[00:01:59] *** Erkan_Yilmaz__ has quit IRC
[00:02:22] <trungl-bot> Tree opened by pkasting at chromium dot org: Tree is open (please use trybots)
[00:02:58] <evmar> i blame... the sea.
[00:03:12] <pkasting> I attack the darkness!
[00:06:37] <evmar> wow, trybots are totally hosed, huh?
[00:07:04] <eglaysher> 92 pending win jobs. 51 pending linux jobs
[00:08:21] <pkasting> OK, I'm going to try to check in my theme_resources.grd change without closing the tree.  I don't think it will break stuff.  If Windows breaks due to the change, and nothing else, I'll keep the tree open while watching to see if it greens up
[00:09:15] * rafaelw crosses fingers
[00:09:17] <thakis> is there a threadsafe ScopedRunnableMethodFactory?
[00:09:35] <pkasting> Woah wait a minute, tree seems fairly red
[00:09:42] <pkasting> I'm not sure I want to check in on a tree this red
[00:09:56] <pkasting> The whole Linux section looks sick.  Is someone looking at it?
[00:10:45] <ojan> layout tests on windows are very broken
[00:10:49] *** phajdan-jr|afk is now known as phajdan-jr
[00:10:53] <ojan> perhaps the tree should be closed
[00:10:56] <ojan> estade_laptop: ^^^
[00:11:00] <rafaelw> looking at linux...
[00:11:07] <ojan> hundreds of failures
[00:11:20] *** shoe has quit IRC
[00:11:45] *** shoe has joined #chromium
[00:12:11] <ojan> apatrick_: your patch is the only one on the blamelist
[00:12:43] <ojan> rafaelw, jam2: maybe we should close the tree?
[00:12:56] <rafaelw> ojan: agree.
[00:13:31] <trungl-bot> Tree closed by rafaelw at chromium dot org: Tree is closed  (please use trybots)
[00:15:11] <phajdan-jr> looking for another pair of eyes... does DownloadRequestLimiter sound like a good name to you? The current name is DownloadRequestManager which is awful, one of the 150 managers in chrome/browser/download.
[00:15:11] * evmar hopes he didn't check in another "rm -rf /" patch accidentally
[00:15:38] <estade_laptop> anyone around to tap apatrick_ on the shoulder?
[00:15:47] <awong> phanjdan-jr: sounds good if what it does is limit the number of download requests.
[00:16:01] <phajdan-jr> awong: that's exactly what it does. Thanks!
[00:17:02] <rafaelw> ojan: it's not clear to me who's to blame for the OpenDatabaseSyncInputs failures. Any chance it was the DEPS change?
[00:17:05] <akalin> evmar: did you actually check in an rm -rf patch?
[00:17:06] <ojan> apatrick_: is there any possibility the webkit failures are from r52390?
[00:17:21] <jamesr> WebKit (dbg) failed with 'update scripts failed'
[00:17:29] <jamesr> both (1) and (3)
[00:17:30] <evmar> akalin: yeah, a few years back
[00:17:37] <ojan> rafaelw: it's possible. it passed on the webkit canaries and the try servers though. my money is on evmar's change. i'm building locally now to try and reproduce. i think evmar is as well.
[00:17:38] <jamesr> doesn't look like it has anything to do with the patch
[00:17:44] <evmar> akalin: the buildbots all red, then black
[00:18:00] <akalin> nice
[00:18:18] <akalin> what did you intend to do
[00:18:49] <leiz> yikes, the try bot queue is 45/55/90 pending for lin/mac/win
[00:18:50] <awong> ojan: complete shot in the dark...is disk full on the builders?  I recall that can happen can cause various database tests to fail.
[00:19:23] <ojan> awong: maybe it's something like that. it's weird that all three windows debug builders would have problems at once
[00:20:01] <awong> ojan: not that weird...they all do run the same data-generating tests all at once.
[00:20:14] <ojan> actually...it looks like they are passing on the following run
[00:20:25] <awong> cool
[00:20:26] <ojan> cosmic rays?
[00:20:49] <ojan> maybe a butterfly flapped it's wings in thailand.
[00:21:15] <awong> that'd explain the remarkably beatufully sunny weather here in seattle.
[00:21:55] <evmar> i did just try M-x butterfly for my first time recently
[00:22:01] <evmar> akalin: it was just a bug
[00:22:06] <akalin> mmm
[00:23:14] *** udp has quit IRC
[00:24:35] <estade_laptop> does anyone sit near apatrick_ before i try reverting his change
[00:25:12] <ojan> estade_laptop: it looks like a fluke
[00:25:21] <ojan> estade_laptop: the bots seem to be passing on the following run
[00:25:48] <estade_laptop> ojan: which bot are you looking at
[00:25:49] <ojan> estade_laptop: which makes sense really...his change seemed like it couldn't possibly affect the layout tests
[00:25:58] <ojan> estade_laptop: webkit win dbg bots
[00:26:05] <rafaelw> phajdan-jr: compile failures on linux appear to be you
[00:26:10] <ojan> estade_laptop: there's just webkit win dbg2 left
[00:26:18] <estade_laptop> ok
[00:26:21] *** bent-mozilla has quit IRC
[00:26:30] <ojan> estade_laptop: did his change cause other regressions?
[00:26:44] <estade_laptop> no, I thought I was looking at linux webkit but it was win
[00:29:20] <rafaelw> evmar: what up? Is OpenDatabaseSyncInputs you?
[00:31:14] <phajdan-jr> rafaelw: looking
[00:31:52] <phajdan-jr> rafaelw: yeah, so sorry, will have a fix in a sec
[00:32:44] <trungl-bot> Tree closed by phajdan.jr at chromium dot org (:phajdan-jr): Tree is closed  (please use trybots) - phajdan.jr fixes compile, so sorry
[00:33:03] *** jrmuizel has quit IRC
[00:34:08] *** jrmuizel has joined #chromium
[00:35:39] *** jrmuizel has joined #chromium
[00:35:42] <jam2> pahjdan-jr: any reason to not use trybot?
[00:35:42] <phajdan-jr> rafaelw: fix in, not sure how this problem was not detected before.
[00:35:48] <rafaelw> ok. thanks.
[00:35:57] <phajdan-jr> jam2: I used one, and compile tested it locally. I'm not sure what happened.
[00:36:12] <jam2> pahjdan-jr: oh, i dont see trybot results here: m
[00:36:18] <jam2> http://codereview.chromium.org/2847050
[00:37:06] <ojan> hm...i reverted evmar's change locally and it still fails
[00:37:23] *** jrmuizel has quit IRC
[00:37:46] <phajdan-jr> jam2: I don't know, maybe indeed I screwed it up. I think I've submitted it at least once. Sorry for breaking the tree. :-/
[00:37:55] *** aroben has quit IRC
[00:38:00] *** rniwa_ has joined #chromium
[00:38:01] *** ChanServ sets mode: +v rniwa_
[00:38:07] *** rniwa has quit IRC
[00:38:07] *** rniwa_ is now known as rniwa
[00:38:26] <jam2> phajdan: i'm just picking on everyone today who breaks the build and doesnt have a change with trybot results ;)
[00:38:27] <rafaelw> ojan: The SyncDatabase test?
[00:38:40] <ojan> rafaelw: yeah. i'm trying revertting the webkit merge now.
[00:38:44] <rafaelw> ok.
[00:39:02] <ojan> oh, itneresting...this test was modified in the merge.
[00:39:05] <ojan> sigh.
[00:39:12] <ojan> why did this pass on the try bots and canaries?
[00:39:16] <pkasting> Hey C++ folks, try this on for size
[00:39:28] * rafaelw retracts his accusatory finger from evmar.
[00:39:34] <pkasting> ListValue node_tags;
[00:39:36] <ojan> evmar: sorry for the blaming.
[00:39:40] <pkasting> std::wstring tag;
[00:39:54] *** awolfson has quit IRC
[00:39:58] <pkasting> node_tags->Remove(StringValue(tag.substr(1)));
[00:40:14] <pkasting> Fails with "StringValue(const StringValue&) is private in this context"
[00:40:15] *** aroben has joined #chromium
[00:40:16] <pkasting> on Mac
[00:40:25] <pkasting> Oh, ListValue* node_tags, sorry
[00:40:29] <phajdan-jr> rafaelw: looks like my compile fix worked
[00:40:52] <pkasting> Relevant info: StringValue(const std::wstring&) is explicit, and StringValue has DISALLOW_COPY_AND_ASSIGN() on it
[00:40:55] *** rniwa has quit IRC
[00:41:01] <rafaelw> phajdan-jr: thanks.
[00:41:02] <pkasting> I have no explanation for this
[00:41:04] *** rniwa has joined #chromium
[00:41:11] *** ChanServ sets mode: +v rniwa
[00:41:14] <pcgod> motownavi: it's "possible" to set cmd-shift-forward delete as a key equiv in IB ... http://0xy.org/clear-browsing-data-shortcut.diff displays the correct shortcut in IB but I don't know if it works
[00:41:28] *** Erkan_Yilmaz__ has joined #chromium
[00:41:31] <pkasting> Note that splitting the line works: "StringValue temp(tag.substr(1)); node_tags->Remove(temp);" compiles fine
[00:42:07] <stuartmorgan> pkasting: you can't pass a reference to non-const temporary variables
[00:42:32] <pkasting> stuartmorgan: What
[00:42:53] <trungl-bot> Tree closed by phajdan.jr at chromium dot org (:phajdan-jr): Tree is closed  (please use trybots) - Linux compile fix in
[00:43:05] <stuartmorgan> pkasting: Remove is pass-by-reference, right?
[00:43:13] <pkasting> Remove takes a const ref
[00:43:17] <pkasting> Not a non-const ref
[00:43:22] <pkasting> I think
[00:43:44] <pkasting> Yes: int Remove(const Value&)
[00:43:52] *** agayev has left #chromium
[00:43:58] <froek> .
[00:44:16] <pkasting> So it seems like the StringValue() call should create a temp whose lifetime is the call
[00:44:27] <pkasting> the outer call, I mean
[00:44:38] <pkasting> And then the compiler supplies a const ref to that temp
[00:44:55] *** Erkan_Yilmaz_ has quit IRC
[00:45:01] <pkasting> But for some reason it's trying to copy the explicit temp to some sort of implicit temp to take its ref
[00:45:13] <pkasting> At least that's how I interpret it...
[00:46:46] <stuartmorgan> pkasting: Yeah, that does seem like it should work. I've just been bitten by porting Windows code that passed temps by non-const reference and assumed that was what was happening here
[00:49:27] *** eseidel has quit IRC
[00:49:28] *** huanr has joined #chromium
[00:49:28] *** ChanServ sets mode: +v huanr
[00:49:45] *** eseidel has joined #chromium
[00:50:14] *** awidegreen has quit IRC
[00:50:29] <ojan> rafaelw: http://codereview.chromium.org/2893017
[00:50:45] <rafaelw> ok.
[00:54:18] *** luxigo has quit IRC
[00:56:08] *** General1337 has quit IRC
[00:56:18] *** janm has quit IRC
[00:56:18] <stuartmorgan> pkasting: http://bytes.com/topic/c/answers/501746-gcc-needs-copy-consturctor-inplace-constructed-object-passed-func-reference-others-not
[00:57:22] <ojan> sigh...i don't get how we run the worker tests. this test started failing on the canary bots only after i committed the webkit merge.
