://']
[16:32:08] <ellyjones> as part of desktopui_UrlFetch.not-live
[16:33:39] <kliegs> ellyjones: I think I've been seeing that error in a few builds as a non-fatal
[16:33:47] <kliegs> like when I was looking at other errors it came along for the ride
[16:34:03] <ellyjones> well
[16:34:05] <ellyjones> that's the test that failed
[16:34:23] <kliegs> 06:09:01 INFO | 		ERROR	desktopui_UrlFetch.not-live	desktopui_UrlFetch.not-live	timestamp=1314709743	localtime=Aug 30 06:09:03	Unexpected web site title.  Expected: Hello World. Returned: about:blank
[16:34:35] <kliegs> I think that's the actual error
[16:35:08] <kliegs> it feels like there's been some flakiness/timing issues introduced to the vmtests lately to me.  but that's just anecdotal
[16:35:50] *** D|sT has quit IRC
[16:40:16] <ellyjones> hrm
[16:43:05] <crosbot> tree became 'Tree is closed (ellyjones: looking at x86-alex-binary)'
[16:44:56] <ellyjones> maybe we are relying on the HTTP server thread being ready immediately after we say .run() and it's not
[16:47:26] <kliegs> ellyjones: its possible. but I'm not really that familiar with how those bits work
[16:47:51] <kliegs> asking selfishly, can you open the tree so I can push?  last build of alex was green so flake seems likely
[16:48:25] *** petermayo has joined #chromium-os
[16:48:25] *** ChanServ sets mode: +v petermayo
[16:50:09] <ellyjones> I really do not like this "it is flaky, reopen the tree" idea
[16:50:54] <ellyjones> there, it's open
[16:51:02] <kliegs> ellyjones: i guess it depends why we close the tree
[16:51:02] <crosbot> tree became 'Tree is open (ellyjones: x86-alex-binary desktop_UrlFetch failure seems spurious)'
[16:52:49] <kliegs> ellyjones: if we're closing the tree to catch bad CL's and isolate what changes broke things, then flaky tests shouldn't necessarily close it
[16:53:15] <kliegs> however the people responsible for the flaky tests should get pressure put on them to unflake them
[16:53:38] <ellyjones> I filed a bug against nirnimesh at chromium dot org
[16:53:48] <ellyjones> who appears to be the one touching desktopui_UrlFetch
[16:54:12] <kliegs> if its an issue with the testframework itself then it can be tougher.  but should still get those as high priority items and make sure someones fixing
[16:54:39] <kliegs> although I've been tempted at times like you to keep the tree closed to force work on fixing flaky tests.  but not sure that's always worth the cost
[16:56:30] <petermayo> ellyjones: (from chump)  hmm warning about red tree not warning -- submitted a CL on mark as stable ... ping me if it gets in the way.
[16:57:38] <ellyjones> petermayo: what could possibly go wrong
[16:57:48] * kliegs whistles and walks away
[16:59:22] <petermayo> * coughs nervously *  nothing of course
[17:16:27] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "tegra2_kaen-binary" from ef7aab90b0ac2ed31fab60c4a280355682015869: apavlov at chromium dot org <apavlov at chromium dot org@0039d316-1c4b-4281-b951-d872f2087c98>)'
[17:19:41] <ellyjones> hmm.
[17:20:18] <ellyjones> that's not good!
[17:20:19] <ellyjones> >=app-admin/eselect-opengl-1.2.4 ~arm
[17:20:37] <ellyjones> petermayo: could your change be related to that at all?
[17:23:54] <crosbot> tree became 'Tree is closed (ellyjones: looking at tegra2_kaen-binary)'
[17:24:26] <ellyjones> vapier's changes, probably
[17:24:46] <vapier> i didnt keyword for arm to be hesitant
[17:24:52] <vapier> easy enough to fix
[17:24:53] <ellyjones> that would do it
[17:24:56] <ellyjones> please do :) thanks
[17:27:53] <vapier> http://gerrit.chromium.org/gerrit/6899
[17:28:18] <vapier> running `./build_packages --board=arm-generic` atm, but presumably that should fix it
[17:28:31] <ellyjones> you believe them to be stable on arm?
[17:28:32] <vapier> dep tree didnt abort, so that's a good thing
[17:28:44] <vapier> the changes should be independent of arch
[17:28:46] <ellyjones> alright
[17:29:05] <ellyjones> approved
[17:29:07] <vapier> i dont have any arm hardware to actually test ;)
[17:30:45] *** fabiand has joined #chromium-os
[17:30:57] <fabiand> good day.
[17:31:01] <fabiand> i just got a samsung chromebook
[17:31:11] <fabiand> and wondered if all bootable efi devices will be booted?