[00:58:09] *** estes has quit IRC
[00:58:30] *** General1337 has joined #chromium
[00:59:04] *** mazda has joined #chromium
[00:59:13] *** estes has joined #chromium
[00:59:14] <pkasting> stuartmorgan: What I take away from that is "there is some reason in the standard that says that you can't guarantee the temp is not copied"
[00:59:20] <evmar> ojan: they pass locally, btw
[00:59:54] <pkasting> stuartmorgan: Which I don't understand at all... but C++ has lots of real hairiness when you open up the floorboards
[01:00:02] <stuartmorgan> pkasting: another site I found referred to http://www.open-std.org/Jtc1/sc22/wg21/docs/papers/2009/n2831.html but I'm having trouble parsing it into English
[01:00:14] <stuartmorgan> (referred to that as the reason)
[01:00:42] *** pdelgallego has quit IRC
[01:02:01] <pkasting> stuartmorgan: I think that's discussing C++0x, which is why I have trouble parsing it ("foo&&" as a type isn't syntax I understand)
[01:02:26] <phajdan-jr> update scripts fail suspiciously often
[01:02:41] <phajdan-jr> and it's a timeout :~/
[01:03:48] *** rdsmith has quit IRC
[01:04:01] <stuartmorgan> pkasting: you're right, I think I conflated another related thread
[01:04:04] <gavinp> i have been made unhappy by that paweE
[01:04:10] <gavinp> :set utf8
[01:04:19] <gavinp> err, pawel.
[01:05:38] <michaeln> ojan: you mean the worker layout tests?
[01:05:45] *** hbono has joined #chromium
[01:05:58] <ojan> michaeln: i htink i know why it didn't fail on the canaries...maybe.
[01:06:12] <ojan> michaeln: actually...no, i don't.
[01:06:44] <michaeln> we only run them in chrome via ui_tests.exe
[01:09:01] <dave_levin> ojan: re the worker test being disabled. It is only run during ui tests due to v8. Also, canaries only run release not debug.
[01:09:26] <ojan> dave_levin: i ran the ui tests on the canaries and the try servers.
[01:09:33] <ojan> dave_levin: also, it failed in both release and debug
[01:09:42] <dave_levin> ojan: how odd...
[01:09:47] <jamesr> phajdan-jr: think the svn connection is timing out?
[01:09:47] <ojan> dave_levin: in face, now that i look at it, it looks like the test was deleted entirely
[01:09:54] *** bmizeran_ has joined #chromium
[01:10:09] <ojan> dave_levin: are the workers test files synced somewhere else?
[01:10:15] <ojan> and maybe those are not kept at ToT?
[01:10:23] <dave_levin> ojan: Yes that is true.
[01:10:31] <ojan> that wouldn't explain the try servers, but it would explain the canaries
[01:10:42] *** bmizerany has quit IRC
[01:10:58] *** dumi has joined #chromium
[01:10:58] *** ChanServ sets mode: +v dumi
[01:11:52] <dave_levin> ojan: See the section in deps "We run these layout tests as UI tests.". Now I don't know why those aren't sync'ed to ToT but that does seem to be true on the canaries.
[01:11:53] <dumi> abarth: got a minute to talk about some v8 bindings?
[01:12:08] <abarth> dumi: sure
[01:12:13] <ojan> dave_levin: yeah...was just looking at that
[01:12:44] <phajdan-jr> jamesr: that's probable, but do we have a way to fix it?
[01:13:19] <jamesr> phajdan-jr: retry logic in gclient? wouldn't the log messages say that it was an svn timeout?
[01:14:08] <dumi> abarth: it's about the CRASH() instructions i added to CodeGeneratorV8, to check if v8 returns empty handles: http://trac.webkit.org/changeset/60754
[01:14:21] <phajdan-jr> jamesr: I don't know, here's an example log: http://build.chromium.org/buildbot/waterfall/builders/Chromium%20Linux%20Builder%20x64/builds/5790/steps/shell/logs/stdio
[01:14:21] <abarth> ok
[01:14:27] <dumi> abarth: we got some crashes that indicate that toV8() returns empty handles indeed
[01:14:50] <dumi> abarth: i asked mads about it, and he basically said "well, it happens sometimes" (because of environment errors, like OOM, not because of v8 bugs)
[01:15:14] <dumi> abarth: so i was wondering if it would be ok to remove those CRASH() instructions, and just return an error when that happens, to avoid the crash
[01:15:17] <jamesr> phajdan-jr: yeah. no way to know what timed out from that log
[01:15:26] <jamesr> dumi: how? OOM = crash
[01:15:32] <jamesr> dumi: in chromium anyway
[01:15:39] <jamesr> you can never OOM without crashing
[01:16:01] <abarth> in webkit too
[01:16:14] <abarth> there are some special allocations that can OOM without crashing
[01:16:16] <abarth> but very rare
[01:16:22] <dumi> jamesr: well, we do crash... is there anything that would make us crash before we get to the v8 bindings?
[01:16:39] <jamesr> it's really unsatisfying to say that it happens 'sometimes' without knowing when :\
[01:16:40] <abarth> depends when you OOM :)
[01:16:41] *** eseidel has quit IRC
[01:16:52] <dumi> jamesr: or you mean that toV8() would crash and not return empty handles if we're oom?
[01:16:53] <ojan> dave_levin: oh. i see the problem.
[01:17:07] <dave_levin> ojan: Cool.
[01:17:11] <jamesr> dumi: if we're OOM we should crash before returning anything from toV8()
[01:17:12] <ojan> not sure how to fix it
[01:17:17] <dumi> jamesr: i agree, but that's the answer i got from the v8 experts
[01:17:24] <thakis> rohitkc: make sure to talk with vangelis before you put too much time into the layer stuff
[01:17:32] <thakis> rohitkc: oops, wrong rohit
[01:17:42] <dumi> jamesr: well, they're not sure it's a OOM problem, but they know it can happen sometime, and apparently it's not a v8 bug or something they can fix
[01:17:54] *** rsesek has quit IRC
[01:18:05] <dave_levin> ojan: I don't see it at all.
[01:18:37] <ojan> dave_levin: the custom_deps for the canaries overwrites third_party/WebKit
[01:18:51] <ojan> dave_levin: not any of the directories for the ui test layout tests
[01:18:54] <dave_levin> oh
[01:19:04] <dumi> abarth: in any case, all these crashes seem to come from one single extension. it seems to me like it might be better to return an error instead of crashing.
[01:19:47] *** ebraminio has joined #chromium
[01:20:01] <ojan> dave_levin: a slow, but correct thing would be to put a full checkout lf LayoutTests in the directory the layout test look at (just on the UI test canaries of course)
[01:20:14] <dave_levin> ojan: Yes.
[01:20:16] <estade_laptop> maruel: we are getting lots of failed update steps on the bots
[01:20:25] <dave_levin> ojan: not too slow if using git
[01:20:35] <dave_levin> ojan: but windows has a had time with that.
[01:20:36] <ojan> i still don't understand why the try servers passed though
[01:20:42] <pkasting> estade_laptop: pinging nsylvain may be faster
[01:20:46] *** rohitkc has quit IRC
[01:20:53] <ojan> i think for the sake of the webkit canaries, that would be fine
[01:20:58] <dumi> jamesr, abarth: from mads:"The bottom line is that any V8 wrapper creation can fail if unlucky. Therefore we do need guarding code like this in the bindings and it is reachable."
[01:21:10] <estade_laptop> nsylvain: ^^
[01:22:28] *** markmentovai has quit IRC
[01:23:11] *** Erkan_Yilmaz_ has joined #chromium
[01:24:30] *** cbentzel has quit IRC
[01:25:53] <tony^work> hmm, looks like something is wrong with a git server?
[01:26:41] *** Erkan_Yilmaz__ has quit IRC
[01:30:36] *** Omen_20 has joined #chromium
[01:30:55] <ojan> where do the .gclient files for the buildbots live?
[01:31:02] <tony^work> evmar: can you fix the git server?
[01:31:05] <ojan> as in...in the repo
[01:31:08] <tony^work> evmar: it's codf17 that's not responding
[01:31:09] <abarth> dumi: that's super lame
[01:31:18] <abarth> dumi: so we have a bunch of untestable crashers
[01:31:50] <tony^work> evmar: e.g., I can't clone from codf17 (it times out)
[01:32:09] <tony^work> oh, there it goes
[01:32:17] <tony^work> it's just really slow and we timeout in the script after 60s
[01:32:17] <ojan> dave_levin: ugh. what i just suggested won't work either. we'd still need to list each individual directory in DEPS to tell it not to pull it.
[01:32:49] <tony^work> ojan: you just want to stop pulling crostools?
[01:33:14] <ojan> tony^work: no, i want that all the webkit code we check out be at the upstream ToT
[01:33:31] <ojan> tony^work: right now, the stuff we check out into chrome/test/data uses the revision in DEPS on the canaries
[01:33:43] <jamesr> dumi, abarth: yeah that's kind of ridiculous. we don't even know what sort of error states we're in when the creation fails?
[01:34:01] <tony^work> ojan: ok, you're fixing a different problem than me :)
[01:35:14] <dave_levin> ojan: I don't understand the canary bots and how they sync to ToT.
[01:35:27] <ojan> dave_levin: actually...i found a way better solution
[01:35:45] *** bmizerany has joined #chromium
[01:35:45] <ojan> if i can figure out where the .gclient files are stored or how they're generated...this should be really easy to fix
[01:36:37] *** cbentzel has joined #chromium
[01:36:38] *** bmizerany has quit IRC
[01:36:56] *** andybons has quit IRC
[01:36:57] *** eseidel has joined #chromium
[01:36:58] *** bmizerany has joined #chromium
[01:37:07] <dumi> abarth: we might be able to unittest them (simulate OOM errors, or whatever makes v8 return empty handles), but i don't think we can test those crashes in a layout test
[01:37:28] *** hclam has quit IRC
[01:37:42] <jamesr> OOM should be impossible
[01:37:48] <jamesr> if we're OOMing and not crashing that is a serious bug
[01:38:29] *** bmizeran_ has quit IRC
[01:38:40] <jamesr> where's the bug for this?
[01:39:03] *** eseidel has quit IRC
[01:41:04] *** agl has quit IRC
[01:41:13] <dumi> jamesr: http://code.google.com/p/chromium/issues/detail?id=42971
[01:41:45] <pkasting> estade_laptop: Apparently the problem is being worked on; a git server is dead and we have a git dependency
[01:41:58] <evmar> wow, we are awesome today
[01:42:04] <pkasting> estade_laptop: Complicating factor: nsylvain knows nothing about git
[01:42:14] <tony^work> good thing evmar is here!
[01:42:33] <pkasting> evmar: <Fry>Fix it fix it fix it fix it fix it fix it fix it!</Fry>
[01:42:44] <nsylvain> The git guys hinted that we should restart the server
[01:42:48] <nsylvain> but they are not there anymore
[01:43:00] <nsylvain> and i dont know if there is any handholding needed when restarting it
[01:43:03] <nsylvain> I might just try anyway
[01:43:36] <pkasting> Maybe one day a week the coders should just take the day off and let machines catch up and infrastructure repair work happen :D
[01:43:51] <pkasting> Stop trying to rebuild the plane in flight!
[01:44:19] <shess> crash the plane first.
[01:44:28] <nsylvain> rebooting it... we're going to lose altitude for the next 5 minutes, but hopefully we should glide long enough to survive this.
[01:48:52] *** luxigo has joined #chromium
[01:49:17] *** mazda has quit IRC
[01:51:55] *** mrossetti_ has joined #chromium
[01:52:13] *** mrossetti_ has quit IRC
[01:52:44] <trungl-bot> Tree closed by estade at chromium dot org: Tree is closed  (please use trybots) - git server rebooting so linux bots can work again
[01:53:34] *** yutak_home has joined #chromium
[01:53:34] *** ChanServ sets mode: +v yutak_home
[01:53:53] *** bweinstein has quit IRC
[01:55:37] *** mrossetti has quit IRC
[01:56:30] *** GeekShadow has quit IRC
[01:57:19] *** akalin has quit IRC
[01:59:16] *** hbono has quit IRC
[01:59:16] *** jshin has quit IRC
[02:00:14] * awong hates shared libraries.
[02:00:58] <pkasting> WTF does "Error: CapabilityDisabledError; please report!" in rietveld mean?
[02:01:16] <tony^work> I don't know, but I'm getting 500 errors when trying to upload a change
[02:01:16] <pkasting> Does it mean "Fuck you if you want to get any work done today"?  Cause that's how I'm beginning to feel
[02:02:24] *** ace__ has quit IRC
[02:02:33] <phajdan-jr> pkasting: yeah, I wonder if we can make appengine-based services more reliable; similar things also happen with tree status app
[02:03:27] *** Erkan_Yilmaz__ has joined #chromium
[02:03:27] <jamesr> pkasting: that's a quota or a transient backend error
[02:03:38] <jamesr> pkasting: maybe we should use bugzilla for code reviews
[02:03:39] * jamesr ducks
[02:03:47] <pkasting> I've gotten it on every one of my retry attempts
[02:03:48] <nsylvain> pkasting, etc: git server is back up. seems more snappy now
[02:04:02] <pkasting> jam2: Are we over some app engine quota?
[02:04:06] <jam2> tony^work: app enigne is going read only for an hour
[02:04:12] <jamesr> gaaah
[02:04:14] <jam2> scheduled maintainence
[02:04:17] <jamesr> why the hell do they do this in the middle of the workday
[02:04:19] <jam2> great timing of course
[02:04:22] <jamesr> well, i guess not technically the middle
[02:04:32] <tony^work> jam2: that's good to know
[02:04:34] *** nvsbl has joined #chromium
[02:04:40] * tony^work switches branches and works on something else
[02:04:58] <pkasting> FUCK
[02:05:03] <pkasting> FUCK YOU APP ENGINE
[02:05:26] <pkasting> Seriously I am so damn sick of that incompetent product.  It's probably not the engineering team's fault, they probably have like 2 people
[02:05:28] *** cbentzel has quit IRC
[02:06:23] *** erikkay has quit IRC
[02:06:33] *** Erkan_Yilmaz_ has quit IRC
[02:06:35] *** dmaclach has quit IRC
[02:07:20] *** Kernel-Panic has joined #chromium
[02:10:27] *** mazda has joined #chromium
[02:10:39] *** bweinstein has joined #chromium
[02:11:28] <oshima> i'm getting server error when i reply on code reviews site. may be same issue as pkasting had?
[02:11:41] <tony^work> oshima: see jam2's comment a few minutes ago
[02:11:44] <pkasting> oshima: Yes, app engine is read only for an hour
[02:11:59] <pkasting> No code review comments or issue creations/deletions/changes for an hour
[02:12:02] <oshima> sigh...
[02:12:25] *** johnny_g has joined #chromium
[02:12:58] *** RT|Chatzilla has joined #chromium
[02:13:58] <trungl-bot> Tree closed by erg at chromium dot org (:eglaysher): Tree is closed  (please use trybots) - git server rebooting so linux bots can work again - appengine down for maintenance
[02:15:02] *** leeight has joined #chromium
[02:15:03] *** leeight has left #chromium
[02:16:05] <abarth> evmar: is there a way in git to search for a string that was removed at some point?
[02:16:17] <abarth> evmar: i'd like find the rev when it was removed
[02:17:02] *** huanr has quit IRC
[02:18:11] *** jcivelli has quit IRC
[02:19:10] <stevenjb> abarth: I've never used it, but you might take a log at git log -S
[02:19:23] <abarth> stevenjb: thanks
[02:19:49] <davidben> There's the pickaxe: git log -Syourstringhere which give you all commits that changed the number of times that string occurred (not quite the same as appearing in the diff). If what you're looking for is more inline with blame/annotate, look at git blame --reverse.
[02:21:29] <dcheng> Quick question: What's chrome/browser/gtk/custom_drag.h used for? It says it's used for programatically initiated drags, but I can't find a location that actually instantiates and uses it?
[02:21:30] *** awolfson has joined #chromium
[02:23:24] <dcheng> Oh never mind... didn't see the subclasses further down in the header.
[02:25:56] *** Gina has joined #chromium
[02:26:54] <kermit> when, if ever, does browsing history expire?
[02:27:18] <Gina> anyone know if there is a chrome browser binary that I can run on my arm device running Ubuntu, or is it just for x86?
[02:27:18] *** akalin has joined #chromium
[02:27:47] *** hbono has joined #chromium
[02:28:33] <oshima> does anyone know what this message means (ths is from chromeos ui_tests)
[02:28:55] <oshima> <window action="destroy" windowid="0" parent="0" session="0" time="1278616726"/
[02:28:55] <oshima>                                                                                ^
[02:28:55] <oshima> error : xmlTextWriterWriteDocCallback : XML error 5 !
[02:28:58] *** mazda1 has joined #chromium
[02:29:01] *** mazda has quit IRC
[02:29:35] <oshima> I asked this a while ago in the chromium-dev mailing-list, but got no reply.
[02:30:22] *** stevenjb has quit IRC
[02:30:34] <oshima> this is not causing test failure, but makes me uneasy.
[02:30:38] <oshima> anyone?
[02:30:49] *** Omen_20 has quit IRC
[02:31:26] <eglaysher> Gina: chrome compiles on arm, though we don't do arm releases yet. Try the ubuntu or debian builds?
[02:32:55] *** seventh has joined #chromium
[02:33:22] <Gina> eglaysher: thank you ... do you mean from here: http://www.google.com/chrome/eula.html?platform=linux&hl=en&hl=en&platform=linux&hl=en
[02:34:04] *** asargent has left #chromium
[02:34:15] <Gina> eglaysher: I have Chrome running on my arm device with Chrome OS ...  should the same Chrome binary work on my arm device with Ubuntu?
[02:36:27] <phajdan-jr> oshima: maybe at some point invalid XML is being produced? It seems it could have missing terminating ">".
[02:39:28] <oshima> phajdan-jr: yes, but my question is rather "what is this xml for?".
[02:39:29] <pkasting> nsylvain: If the git server is up, do we need to kick all the build bots to make them retry?
[02:39:56] <nsylvain> did not notice the tree was still closed
[02:40:06] <tony^work> seems like someone can just check in a change
[02:40:08] <nsylvain> I guess it should fix itself once someone commit. I can try
[02:40:11] *** stevenjb has joined #chromium
[02:40:11] <nsylvain> ok
[02:40:12] <nsylvain> yeah
[02:40:32] <nsylvain> yeah
[02:40:37] <nsylvain> it's going to turn green
[02:40:40] <oshima> phajdan-jr: do you know what is this xml for?
[02:42:03] <oshima> phajdan-jr: may be i should have said: "everytest has this message"
[02:42:24] <oshima> so this is something to do with test framework rather than individial test
[02:43:02] <oshima> actually im' wrong.
[02:43:22] *** Omen_20 has joined #chromium
[02:43:32] <oshima> but there are several across tests. at least it looks like that.
[02:45:13] *** Kernel-Panic has quit IRC
[02:45:33] *** stevenjb has quit IRC
[02:45:46] *** stevenjb has joined #chromium
[02:46:27] <phajdan-jr> oshima: well, I don't really know then
[02:47:06] *** Kernel-Panic has joined #chromium
[02:48:52] <oshima> phajdan-jr: ok, i'll ask around.
[02:48:53] *** eseidel has joined #chromium
[02:49:18] <hbono> oshima: Sorry for my off-the-topic question in advance.  I was not able to find the e-mail about your question in the chromium-dev ML <http://groups.google.com/a/chromium.org/group/chromium-dev/topics>. Is it possible to give me the URL?
[02:49:37] *** MikeSmithW3C has quit IRC
[02:50:04] *** morrita has quit IRC
[02:50:59] *** Kernel-Panic has quit IRC
[02:51:03] <oshima> hbono: doh, seems like i sent to chromium-os-dev. my bad.
[02:51:50] *** AaronMT has quit IRC
[02:53:18] *** nvsbl has quit IRC
[02:53:46] *** Kernel-Panic has joined #chromium
[02:54:44] <hbono> oshima: no problem. :)
[02:54:55] *** morrita has joined #chromium
[02:54:55] *** ChanServ sets mode: +v morrita
[02:55:29] <pkasting> nsylvain: So, someone should commit something to trip a build?
[02:56:54] <pkasting> OK, committed a whitespace change, let's see what happens
[02:57:28] *** eroman has joined #chromium
[02:57:28] *** ChanServ sets mode: +v eroman
[03:00:05] *** tedoc2000 has quit IRC
[03:04:40] <thakis> xcode is so crappy
[03:06:08] *** nvsbl has joined #chromium
[03:06:32] <trungl-bot> Tree closed by pkasting at chromium dot org: Tree is closed (waiting to see if it turns green)
[03:06:55] <pkasting> Git server reboot is complete, app engine is back up, tree looks like it may at least partially come back online
[03:07:09] <pkasting> Not sure about all the busted "release full" bots though
[03:10:19] *** til_ has joined #chromium
[03:10:19] *** til has quit IRC
[03:10:19] *** til_ is now known as til
[03:12:05] *** yutak_home has quit IRC
[03:13:14] <rafaelw> pkasting: they look like they are coming back.
[03:17:05] *** sergeyu has quit IRC
[03:19:48] <pkasting> rafaelw: Well, they were operating, they were just failing tests.  We have to wait to see what happens.
[03:20:29] <pkasting> Wondering if I should opportunistically commit my grd change now :P
[03:21:12] <jamesr> pkasting: oh what a jokester
[03:22:44] <jamesr> pkasting: according to the waterfall your whitespace change broke a test on snow leopard
[03:23:01] *** trungl has quit IRC
[03:23:21] <pkasting> NOOOOOO I should have trybotted it!!
[03:23:36] *** rniwa has quit IRC
[03:24:00] <dumi> question about worker tests: i have 2 tests, that both pass when run separately. if i put run them in the same TEST_F() test though, the execution is blocked on GetActiveTab() after the second test (the output looks good for both of them). any idea why this might be happening?
[03:24:24] *** cying has joined #chromium
[03:24:55] <phajdan-jr> dumi: do you have the code somewhere? I can take a look at your desk if you prefer
[03:25:11] <dumi> phajdan-jr: the code is not submitted yet
[03:25:11] *** slightlyoff has quit IRC
[03:25:46] *** leeight has joined #chromium
[03:26:00] <dumi> phajdan-jr: it doesn't do anything weird though (as far as i can tell): set up the test directories, run the tests
[03:26:11] *** nvsbl has quit IRC
[03:26:49] *** leeight has left #chromium
[03:28:43] *** phajdan-jr is now known as phajdan-jr|afk
[03:29:56] *** up365 has quit IRC
[03:31:44] <akalin> oh god
[03:31:52] <akalin> ld now fails for me even with enable_svg=0 D:
[03:32:22] <dumi> phajdan-jr|afk: nvm, looks like i was initializing dirs and what not for every single test, and i should've done it only once
[03:33:36] <pkasting> akalin: It was only ever going to be a temporary fix anyway...