[17:33:39] <ellyjones> vapier: I have arm hardware over here but it's awkward to use
[17:33:56] <ellyjones> fabiand: in dev mode, if you it C-u, we boot USB
[17:34:20] <ellyjones> vapier: tree reopened, thank you :)
[17:34:25] <crosbot> tree became 'Tree is open (ellyjones: tegra2_kaen-binary fix in from vapier)'
[17:34:35] <fabiand> ellyjones: yep. I just tried a stock fedora live image, that does not get recognized.
[17:34:47] <fabiand> i expected something like this, as there is this script for installing ubuntu around ..
[17:34:59] <ellyjones> fabiand: on a USB stick? yeah, we only boot GPT, not regular partition tables
[17:36:33] <fabiand> ellyjones: would booting a efi-livedisk work?
[17:36:59] <fabiand> or is there some magic-gpt-variant used by chromeos?
[17:38:01] <ellyjones> we use 'plain gpt' as far as I know...
[17:38:09] <ellyjones> I have no idea what an efi-livedisk is, though
[17:43:23] <fabiand> ellyjones: there is a small image for uefi systems (http://ftp.uni-kl.de/pub/linux/fedora/linux//releases/15/Fedora/x86_64/os/images/efidisk.img) - i hoped that this could be booted, does not seem so ..
[17:43:36] <fabiand> maybe i get some more time to see why it does not work ..
[17:45:10] <rhashimoto> West coast sheriff reporting for duty.
[17:47:14] *** reveman has joined #chromium-os
[17:48:42] <fabiand> ellyjones: thanks for your suggestions. off for now, but i'll get back :)
[17:49:04] <ellyjones> later
[17:50:03] *** fabiand has quit IRC
[17:53:43] *** wfrichar has joined #chromium-os
[17:53:44] *** ChanServ sets mode: +v wfrichar
[18:01:05] <ellyjones> anyone familiar with building chromeos-chrome for tegra2? I tried a build just now and got this:
[18:01:08] <ellyjones> {standard input}:1711: Error: selected processor does not support `vshrn.i16 d0,q2,#8'
[18:01:11] <ellyjones> make: *** [c/Debug/obj.target/skia/third_party/skia/src/core/SkBitmapProcState.o] Error 1
[18:07:59] *** jujugre has quit IRC
[18:11:26] <ellyjones> east coast sheriff off to lunch :)
[18:13:20] *** [S]olet has quit IRC
[18:24:46] *** Solet has joined #chromium-os
[18:30:10] *** gfrog has quit IRC
[18:33:01] *** gfrog has joined #chromium-os
[18:36:37] *** lipsinV2 has quit IRC
[18:37:22] *** saintlou has joined #chromium-os
[18:37:23] *** ChanServ sets mode: +v saintlou
[18:45:14] <ellyjones> east coast sheriff back on duty
[18:47:04] *** Styx has quit IRC
[19:00:00] *** saintlou has quit IRC
[19:02:41] *** rbyers_ has joined #chromium-os
[19:07:27] *** dennisjeffrey has joined #chromium-os
[19:07:27] *** ChanServ sets mode: +v dennisjeffrey
[19:08:33] *** grundler has joined #chromium-os
[19:08:33] *** ChanServ sets mode: +v grundler
[19:11:42] *** rginda has quit IRC
[19:12:36] *** rginda has joined #chromium-os
[19:12:36] *** ChanServ sets mode: +v rginda
[19:35:08] *** unreal has quit IRC
[19:38:49] *** petermayo has quit IRC
[19:47:42] *** unreal has joined #chromium-os
[19:52:22] *** rbyers_ has quit IRC
[19:53:36] *** patcito has joined #chromium-os
[20:03:48] *** sergiu has joined #chromium-os
[20:03:48] *** sergiu has joined #chromium-os
[20:10:09] *** saintlou has joined #chromium-os
[20:10:20] *** ChanServ sets mode: +v saintlou
[20:19:21] *** srao has quit IRC
[20:19:36] *** srao has joined #chromium-os
[20:19:36] *** ChanServ sets mode: +v srao
[20:19:42] *** Styx has joined #chromium-os
[20:25:48] *** oshima has joined #chromium-os
[20:25:48] *** ChanServ sets mode: +v oshima
[20:33:21] *** Adys has quit IRC
[20:34:49] *** ferringb has quit IRC
[20:46:55] *** saintlou is now known as saintlou|away
[20:47:01] *** Adys has joined #chromium-os
[20:47:14] *** ferringb has joined #chromium-os
[20:47:14] *** ChanServ sets mode: +v ferringb
[20:50:04] <BigWookie> :/ can at least someone point me in the right direction? I have to start somewhere
[20:50:18] <ellyjones> the right direction for what?