[03:34:08] *** nvsbl has joined #chromium
[03:34:36] *** cying has quit IRC
[03:35:04] <pkasting> sheriffs: It looks like the Mac startup perf regression is probably real
[03:35:45] <pkasting> Actually, are any of the sheriffs even around anymore?
[03:35:50] <rafaelw> i'm here,
[03:35:53] <rafaelw> i'm looking.
[03:36:06] <rafaelw> it see intl2 failed then dhtml
[03:36:33] <pathorn> pkasting: there is a noticable startup regression here:
[03:36:36] <pathorn> http://build.chromium.org/buildbot/perf/mac-release-10.5/startup/report.html?history=150&rev=-1&graph=several_tabs
[03:37:10] <pathorn> the 10.5 bot seems to gradually increase over time, if you look at a long history, the bot was increasing until the machine rebooted I think two weeks ago
[03:37:24] <pkasting> rafaelw: Wrong bot, look at mac 10.5 perf (2)
[03:37:53] <rafaelw> ok. with you now.
[03:39:16] <rafaelw> are you looking at the bump around r52111?
[03:39:52] *** MikeSmithW3C has joined #chromium
[03:40:18] <pkasting> I'm just looking at the bot -- it's been consistently red several times in a row.
[03:40:38] <pkasting> I don't understand the perf graphs well enough to know which lines I'm supposed to be looking at where the bot is complaining.
[03:40:42] <pkasting> chase: yt?
[03:40:48] <rafaelw> nor i =-(
[03:41:17] <pathorn> http://build.chromium.org/buildbot/perf/mac-release-10.5/startup/report.html?history=350&rev=-1&graph=warm&trace=t&trace=t_ref
[03:41:31] *** eseidel has quit IRC
[03:42:01] <pkasting> Wow
[03:42:08] <pathorn> I talked with chase about this a few weeks ago when I was working on the new startup tests
[03:42:14] <pathorn> the mac 10.5 bot has been acting up though
[03:42:18] <pkasting> So, looks like the bot is getting slower in general, but there also is some real regressing going on
[03:42:33] <pkasting> (since t grows faster than t_ref)
[03:43:04] <pathorn> if you see here, it shows the problem that happened before the last reboot, so you have to keep that in mind:
[03:43:04] <pathorn> http://build.chromium.org/buildbot/perf/mac-release-10.5/startup/report.html?history=600&rev=-1&graph=warm&trace=t&trace=t_ref
[03:43:26] <pkasting> The bots started complaining around r52409, where it does look like things might have an issue
[03:43:29] <pathorn> so yeah ideally if you could take t/t_ref or something to account for that
[03:43:49] <pkasting> Woah
[03:44:17] <pkasting> That still suggests to me that there's some real regression here, but it seems hard to identify
[03:45:07] <pkasting> pathorn: Are you looking into this issue in general still?
[03:45:26] <pkasting> I kinda don't think the tree should be clsoed for this, but I don't really want it to just sit red either...
[03:45:36] <pathorn> no I'm not
[03:45:52] *** leeight has joined #chromium
[03:45:54] *** leeight has left #chromium
[03:46:06] <pathorn> I've just been monitoring the graphs a bit, and as far as I know 10.5 Perf(2) is the only bot with the proiblem
[03:46:59] <rafaelw> i have to go otherwise my car gets locked in the garage. i'll be back online at home shortly.
[03:47:03] *** rafaelw has left #chromium
[03:47:24] *** johnny_g has quit IRC
[03:49:57] *** up365 has joined #chromium
[03:50:36] *** phajdan-jr|afk has quit IRC
[03:54:31] *** nvsbl has quit IRC
[03:55:01] <trungl-bot> Tree opened by pkasting at chromium dot org: Tree open (don't submit without green trybot run! Mac 10.5 Perf -> ?)
[03:55:18] <pkasting> Please do not all kill the tree at once...
[03:58:03] *** stuartmorgan has quit IRC
[03:59:06] *** nvsbl has joined #chromium
[03:59:20] *** csilv has left #chromium
[04:01:27] *** SRabbelier has quit IRC
[04:02:57] *** fqian has quit IRC
[04:04:36] *** pathorn has quit IRC
[04:05:19] *** dr_win has quit IRC
[04:08:26] *** MikeSmithW3C has quit IRC
[04:13:28] *** rafaelw2 has joined #chromium
[04:14:45] *** bweinstein has quit IRC
[04:18:02] *** rafaelw2 has quit IRC
[04:18:34] *** ebraminio has quit IRC
[04:19:21] *** peavey has quit IRC
[04:19:31] *** MikeSmithW3C has joined #chromium
[04:20:50] *** nvsbl has quit IRC
[04:23:53] <oshima> updates on linux bots are failing again.
[04:23:58] <oshima> :(
[04:24:42] *** estes has quit IRC
[04:25:40] <oshima> nsylvain: ping
[04:26:13] <nsylvain> yeah
[04:26:15] <nsylvain> coming back now
[04:26:22] <nsylvain> we had to restart the git server again
[04:26:32] <oshima> i see. thanks
[04:27:27] *** estes has joined #chromium
[04:28:05] *** jrmuizel has joined #chromium
[04:30:50] *** peavey has joined #chromium
[04:33:17] *** xji has quit IRC
[04:33:21] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Webkit Builder (dbg)" from 52425: wtc at chromium dot org)
[04:34:34] *** eseidel has joined #chromium
[04:37:03] <thakis> is wtc around?
[04:37:19] <thakis> Error executing rc.exe (tool returned code: 1)
[04:37:22] <thakis> probably not his fault
[04:37:48] <thakis> maruel, nsylvain: ^ bot problem?
[04:37:48] *** til has quit IRC
[04:38:02] *** til has joined #chromium
[04:38:13] *** SRabbelier has joined #chromium
[04:39:52] <leiz> thakis: how about I submit a trivial CL and see what happens
[04:40:03] <thakis> leiz: sgtm
[04:40:05] <leiz> I think it's just VC crappiness
[04:40:08] <thakis> yeah
[04:40:57] <leiz> r52429
[04:42:56] <leiz> try bot pending queue is 0/36/109 for L/M/W
[04:43:20] <leiz> we'll see if the bot goes green in ~5 min
[04:44:13] *** luxigo has quit IRC
[04:46:28] *** malavv has quit IRC
[04:47:26] <trungl-bot> Tree closed by thakis at chromium dot org: Tree is closed (Automatic: "compile" on "Webkit Builder (dbg)" from 52425: wtc at chromium dot org -> flake? rc compiler error)
[04:50:01] *** tonyg-cr has quit IRC
[04:51:31] *** dumi has quit IRC
[04:53:40] <leiz> green
[04:54:30] <trungl-bot> Tree opened by thestig at chromium dot org (:leiz): Tree open (don't submit without green trybot run! Mac 10.5 Perf -> ?)
[04:54:49] *** eseidel is now known as eseidel|food
[04:56:02] *** rubenbb has joined #chromium
[04:59:31] *** jrmuizel has quit IRC
[05:01:52] *** Gina has quit IRC
[05:03:55] *** asargent2 has joined #chromium
[05:03:55] *** ChanServ sets mode: +v asargent2
[05:05:42] *** peavey has quit IRC
[05:09:42] *** rubenbb has quit IRC
[05:12:02] *** jamesr has quit IRC
[05:15:42] *** jrmuizel has joined #chromium
[05:17:53] *** peavey has joined #chromium
[05:18:29] *** rubenbb has joined #chromium
[05:18:53] *** tonikitoo has joined #chromium
[05:23:09] *** dumi has joined #chromium
[05:23:09] *** ChanServ sets mode: +v dumi
[05:30:32] *** eseidel has joined #chromium
[05:33:10] *** trungl has joined #chromium
[05:33:10] *** ChanServ sets mode: +v trungl
[05:34:09] *** eseidel|food has quit IRC
[05:37:01] *** awolfson has quit IRC
[05:37:01] *** Kernel-Panic has quit IRC
[05:38:00] *** Kernel-Panic has joined #chromium
[05:38:04] *** loislo has joined #chromium
[05:40:56] *** viro101 has joined #chromium
[05:43:40] *** eseidel_ has joined #chromium
[05:43:40] *** eseidel has quit IRC
[05:43:40] *** eseidel_ is now known as eseidel
[05:43:52] *** dr_win has joined #chromium
[05:46:30] *** eseidel_ has joined #chromium
[05:46:30] *** eseidel has quit IRC
[05:46:33] *** eseidel_ is now known as eseidel
[05:47:11] *** estes has quit IRC
[05:47:55] *** eseidel_ has joined #chromium
[05:47:55] *** eseidel has quit IRC
[05:47:59] *** eseidel_ is now known as eseidel
[05:53:23] *** bandu has joined #chromium
[05:59:58] *** Inc has joined #chromium
[06:00:01] <Inc> sooo
[06:00:23] <Inc> pretty good crash.... http://pastebin.ca/1900829
[06:03:43] *** viro101 has quit IRC
[06:04:58] *** bmizerany has quit IRC
[06:09:12] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Mac10.5 Tests (dbg)(1)" from 52438: shess at chromium dot org)
[06:10:10] <shess> looking.
[06:13:14] <trungl-bot> Tree opened by shess at google dot com: Tree is open (shess: believe Mac unit_tests is flake; don't submit without green trybot run! Mac 10.5 Perf -> ?)
[06:23:03] *** yusukes has joined #chromium
[06:23:40] *** bers has joined #chromium
[06:24:58] *** cuba33ci has joined #chromium
[06:26:00] *** johnny_g has joined #chromium
[06:31:10] *** asargent2 has left #chromium
[06:32:10] *** dumi has quit IRC
[06:32:38] *** johnny_g has quit IRC
[06:36:44] *** loislo has quit IRC
[06:38:11] *** bandu is now known as coyo
[06:39:35] *** Kernel-Panic has quit IRC
[06:47:15] *** cuba33ci has quit IRC
[06:48:30] *** cuba33ci has joined #chromium
[06:49:25] *** Buglouse has quit IRC
[06:49:37] <trungl-bot> Tree opened by shess at google dot com: Tree is open (don't submit without green trybot run! Mac 10.5 Perf -> ?)
[06:55:41] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Linux Builder (ChromiumOS)" from 52444: brettw at chromium dot org, jar at chromium dot org)
[06:56:42] <trungl-bot> Tree closed by brettw at chromium dot org: Tree is closed (Brett looking at ChromeOS issue)
[06:57:42] <trungl-bot> Tree opened by brettw at chromium dot org: Tree is open (Brett fixing trivial ChromeOS issue)
[07:00:04] *** michaeln has quit IRC
[07:01:45] <trungl-bot> Tree opened by brettw at chromium dot org: Tree is open (Brett should have fixed ChromeOS/Linux Views/Arm compile)
[07:04:34] *** BryanWB has joined #chromium
[07:05:24] *** jrmuizel has quit IRC
[07:07:37] *** tommi2 has joined #chromium
[07:08:23] *** BryanWB_ has joined #chromium
[07:09:07] *** magn3ts has joined #chromium
[07:09:20] *** tommi has quit IRC
[07:10:24] *** estes has joined #chromium
[07:11:57] *** tommi2 has quit IRC
[07:12:38] *** tommi has joined #chromium
[07:30:59] *** magn3ts has quit IRC
[07:32:41] *** loislo has joined #chromium
[07:35:50] *** estes has joined #chromium
[07:36:59] *** BryanWB_ has quit IRC
[07:42:54] *** roc has quit IRC
[07:44:58] *** thakis_ has joined #chromium
[07:44:59] *** ChanServ sets mode: +v thakis_
[07:48:04] *** stevenjb has quit IRC
[07:48:09] *** bers has quit IRC
[07:51:31] *** BryanWB has quit IRC
[07:54:30] *** MikeSmithW3C has quit IRC
[07:54:55] *** MikeSmithW3C has joined #chromium
[07:59:46] *** victorw has left #chromium
[08:02:23] *** MichealH has joined #chromium
[08:04:18] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Builder (dbg)" from 52449: akalin at chromium dot org, jshin at chromium dot org)
[08:05:55] <thakis_> webcore.lib(npruntime.obj) : fatal error LNK1318: Unexpected PDB error; UNKNOWN (24) ''
[08:06:04] <thakis_> flake?
[08:07:19] <trungl-bot> Tree opened by thakis at chromium dot org: Tree is open ("Chromium Builder (dbg)" looks like a flake)
[08:07:45] *** akalin2 has joined #chromium
[08:07:48] <akalin2> greetings
[08:07:58] <akalin2> i looked at the compile failure and i don't think it's me
[08:08:03] <akalin2> i'm not sure if it's jshin's change either, though
[08:08:16] <akalin2> does the compile error look familiar to anyone?
[08:08:25] <akalin2> http://build.chromium.org/buildbot/waterfall/builders/Chromium%20Builder%20(dbg)/builds/22255
[08:09:30] <akalin2> thanks thakis!
[08:11:41] *** hrna has joined #chromium
[08:23:58] *** dr_win has quit IRC
[08:27:29] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Linux Tests (dbg)(2)" from 52447: akalin at chromium dot org)
[08:27:45] <akalin2> whattt
[08:28:15] <akalin2> is this another flake
[08:37:38] *** stevenjb has joined #chromium
[08:43:35] <trungl-bot> Tree opened by akalin at chromium dot org: Tree is open ("Linux Tests (dbg)(2)" looks like a similar flake to the last one)
[08:45:18] *** Erkan_Yilmaz__ has quit IRC
[08:45:52] *** [Nef] has quit IRC
[08:46:00] *** kbrosnan has joined #chromium
[08:47:02] *** estes has quit IRC
[08:48:55] *** stevenjb has quit IRC
[08:51:29] <jochen__> good morning
[08:56:05] *** MichealH has quit IRC
[08:56:10] <abarth> jochen__: morning
[08:56:56] *** Zaba has quit IRC
[08:59:32] <jochen__> 60 tryjobs pending for win
[08:59:39] *** pdelgallego has joined #chromium
[08:59:41] <jochen__> so basically, no tryjobs for me today :(
[08:59:53] *** FullFlannelJacke has quit IRC
[09:00:34] *** Zaba has joined #chromium
[09:04:27] *** FullFlannelJacke has joined #chromium
[09:16:21] *** ph0nk has quit IRC
[09:21:39] *** star-affinity has joined #chromium
[09:22:17] *** eseidel has quit IRC
[09:25:20] *** FullFlannelJacke has quit IRC
[09:27:04] *** dave_levin has quit IRC
[09:34:41] *** maikmerten has joined #chromium
[09:36:34] *** pfeldman has joined #chromium
[09:41:10] *** eseidel has joined #chromium
[09:41:24] *** FullFlannelJacke has joined #chromium
[09:41:24] *** leeight has joined #chromium
[09:41:29] *** leeight has left #chromium
[09:51:22] *** General13372 has joined #chromium
[09:52:09] *** General1337 has quit IRC
[09:55:48] *** bauerb has joined #chromium
[09:55:48] *** ChanServ sets mode: +v bauerb
[09:59:58] *** dave_levin has joined #chromium
[10:00:16] <dave_levin> trungl-bot: help
[10:00:16] <trungl-bot> dave_levin: The help command is "halp" (blame thakis for this).
[10:00:26] <dave_levin> trungl-bot: halp
[10:00:27] <trungl-bot> dave_levin: Valid commands (use "halp <command>" for more): bug, commit, commits, cookie, fortune, halp, help, lkgr, offices, settreestatus, time, treestatus, uptime, webkitbug, webkitcommit, webkitcommits, whois. (If you whisper to me, I'll whisper back.)
[10:00:37] <dave_levin> trungl-bot: lkgr
[10:00:39] <trungl-bot> dave_levin: 52431
[10:01:18] *** [Nef] has joined #chromium
[10:01:18] *** dave_levin has quit IRC
[10:02:50] *** Imperator_ has left #chromium
[10:11:18] <tyoshino> hmm. mac official is failing on compilation for about one week?
[10:15:29] *** elmargol has quit IRC
[10:16:13] *** elmargol has joined #chromium
[10:17:43] <tyoshino> trying clobber.
[10:22:16] <jochen__> tyoshino: it runs out of memory
[10:28:38] <tyoshino> jochen_: thanks. do you know if anyone is already working on that?
[10:30:05] <dcheng> I thought the fix was to move to building on 10.6?
[10:30:08] *** hrna has quit IRC
[10:30:15] <jochen__> yes
[10:37:08] *** MikeSmithW3C^ has joined #chromium
[10:37:25] *** podivilov has joined #chromium
[10:39:30] *** MikeSmithW3C has quit IRC
[10:41:04] *** ph0nk has joined #chromium
[10:41:57] *** ROBOd has joined #chromium
[10:44:38] <tyoshino> ok. so it takes some time.
[10:47:31] *** GeekShadow has joined #chromium
[10:48:06] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open ("Linux Tests (dbg)(2)" looks like a similar flake to the last one). Mac official -> see thread on crdev
[10:49:07] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Builder (dbg)" from 52464: mnissler at chromium dot org, satish at chromium dot org)
[10:50:14] *** satish_ has joined #chromium
[10:50:28] <satish_> mnissler: build break?
[10:50:30] <mnissler> ye olde PDB error is back...
[10:51:00] <satish_> is it flakiness which gets resolved in the next build?
[10:51:05] <mnissler> let's wait for the next build, we've had that before.
[10:51:18] <mnissler> yes, let's hope so :)
[10:51:25] <mnissler> seen that a couple of times in the last weeks
[10:51:28] <satish_> apparently the next buidl (mine) also failed
[10:51:46] <satish_> the one after that is yours, going on now.. hopefully will succeed :)
[10:52:07] *** trungl has quit IRC
[10:52:40] *** mylene_ has joined #chromium
[10:54:19] <mnissler> satish_: yours and mine are actually the same, check the build number or click the merge link at the bottom of the page.
[10:54:41] <satish_> oh right, sorry.
[10:54:46] *** General13372 has quit IRC
[10:58:58] *** nkostylev has joined #chromium
[10:59:27] *** ChanServ sets mode: +v nkostylev
[11:00:56] <tyoshino> ok. wait an see.
[11:01:00] <tyoshino> and
[11:01:39] <satish_> tyoshino: perhaps fix the tree status to say we are waiting to cycle green?
[11:01:49] *** General1337 has joined #chromium
[11:04:13] *** jparent has quit IRC
[11:04:13] *** jparent has joined #chromium
[11:04:13] *** jparent has joined #chromium
[11:04:13] *** ChanServ sets mode: +v jparent
[11:05:16] *** MikeSmithW3C^ has quit IRC
[11:06:33] *** jparent_ has joined #chromium
[11:06:33] *** ChanServ sets mode: +v jparent_
[11:07:07] <tyoshino> done
[11:07:22] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open (hopefully cycle green) (Automatic: "compile" on "Chromium Builder (dbg)" from 52464: mnissler at chromium dot org, satish at chromium dot org)
[11:08:12] *** steinex_ has joined #chromium
[11:08:59] <mnissler> tyoshino, satish_ : everything ok, compile succeeded.
[11:09:11] <tyoshino> cool!
[11:09:27] *** jparent has quit IRC
[11:09:27] *** pfeldman has quit IRC
[11:09:27] *** Zaba has quit IRC
[11:09:28] *** cuba33ci has quit IRC
[11:09:29] *** reiko has quit IRC
[11:09:29] *** Inc has quit IRC
[11:09:30] *** joaop5 has quit IRC
[11:09:30] *** brettw has quit IRC
[11:09:30] *** cyberix has quit IRC
[11:09:31] *** ScriptRipper has quit IRC
[11:09:31] *** alyssad has quit IRC
[11:09:31] *** Forsaken|GER has quit IRC
[11:09:32] *** steinex has quit IRC
[11:09:32] *** fahadsadah has quit IRC
[11:09:32] *** sash_ has quit IRC
[11:09:33] *** nirnimesh has quit IRC
[11:09:33] *** Derevko has quit IRC
[11:09:33] *** kcliu has quit IRC
[11:09:33] *** bryeung has quit IRC
[11:09:34] *** star-affinity has quit IRC
[11:09:35] *** shoe has quit IRC
[11:09:35] *** dhollowa has quit IRC
[11:09:35] *** sbyer has quit IRC
[11:09:36] *** joth_ has quit IRC
[11:09:36] *** satorux has quit IRC
[11:09:36] *** oshima has quit IRC
[11:09:36] *** tkent has quit IRC
[11:09:36] *** willchan has quit IRC
[11:09:36] *** zuh has quit IRC
[11:09:37] *** faitz has quit IRC
[11:09:37] *** Malmis has quit IRC
[11:09:38] *** zer0her0 has quit IRC
[11:09:38] *** adiumtestuser has quit IRC
[11:09:39] *** mattijle has quit IRC
[11:09:39] *** maruel has quit IRC
[11:09:39] *** estade has quit IRC
[11:09:39] *** anders has quit IRC
[11:09:39] *** kbr_google has quit IRC
[11:09:40] *** j^ has quit IRC
[11:09:40] *** adlr has quit IRC
[11:09:40] *** garykac has quit IRC
[11:09:40] *** Herald has quit IRC
[11:09:41] *** praseodym has quit IRC
[11:09:41] *** ss has quit IRC
[11:09:41] *** D-V has quit IRC
[11:09:41] *** steinex_ is now known as steinex
[11:09:41] *** dhollowa_ is now known as dhollowa
[11:09:47] <mnissler> tyoshino: any idea on the Linux webkit builders?
[11:10:02] *** cuba33ci has joined #chromium
[11:10:30] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open ("Linux Tests (dbg)(2)" looks like a similar flake to the last one). Mac official -> see thread on crdev
[11:10:42] <mnissler> tyoshino: seems the tests run fine, but update_scripts somehow failed.
[11:11:10] *** bgmerrell has quit IRC
[11:12:43] <nkostylev> tyoshino: Is someone looking into Linux Builder (ChromiumOS) unit_tests?