[20:53:07] <BigWookie> i tried to follow the guide to add an ebuild to the image build, but it won't be included (nano), and my image won't connect to my devserver even though it works
[20:53:53] <BigWookie> i added app-editors/nano in the line after vim, and ran build_packages, and then build_image ( with the according parameters )
[20:54:44] <BigWookie> nano is in /build/x86-generic/usr/bin/ but not in the final image, I also incremented the revision for chromeos ebuild
[20:56:07] <BigWookie> Guide is this one http://goo.gl/RK8Ka , after it didn't work, i also copied the nano ebuild to the same directory where vim is located
[20:57:09] <cros_> Anyone here know how to build the chromium browser? Posted instructions are very unclear.
[20:58:17] <BigWookie> And about the devserver, it is listening to the right address and port, I can open it via browser ( except the /update link, which leads to a cherrypy error ), there is a new image avalible and update_engine_client -update returns an error :/
[20:59:09] <ellyjones> BigWookie: does the build number at the bottom left say the right date and such?
[20:59:32] <BigWookie> yep
[20:59:38] <BigWookie> and the devserver doesn't even log a request
[20:59:55] <BigWookie> (when I try to update, there are logs if I try with the browser)
[21:00:04] <msb__> BigWookie: you added nano to chromeos-base/chromeos and then uprevved it?
[21:00:36] <BigWookie> msb__: yes ( I moved the symlink to a new revision, i hope that was correct )
[21:02:46] *** rbyers_ has joined #chromium-os
[21:03:10] <msb__> BigWookie: emerge-$BOARD -ep chromeos | grep nano
[21:03:37] <BigWookie> no return
[21:05:06] <msb__> BigWookie: emerge-$BOARD -avp chromeos
[21:05:19] <msb__> BigWookie: emerge-$BOARD -vp chromeos
[21:06:52] <BigWookie> oh oh :S, masked because of inavlid token, i'll investigate
[21:09:03] <BigWookie> okay fixed that, now it is in the tree, rebuilding image to make sure it is included ( I tired to comment out a line in the string...
[21:09:16] *** saintlou|away is now known as saintlou
[21:10:41] <BigWookie> *tried, and thank you very much :) ( this should've been obvious to me ;_; )
[21:12:24] <BigWookie> now this (hopefully) leaves me only with the devserver problem
[21:22:45] <BigWookie> :O i just recognized, i don't have gmerge in my image, and the crosh dosn't recognize the shell command
[21:23:59] <BigWookie> but it is build with --withdev and shows the dev image box on the login screen
[21:24:34] <kliegs> BigWookie: I think --withdev is the default. and that the shell option has been removed from crosh
[21:25:31] <BigWookie> and gmerge?
[21:26:41] <kliegs> BigWookie: its often worth running --test when calling build_image
[21:26:42] <oshima> ellyjones: arm build must be fixed now?
[21:26:56] <kliegs> BigWookie: gmerge should be there but its a system command, not part of crosh
[21:26:59] <ellyjones> hm?
[21:27:12] <ellyjones> oshima: which arm build are you talking about?
[21:27:15] <oshima> ellyjones: tegra2
[21:27:17] <BigWookie> kliegs: I know but as root find /|grep gmerge returns nothing
[21:27:46] <oshima> ellyjones: i'm having compilation error in SkBitmapProcState_matrixProcs.cpp
[21:27:57] <ellyjones> oshima: me too
[21:27:57] <kliegs> BigWookie: hmm.. weird
[21:28:19] <kliegs> It's in /usr/local/bin/gmerge for me
[21:28:19] <oshima> ellyjones: so you still have this issue?
[21:28:23] <BigWookie> :/ and the devserver dosn't work
[21:28:30] <ellyjones> I had it this morning, I haven't tried to build since
[21:29:06] <oshima> ellyjones: got it.
[21:34:39] *** unreal has quit IRC
[21:46:14] *** fabiand has joined #chromium-os
[21:52:07] <fabiand> ellyjones: good evening.
[21:52:42] <fabiand> you responded a couple of hours ago, regarding what a chromebook expects from a boot medium.
[21:52:56] <ellyjones> fabiand: yeah, but I still have no clue :)
[21:53:17] <fabiand> :) ah okay - it sounded as if you had ;)
[21:55:41] * Flameeyes still can't build python on arm-generic.. time to re-setup-board
[22:05:59] <fabiand> so, does someone know what a chromebook is looking for - if i want to boot from an usb stick?