[11:13:02] <tyoshino> not yet
[11:13:27] *** sash_ has joined #chromium
[11:13:40] <tyoshino> mnissler: looking
[11:14:55] *** nirnimesh has joined #chromium
[11:14:55] *** ChanServ sets mode: +v nirnimesh
[11:14:57] *** alyssad has joined #chromium
[11:14:58] *** ChanServ sets mode: +v alyssad
[11:15:41] *** dhollowa is now known as 77CAAXZMM
[11:15:50] *** reiko has joined #chromium
[11:15:50] *** pfeldman has joined #chromium
[11:15:50] *** star-affinity has joined #chromium
[11:15:50] *** Zaba has joined #chromium
[11:15:50] *** Inc has joined #chromium
[11:15:50] *** shoe has joined #chromium
[11:15:50] *** dhollowa has joined #chromium
[11:15:50] *** j^ has joined #chromium
[11:15:50] *** sbyer has joined #chromium
[11:15:50] *** joth_ has joined #chromium
[11:15:50] *** satorux has joined #chromium
[11:15:50] *** zelazny.freenode.net sets mode: +vvvv dhollowa sbyer joth_ satorux
[11:15:50] *** oshima has joined #chromium
[11:15:50] *** tkent has joined #chromium
[11:15:50] *** joaop5 has joined #chromium
[11:15:50] *** willchan has joined #chromium
[11:15:50] *** zuh has joined #chromium
[11:15:50] *** faitz has joined #chromium
[11:15:50] *** Malmis has joined #chromium
[11:15:51] *** ScriptRipper has joined #chromium
[11:15:51] *** zer0her0 has joined #chromium
[11:15:51] *** Forsaken|GER has joined #chromium
[11:15:51] *** adiumtestuser has joined #chromium
[11:15:51] *** zelazny.freenode.net sets mode: +vvvv oshima tkent willchan adiumtestuser
[11:15:51] *** D-V has joined #chromium
[11:15:51] *** ss has joined #chromium
[11:15:51] *** praseodym has joined #chromium
[11:15:51] *** garykac has joined #chromium
[11:15:51] *** adlr has joined #chromium
[11:15:51] *** Herald has joined #chromium
[11:15:51] *** kbr_google has joined #chromium
[11:15:51] *** anders has joined #chromium
[11:15:51] *** estade has joined #chromium
[11:15:51] *** maruel has joined #chromium
[11:15:51] *** zelazny.freenode.net sets mode: +vvvv garykac kbr_google estade maruel
[11:15:51] *** mattijle has joined #chromium
[11:15:51] *** Derevko has joined #chromium
[11:15:51] *** kcliu has joined #chromium
[11:15:51] *** bryeung has joined #chromium
[11:15:51] *** ChanServ sets mode: +v reiko
[11:16:54] <tyoshino> just retrying
[11:17:30] *** brettw has joined #chromium
[11:17:30] *** ChanServ sets mode: +v brettw
[11:17:48] *** cyberix has joined #chromium
[11:17:49] *** satish_ has quit IRC
[11:17:56] <tyoshino> hmm. failed again
[11:18:38] *** fahadsadah has joined #chromium
[11:20:43] <mnissler> tyoshino: seems the update scripts now fails consistently on linux.
[11:21:16] <tyoshino> oh. so many reds...
[11:22:31] <mnissler> but the builder compiled just fine.
[11:22:52] *** MikeSmithW3C has joined #chromium
[11:25:20] <mnissler> tyoshino: looks like an svn timeout of some sort...
[11:27:51] *** eroman has quit IRC
[11:29:51] <trungl-bot> Tree closed by tyoshino at chromium dot org: Tree is closed (too many linux redness)
[11:30:42] *** roc has joined #chromium
[11:35:34] <mnissler> tyoshino: seems like crostools update is really slow, leading to a timeout.
[11:37:10] <mnissler> tyoshino: I tried accessing the git repo this fetched from and it's really slow...
[11:38:47] *** mylene_ has left #chromium
[11:41:19] <tyoshino> mnissler: thanks for investigation
[11:42:52] *** eseidel has quit IRC
[11:48:55] <tyoshino> mnissler: i tried manually and it seems to be fine now.
[11:49:07] <jochen__> yeah, me too
[11:49:22] <jochen__> i can commit a small change and see whether it turns green
[11:49:42] <yusukes> This failure http://build.chromium.org/buildbot/waterfall/builders/Linux%20Builder%20(ChromiumOS)/builds/9086/steps/unit_tests/logs/GetTextForIndicatorTest seems to be caused by my change.
[11:49:45] <yusukes> will submit a fix asap..
[11:50:23] <mnissler> tyoshino, jochen__ : still not exactly fast, but much better for me as well.
[11:51:05] <tyoshino> jochen_: i'
[11:51:14] <tyoshino> ve committed small change
[11:51:19] <tyoshino> to trigger
[11:51:46] <jochen__> me too
[11:51:55] <jochen__> t'will work probably anyway
[11:51:56] <mnissler> tyoshino: do you want to wait for a cycle to see if everything is back to normal?
[11:52:37] <jochen__> we could wait for yusukes to land his compile fix
[11:53:01] <tyoshino> mnissler: let me wait and see some bots. if fine, i'll open the tree
[11:53:31] <tyoshino> jochen_: that's also good
[11:56:33] <yusukes> hmm...  tyoshino: can you clobber the Linux-ChromeOS bot? i think that should fix the redness of the bot.
[11:56:47] <tyoshino> yusukes: ok
[11:59:35] *** ager has joined #chromium
[12:00:46] *** hbono has quit IRC
[12:01:12] *** chrisccoulson has quit IRC
[12:07:45] *** abarth is now known as abarth|zZz
[12:08:28] *** chrisccoulson has joined #chromium
[12:10:29] *** General1337 has quit IRC
[12:10:35] *** General13372 has joined #chromium
[12:11:09] *** shoe has quit IRC
[12:12:36] *** leeight has joined #chromium
[12:12:42] *** leeight has left #chromium
[12:13:50] *** shoe has joined #chromium
[12:17:15] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open ("Linux Tests (dbg)(2)" looks like a similar flake to the last one). Mac official -> see thread on crdev, linux cros -> yusukes cycling
[12:19:17] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "svnkill" on "XP Tests (dbg)(2)")
[12:19:36] <yusukes> :(
[12:22:52] *** caac has joined #chromium
[12:23:29] <caac> hi, how come desktop apps give 2 taskbar entries when opening one? after a while the 2nd entry disappears though
[12:23:54] <jochen__> flaky
[12:23:57] <mnissler> tyoshino: flakiness?
[12:24:08] <jochen__> and the crostools git repo starts to timeout again :(
[12:27:45] <tyoshino> well...
[12:27:58] -tyoshino- ERROR: The process "svn.exe" not found.
[12:30:56] <tyoshino> ok. it's fine now -> XP Tests (dbg)(2)
[12:31:23] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open ("Linux Tests (dbg)(2)" looks like a similar flake to the last one). Mac official -> see thread on crdev, linux cros -> yusukes cycling
[12:32:14] <caac> how come desktop apps give 2 taskbar entries when opening one? after a while the 2nd entry disappears though
[12:32:36] *** satish_ has joined #chromium
[12:33:11] *** leeight has joined #chromium
[12:33:13] *** leeight has left #chromium
[12:33:26] <satish_> Some of my try-server runs are failing at 'update scripts' stage, with a message '60 seconds without output'.. any suggestions? http://build.chromium.org/buildbot/try-server/builders/linux/builds/38283/steps/shell/logs/stdio
[12:35:21] <tyoshino> satish_: we saw the same problem on main build bots
[12:35:35] <satish_> ah.
[12:35:36] <tyoshino> fetching crostools is too slowwww
[12:36:22] <satish_> is crosstools a google code project? i remember a mail thread about google code hosting being slow at times
[12:38:04] *** MikeSmithW3C has quit IRC
[12:45:52] *** mazda1 has quit IRC
[12:54:54] *** peper has quit IRC
[12:56:06] *** General1337 has joined #chromium
[12:58:04] *** General13372 has quit IRC
[13:01:44] *** caac has left #chromium
[13:05:46] *** satish_ has quit IRC
[13:10:48] <tyoshino> it's internal stuff
[13:11:04] *** barcon332 has joined #chromium
[13:12:28] *** awolfson has joined #chromium
[13:12:48] <trungl-bot> Tree opened by tyoshino at chromium dot org: Tree is open. be careful. linux bots are flaky due to crostools check out slowness ("Linux Tests (dbg)(2)" looks like a similar flake to the last one). Mac official -> see thread on crdev, linux cros -> yusukes cycling
[13:16:52] *** MikeSmithW3C has joined #chromium
[13:21:51] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "test_shell_tests" on "Webkit" from 52476: ager at chromium dot org, joth at chromium dot org)
[13:22:05] <jorlow> bulach's pig is faster than trungl-bot
[13:22:51] <hwennborg> we should build a pig-bot
[13:23:12] <jorlow> yes...so pig-bot and trungl-bot can race
[13:23:18] <jorlow> and pig-bot can win
[13:23:40] <jorlow> pig bot needs to periodically complain about marcus' habit of eating pork tho
[13:24:28] *** peper has joined #chromium
[13:26:41] <mnissler> any chance we can get a pig for the MUC office too?
[13:27:02] <mnissler> ;)
[13:27:19] <bulach> mnissler: if you have one of those LEDs, I can send you a pig.. ;)
[13:28:03] <mnissler> bulach: I don't think we have one.. :(
[13:28:24] <jorlow> awwe...well work on getting test certified
[13:28:28] <jorlow> so u can get one
[13:28:34] <jorlow> but yeah...we can give you LOTS of spare pigs
[13:30:01] <jochen__> bulach: where can we buy one?
[13:30:08] <bulach> mnissler, jochen: I'll check with the test people here if they have any spare..
[13:30:16] <mnissler> so, the failing WebKit test looks like flakiness, no?
[13:30:29] <jochen__> having the brand/model would be enough, i can then just order one
[13:30:38] <jorlow> it's made by a googler
[13:30:44] <jorlow> not sure if you can buy them
[13:30:44] <jochen__> oh
[13:31:18] <jochen__> with the state of the tree and the trybots, we'll run into trouble with the feature freeze
[13:31:41] *** tonikitoo has quit IRC
[13:32:00] <jorlow> we need to shard some
[13:32:04] <jorlow> would be easy with git  :-)
[13:32:48] <jochen__> the reason why all linux bots are red is git...
[13:34:18] *** BryanWB has joined #chromium
[13:34:48] <mnissler> it's probably not git, but an overloaded machine.
[13:35:02] *** andrix has joined #chromium
[13:35:02] <mnissler> don't try this with svn :)
[13:35:18] <jochen__> the test_shell_test looks flaky to me
[13:35:31] <ager> Yes, can't repro test_shell_tests either.
[13:36:18] <jochen__> looks like it just couldn't start up the external server
[13:36:32] <jochen__> but, there's more redness bubbling up from below
[13:38:00] <trungl-bot> Tree closed by jochen at chromium dot org (:jochen__): Tree is closed (too much rendess bubbling up)
[13:38:24] <jochen__> i'd like to have all bots finished before we reopen again
[13:38:33] <jochen__> everything else would be suicide in this state
[13:40:19] <tyoshino> makes sense.
[13:42:57] *** apavlov has joined #chromium
[13:47:08] <jochen__> bulach: so the pig is only glowing red or not
[13:47:10] <apavlov> hey folks. I send a CL to try bots and all of them fail on the "update scripts" step timing out on the line "_____ scripts/crostools at refs/heads/master" (i.e. its update never completes: "command timed out: 60 seconds without output, killing pid ###"). Any clues?
[13:47:18] <jochen__> the pigs we have are made from red plastic..
[13:47:26] <jochen__> apavlov: ignore it
[13:47:36] <apavlov> jochen__: thanks! :)
[13:50:21] *** hamaji has quit IRC
[13:52:05] *** JakeSays has quit IRC
[13:53:18] *** JakeSays has joined #chromium
[13:59:06] *** zub has joined #chromium
[13:59:59] <zub> hi
[14:00:56] <tyoshino> Builder WebKit test_shell_tests is fine now
[14:01:00] <tyoshino> hi
[14:01:21] <zub> Does WebGL in chromium work on GL ES?
[14:02:04] <jorlow> WebGL is only GL ES
[14:02:33] <zub> but when using an ARM build, it wants OpenGL (GLX)
[14:02:43] <jorlow> odd
[14:02:50] <jorlow> i don't know anything about it, unfrotunately
[14:02:56] <zub> but I don't know my way around the sources
[14:03:05] <zub> thanks anyway, I'm investigating so far
[14:03:17] <zub> maybe I just need to tell it to use EGL
[14:04:55] *** peavey has quit IRC
[14:06:50] *** peavey has joined #chromium
[14:07:24] *** Omen_20 has quit IRC
[14:11:02] *** janm has joined #chromium
[14:12:58] *** BryanWB has quit IRC
[14:18:14] *** awolfson has quit IRC
[14:19:21] *** coyo has quit IRC
[14:21:42] *** jeremymos has quit IRC
[14:23:01] *** peavey has quit IRC
[14:24:26] *** peavey has joined #chromium
[14:34:13] *** tonikitoo has joined #chromium
[14:37:05] *** paloris_J has joined #chromium
[14:37:57] <paloris_J> I can not delet my history.  Could some one help
[14:38:44] <paloris_J> Any thoughts
[14:40:34] *** ebraminio has joined #chromium
[14:41:27] *** brettw has quit IRC
[14:41:40] *** peper has quit IRC
[14:41:41] *** brettw has joined #chromium
[14:41:42] *** ChanServ sets mode: +v brettw
[14:43:46] <pfeldman> mnissler: ping
[14:44:31] <mnissler> pfeldman: hi, what's up?
[14:44:47] <pfeldman> mnissler: do you think your stuff should be rolled out? too much redness started with you
[14:46:01] <mnissler> pfeldman: I think it's just flakiness. My understanding of the status is to let the tree settle and see whether there's any real breakage.
[14:46:19] <jochen__> pfeldman: which revision do you refer to?
[14:46:41] <pfeldman> jochen__: r52466
[14:46:55] <pfeldman> and r52463
[14:47:11] <jochen__> that's the crostools git server being slow
[14:47:11] <mnissler> pfeldman: that's a not very responsive git server
[14:47:41] <jochen__> paloris_J: see topic
[14:47:46] *** mnaganov has joined #chromium
[14:47:51] <mnissler> pfeldman: IMHO it's save to reopen. We need to watch linux closely though in order to prevent any real breakage from slipping through
[14:47:52] <pfeldman> hm
[14:48:16] <pfeldman> mnissler: open? this is nuts. i'd jsut force failed builds on waterfall
[14:48:18] *** AaronMT has joined #chromium
[14:49:23] <jochen__> pfeldman: the only real failures are the morejs perf test and the mac bot not being able to link
[14:49:55] <tyoshino> and, mac perf
[14:50:08] <pfeldman> jochen__: this is more than enough to keep the tree closed/
[14:50:11] <jochen__> i guess we should rebuild the webkit and mac10.6 test
[14:50:29] <pfeldman> jochen__: will you kick them?
[14:50:38] <mnissler> pfeldman: I'm not aware of anything on the main waterfall that looks like a real problem, but I agree it looks very red.
[14:50:59] <jochen__> pfeldman: both of them are failing since days
[14:51:11] <jochen__> pfeldman: will do
[14:51:38] <pfeldman> mnissler, jochen__: what about XP Perf? this does look like r52467
[14:52:07] <paloris_J> jochen__:  I know how to do it. It is not working.   I got this running it through a terminal [9851:9872:17167914621:ERROR:chrome/browser/history/history_backend.cc(2024)] Main history could not be cleared
[14:52:09] <pfeldman> oh
[14:52:10] <pfeldman> ok
[14:52:13] <pfeldman> it has been flaky
[14:52:44] <pfeldman> but r52467 made it failing. i'd revert :)
[14:53:13] <jochen__> paloris_J: file a bug then please
[14:55:03] <paloris_J> I have never filed a bug report befor. Wouldnt have a clue how to do that.  And besids I am not even sure weather it is a bug or some thing wrong with my system
[14:55:04] <mnissler> 52467 is only concerned with the options dialog
[14:55:50] <pfeldman> mnissler: someone explain why it triggered perf bot then...
[14:57:26] <mnissler> pfeldman: I agree it looks like the culprit and I'm happy to revert, but I don't think it'll fix our problems.
[14:57:40] *** apavlov has quit IRC
[14:57:53] <pfeldman> mnissler: it probably crossed the critical mass of the code
[14:58:26] *** paloris_J has quit IRC
[14:58:30] <mnissler> pfeldman: I don't understand?
[14:59:00] <pfeldman> mnissler: well, more code, larger footprint, slower browser, failing perf
[15:00:15] *** apavlov has joined #chromium
[15:00:38] <mnissler> pfeldman: fair enough, so you think reverting will help?
[15:01:41] <pfeldman> mnissler: i am not sure. given that the tree is closed, reverting would be a nice experiment (also will kick bots)
[15:02:41] <mnissler> pfeldman: but if you're theory is correct, shouldn't the log show at least some output instead of nothing?
[15:03:06] <jochen__> pfeldman: i'd like to reboot the xp perf (dbg) bot
[15:03:22] <jochen__> so let's wait until the current run is finished
[15:05:01] <tyoshino> ~20min eta
[15:05:36] *** apavlov has quit IRC
[15:05:53] *** apavlov has joined #chromium
[15:06:19] <jochen__> xp perf is done
[15:06:29] <jochen__> i reboot it now
[15:07:30] <jochen__> once it's back up, i'll force a build
[15:07:36] *** monreal has joined #chromium
[15:07:47] <mnissler> and if that doesn't fix things, I'll revert.
[15:08:07] <tyoshino> good.
[15:12:40] *** taf2_ has joined #chromium
[15:12:43] *** satish_ has joined #chromium
[15:13:11] <satish_> tyoshino: Looks like builds are still failing at the same update-scripts phase?
[15:14:06] <jochen__> yes, that's due to a slow git server
[15:14:25] <tyoshino> yes. try bots are failing due to that.
[15:15:06] <satish_> any workaround? or should we just wait for a few more hours hoping the git server catches up?
[15:15:24] <jochen__> wait it is :(
[15:15:32] *** taf2 has quit IRC
[15:15:45] <jochen__> basically, we need to wait until somebody with the right credentials wakes up and fixes the server
[15:15:48] <satish_> i see at least the linux try bot is consistently failing in there where as mac and windows seem ok
[15:15:53] <satish_> which is interesting.
[15:17:32] <tyoshino> crostools dep is written only for linux platform
[15:17:33] <jochen__> crostools is chromeos
[15:18:26] *** tonikitoo has quit IRC
[15:20:07] <satish_> ok.
[15:20:18] <jochen__> pfeldman: the page cycler is green
[15:20:24] <satish_> good to know, i'll try the builds on mac and win for now.
[15:20:53] <jochen__> one of the mac bots is green
[15:21:01] <jochen__> the mac perf bot and the webkit bot are still running
[15:21:11] <tyoshino> thanks,
[15:21:28] <pfeldman> jochen__: great, thanks!
[15:24:57] *** Viclame has joined #chromium
[15:27:14] *** satish_ has quit IRC
[15:28:02] *** davirtavares has joined #chromium
[15:29:04] <jochen__> no idea
[15:29:19] <jochen__> i'd say it's as green as it gets without help from some cros folks
[15:29:39] <davirtavares> how much effort is needed to make Chromium run headless (without X Server, I mean)?
[15:30:55] <mnissler> mac 10.6 perf(2) is still red, but my change only touched windows code and it's failed lately at 52474 and 52475
[15:31:55] *** peper has joined #chromium
[15:32:32] <jochen__> one of those is a v8 roll... no idea
[15:33:02] <tyoshino> seems it started to failing around build #5162
[15:33:43] *** tonikitoo has joined #chromium
[15:34:02] <ager> The V8 roll contains only 4 V8 revisions. Nothing that should affect perf bots at all.
[15:34:10] *** awolfson has joined #chromium
[15:40:41] *** dave_levin has joined #chromium
[15:40:41] *** yutak_home has joined #chromium
[15:40:42] *** ChanServ sets mode: +v yutak_home
[15:40:53] <dave_levin> trungl-bot: lkgr
[15:40:54] <trungl-bot> dave_levin: 52458
[15:42:54] <jochen__> nicolas woke up
[15:47:24] *** dave_levin has quit IRC
[15:50:26] *** rohitrao has joined #chromium
[15:50:26] *** ChanServ sets mode: +v rohitrao
[15:50:34] *** dob1 has joined #chromium
[15:52:01] *** hebz0rl has joined #chromium
[15:53:26] <trungl-bot> Tree closed by mnissler at chromium dot org: Tree is closed (too much rendess bubbling up, desperately need someone with privileges to fix crostools repository)
[15:54:41] *** rsesek has joined #chromium
[15:54:41] *** ChanServ sets mode: +v rsesek
[15:57:45] *** shreyas_ has joined #chromium
[16:04:27] <thakis_> tears
[16:04:35] *** bgmerrell has joined #chromium
[16:05:11] <rsesek> we should cut down this tree and plant a new one. maybe an oak or a pine
[16:07:33] *** agl has joined #chromium
[16:07:33] *** ChanServ sets mode: +v agl
[16:09:05] <jorlow> well, we should plant more than one
[16:09:18] *** trungl has joined #chromium
[16:09:18] *** ChanServ sets mode: +v trungl
[16:09:20] *** thakis_ has quit IRC
[16:09:27] <rsesek> you're right. this is a scaling issue; we need a forest
[16:10:01] *** cbentzel has joined #chromium
[16:11:01] <trungl> crap
[16:11:12] <jochen__> it's not as bad as it looks like
[16:11:22] * trungl should have committed late last night when he had the chance.
[16:11:25] <jochen__> we suspect that all failures are related to crostools
[16:11:42] <agl> Where's crostools referenced from? (git grep crostools shows nothing)
[16:11:43] <jochen__> i expect things to become real bad once we open the tree again
[16:11:56] <jochen__> in the slave DEPS file
[16:12:14] * trungl weeps.
[16:12:17] <jochen__> when we open the tree again, everybody will commit a bunch of CLs instantly
[16:12:25] <agl> jochen__: we'll have to rate limit it manually
[16:12:31] <agl> (once I've checked in, obviously)
[16:12:36] <jochen__> yes
[16:12:53] <jochen__> not that it was thursday evening here, which makes tomorrow the feature freeze
[16:12:59] <jochen__> if you don't intend to work all weekend
[16:14:09] <trungl> is anyone (with privileges) looking at crostools yet?