[22:07:27] *** rbyers_ has quit IRC
[22:08:20] *** rbyers_ has joined #chromium-os
[22:08:20] *** ChanServ sets mode: +v rbyers_
[22:17:42] <cros_> Anyone know how to build the chromium browser? I can't follow the instructions...very confusing.
[22:21:20] <rhashimoto> ellyjones, wfrichar: x86-alex-binary failed to update chrome.  The chrome tree is closed because their svn server failed.
[22:21:32] <ellyjones> rhashimoto: loss
[22:21:43] <wfrichar> 'k
[22:23:50] <wfrichar> fabiand: what do you mean?
[22:23:51] *** fabiand has left #chromium-os
[22:24:40] <rhashimoto> ellyjones, wfrichar: Okay to go pick up some lunch, ~15 min?
[22:24:58] <ellyjones> No! No lunch for you!
[22:25:07] <ellyjones> (yes, go :P)
[22:25:24] <rhashimoto> :-) Thanks!
[22:27:50] <wfrichar> I don't want to jinx it, but the you-know-what has been #00ff00 for a suspiciously long time.
[22:27:52] *** thieule has joined #chromium-os
[22:27:52] *** ChanServ sets mode: +v thieule
[22:28:36] <ellyjones> what is the you-know-what, wfrichar?
[22:30:04] <wfrichar> The T-R-E-E
[22:35:53] <ellyjones> what?
[22:36:04] <ellyjones> it's green
[22:36:08] <ellyjones> crosbot: tree?
[22:36:09] <crosbot> ellyjones: tree: Tree is open (ellyjones: tegra2_kaen-binary fix in from vapier)
[22:36:14] <ellyjones> that's what I see in the web ui too
[22:37:47] <rhashimoto> wfrichar, ellyjones: Back.
[22:38:23] <wfrichar> yes. It's been that way for much too long. It's unnatural.
[22:38:50] <wfrichar> It should have fallen over three times by now.
[22:38:54] <ellyjones> well
[22:38:58] <ellyjones> we can close it a bit if you want? :)
[22:39:03] <wfrichar> Heh.
[22:39:18] <wfrichar> I was thinking of maybe throttling it just to be sure it stays that way.
[22:49:42] <Flameeyes> can somebody confirm/deny whether python builds on arm-generic?
[22:52:00] <crosbot> tree became 'Tree is closed (Automatic: "cbuildbot" on "stumpy canary" from None: )'
[22:53:39] <ellyjones> wfrichar: you got your wish
[22:53:53] <wfrichar> DARN IT>
[22:54:18] <rhashimoto> Looks like the chromium SVN server died in the middle of the update.
[22:54:19] <wfrichar> We scared it. I knew I shouldn't have mentioned it. You don't mention a no-hitter.
[22:55:21] <crosbot> tree became 'Tree is open (chromium SVN server died)'
[23:02:16] *** rbyers_ has quit IRC
[23:11:14] *** petermayo has joined #chromium-os
[23:11:14] *** ChanServ sets mode: +v petermayo
[23:19:57] *** Sarten-X has quit IRC
[23:24:17] *** Solet has quit IRC
[23:30:01] *** Sarten-X has joined #chromium-os
[23:31:15] <njw> my builds of things using dbus-c++ are failing with this error: "dbusxx-xml2cpp: /usr/lib/gcc/x86_64-pc-linux-gnu/4.3.4/libstdc++.so.6: version `GLIBCXX_3.4.11' not found (required by dbusxx-xml2cpp)". Ring any bells?
[23:32:21] <ellyjones> remerge it on the host
[23:32:38] <ellyjones> we bumped libstdc++ but didn't rebuild packages built against the old one, I think
[23:33:25] <njw> ugh. don't we have upgrade scripts for this?
[23:34:06] <ellyjones> haha.
[23:34:15] <ellyjones> your expectation of childlike simplicity delights me
[23:35:29] <njw> Last time I broke something like this I had to write the damn script.
[23:40:49] <wfrichar> Must .. resist... saying.. "then it'll be that much simpler for you this time".
[23:40:58] <wfrichar> Oh, darn. Sorry. :-P
[23:41:17] *** cros_ has quit IRC
[23:42:02] <vapier> you shouldnt need to do that
[23:42:28] <vapier> run `sudo gcc-config 4.4.6`, then enter/exit the chroot
[23:43:49] <zbehan> argh, gotta push the host gcc upgrade script finally
[23:48:32] *** Justasic has quit IRC
[23:49:26] *** jaharkes has quit IRC
[23:53:16] *** vmil86 has quit IRC
[23:54:18] *** Justasic has joined #chromium-os





top