[16:14:15] <jochen__> no
[16:14:24] <jochen__> i can't log in with chrome-bot credentials
[16:14:31] <jochen__> on #chromium-os nobody knew what to do
[16:14:38] *** wr| has quit IRC
[16:14:48] <trungl> is maruel around?
[16:14:53] <jochen__> i pinged him
[16:14:58] <jochen__> i guess he's showering or smth
[16:15:16] <agl> how about we delete crostools and let the ChromeOS builders fall?
[16:15:26] <agl> I mean, shouldn't it already be checked out on them anyway?
[16:15:29] <rsesek> it will try to check them out again
[16:15:34] <rsesek> we'd need to remove it from the gclient spec
[16:15:35] <agl> they might just keep working
[16:15:41] <rsesek> but that could work
[16:15:53] <agl> I can remove it from the slave DEPS file and see what happens
[16:16:04] <agl> are any sherriffs awake?
[16:16:24] <jochen__> one was an hour ago
[16:16:43] <jochen__> maybe he still is
[16:16:50] <agl> I'll take that as a no then and give this a shot.
[16:17:15] <jochen__> have fun
[16:17:23] *** tonikitoo has quit IRC
[16:17:28] <jochen__> i just need to hit enter to land my cl if you need something to trigger a build ;)
[16:17:44] <agl> jochen__: just doing whitespace changes for now ;)
[16:17:53] *** tonikitoo has joined #chromium
[16:18:01] <agl> I've tweaked Linux x86. We'll see if it does any good.
[16:20:16] <agl> nope
[16:20:42] <agl> although we get a different error message
[16:21:54] *** motownavi_ has joined #chromium
[16:21:54] *** ChanServ sets mode: +v motownavi_
[16:22:02] *** motownavi_ has left #chromium
[16:22:27] <trungl> mmoss: ping
[16:25:06] <agl> trungl: it's only 7:30am mmoss time
[16:25:13] <trungl> I know
[16:25:21] * trungl time = mmoss time
[16:25:35] <trungl> he's showing up as logged in on gchat
[16:25:45] <trungl> one can hope
[16:26:06] <agl> ok, Linux x64 is going now.
[16:26:08] *** pinkerton has joined #chromium
[16:26:08] *** ChanServ sets mode: +v pinkerton
[16:26:08] <agl> I can do the rest
[16:26:27] *** tedoc2000 has joined #chromium
[16:26:33] <trungl> pinkerton!
[16:27:21] <pinkerton> trungl!
[16:27:22] *** wr| has joined #chromium
[16:27:39] <trungl> pinkerton: we made the tree nice and pink for your return
[16:28:37] <pinkerton> yay?
[16:28:57] <agl> I'm working on the Linux builders
[16:29:07] <pinkerton> i think i'm sheriff on monday
[16:29:45] <trungl> maybe the craziness will be over by then
[16:29:49] *** ebraminio has quit IRC
[16:30:04] *** bulach has quit IRC
[16:30:04] <trungl> though somehow that seems unlikely
[16:30:30] <pinkerton> maybe if i press the tree in the lower left corner, the red will go away
[16:30:47] <trungl-bot> Tree closed by agl at chromium dot org: Tree is closed (too much rendess bubbling up, desperately need someone with privileges to fix crostools repository. agl working on getting Linux builders going again.)
[16:31:04] *** jorlow has quit IRC
[16:31:05] *** hwennborg has quit IRC
[16:32:34] <agl> How does "Linux Perf (1)" have a hostname of chrome-linux-perf? Anyone know where it really is?
[16:32:40] *** hwennborg has joined #chromium
[16:33:41] <agl> ah, found a Wiki page listing it's IP address
[16:33:47] *** joth_ has quit IRC
[16:34:08] *** andreip__ has quit IRC
[16:34:48] *** jorlow has joined #chromium
[16:37:06] *** Gina has joined #chromium
[16:38:18] <agl> 4 builders done
[16:38:55] *** erikkay has joined #chromium
[16:38:55] *** ChanServ sets mode: +v erikkay
[16:39:03] *** Buglouse has joined #chromium
[16:40:53] <jochen__> who's taking bets on whether the git server is first up again, or agl fixes all builders
[16:42:12] *** Ghent has joined #chromium
[16:42:28] <agl> jochen__: if your patch is pretty safe, you can land it now
[16:42:41] <agl> jochen__: (I want something to check the builders that I think should be working now)
[16:43:17] <Ghent> I'm not having much luck finding an answer on google, I just downloaded the latest osx build of chromium and now when I try to see a youtube video it says "missing plugin" where the video should be.  Refreshing doesn't seem to help.  Is there some sort of fix?
[16:43:46] <jochen__> it includes a grd change...
[16:43:56] <agl> jochen__: maybe not then
[16:44:30] * trungl has a fairly safe (linux-only) change. At least he hopes it's fairly safe (famous last words).
[16:44:56] <agl> trungl: go for it
[16:45:23] <trungl> agl: ok, thanks
[16:45:23] *** ager has quit IRC
[16:45:40] <mmoss> trungl: hi
[16:45:50] <pinkerton> Ghent: see topic (#chromium-support) but i can't say i've seen that with nightly builds
[16:45:56] <rsesek> pinkerton!
[16:46:02] <pinkerton> heh
[16:46:05] <pinkerton> rsesek!
[16:46:08] <rsesek> hi :)
[16:46:12] <Ghent> pinkerton: my mistake!  thank you for the heads up :)
[16:46:19] *** Ghent has left #chromium
[16:46:41] <trungl> mmoss: I was wondering if you were awake enough to help out with the bots (agl has more details and is doing something that should get them running again -- mostly?).
[16:47:12] <mmoss> trungl: yeah, looking
[16:47:19] <trungl> mmoss: thanks
[16:48:32] <rsesek> agl: fyi for whitespace we have src/build/whitespace_file.txt
[16:48:40] <agl> mmoss: I'm removing the crostools from the DEPS for the moment and running gclient sync twice, manually
[16:48:47] <agl> rsesek: oh, thanks.
[16:49:49] *** yutak_home has quit IRC
[16:50:50] *** thakis_ has joined #chromium
[16:50:50] *** ChanServ sets mode: +v thakis_
[16:51:36] <mmoss> agl: everywhere?
[16:51:48] <agl> mmoss: on all the Linux builders that I can
[16:52:30] <agl> trungl: landing?
[16:52:51] <mmoss> agl: would it be easier/faster to just send a couple noop changelists to kick gclient? or are they hosed so bad it needs to be manual?
[16:52:52] <trungl> agl: unexpectedly having to rebase, and the merge failed :(
[16:53:23] <trungl> agl: so go for a whitespace change if you need
[16:53:36] <agl> trungl: ok
[16:53:51] <agl> mmoss: gclient seems to wedge, trying to print a warning to a non-terminal
[16:54:11] <thakis_> agl: looking for a cl to land?
[16:54:18] <thakis_> i have http://codereview.chromium.org/2963009
[16:54:33] <agl> thakis_: if you can do it in 10 seconds, go for it
[16:54:50] <agl> thakis_: otherwise I just landed a whitespace change to (hopefully) green the builders
[16:54:58] <thakis_> doing?
[16:55:34] <thakis_> agl: gah, merge conflicts
[16:55:37] <thakis_> nvm then
[16:56:29] <agl> thakis_: thanks anyway
[16:56:36] <thakis_> was an easy merge
[16:56:39] <thakis_> i'd be ready now
[16:56:40] <trungl> me too
[16:56:41] <thakis_> if it stil lhelps
[16:57:01] *** MichealH has joined #chromium
[16:57:04] <trungl> no, me!
[16:57:06] *** rdsmith has joined #chromium
[16:57:10] <trungl> me! me!
[16:57:10] *** star-affinity has quit IRC
[16:58:06] <agl> trungl: thakis_: let's see if anything goes more green for a minute
[16:58:22] <thakis_> me! me!
[16:58:25] <agl> then we need to start landing real changes otherwise we'll have a pile on
[16:58:34] <rsesek> I also have a CL waiting to land
[16:58:48] <mnissler> guys, I guess we all have CLs ready to be landed :)
[16:58:51] <trungl> rsesek: get back in line, young man
[16:58:57] <thakis_> i won't lgtm any more CLs for this week
[16:59:01] <thakis_> to keep the tree green for me
[16:59:02] <rsesek> :(
[16:59:07] <trungl> rsesek: (and get off my lawn too)
[16:59:18] <rsesek> :p
[16:59:23] *** tonyg-cr has joined #chromium
[16:59:23] *** ChanServ sets mode: +v tonyg-cr
[16:59:59] *** malavv has joined #chromium
[17:00:05] <trungl> we should run a manual commit queue
[17:00:07] <agl> the shell stages are looking much better
[17:00:08] *** ebraminio has joined #chromium
[17:00:11] <trungl> "just take a number"
[17:00:13] <rsesek> trungl: take a number. now serving _
[17:00:14] <rsesek> ha
[17:00:14] <rsesek> yes
[17:00:19] <thakis_> agl: update scripts failed again
[17:00:28] <thakis_> tests x64
[17:00:30] <thakis_> linux
[17:01:03] <trungl-bot> Tree closed by mnissler at chromium dot org: Tree is closed (too much rendess bubbling up, desperately need someone with privileges to fix crostools repository. agl working on getting Linux builders going again.)
[17:01:20] <agl> thakis_: thanks, I missed that one.
[17:01:25] <agl> (and all the release builders I note)
[17:01:29] *** satish_ has joined #chromium
[17:01:29] *** fusion44 has joined #chromium
[17:02:15] <thakis_> agl: and once you're waiting for stuff to cycle and twiddling your thumbs, maybe you can take a look at https://bugs.webkit.org/show_bug.cgi?id=42214 ? :-)
[17:02:30] <thakis_> (it's a patch, not a bug)
[17:06:14] <satish_> I'm looking for a way in chromium (browser process code) to connect to a url and stream out a large chunk of binary data (something like recorded audio). Anyone familiar enough with the net modules to suggest where I can start looking? Something like URLFetcher but which allows me to incrementally keep sending data as the http request's message body (a POST message) would be ideal.
[17:06:28] *** shreyas_ has quit IRC
[17:06:29] *** wr| has quit IRC
[17:08:47] *** bent-mozilla has joined #chromium
[17:09:47] *** maikmerten has quit IRC
[17:10:19] <agl> trungl: land
[17:10:40] <agl> trungl: (this offer expires in 60 seconds ;)
[17:10:51] *** awidegreen has joined #chromium
[17:10:55] *** Venom_X has joined #chromium
[17:11:01] *** m0 has quit IRC
[17:11:01] *** cdn has quit IRC
[17:11:02] *** gkelly has quit IRC
[17:11:02] *** trungl-bot has quit IRC
[17:11:04] *** cdn has joined #chromium
[17:11:04] *** gkelly has joined #chromium
[17:11:04] *** m0 has joined #chromium
[17:11:05] *** jibot has joined #chromium
[17:11:06] <agl> thakis_: this is Dean's bug?
[17:11:07] <trungl> agl: landing
[17:11:09] *** ChanServ sets mode: +v cdn
[17:11:44] <thakis_> agl: it's a bug where dean pointed out how to fix it, if that was the question
[17:12:07] *** dr_win has joined #chromium
[17:12:26] <thakis_> if trungl doesn't want to land, i can?
[17:12:40] *** MikeSmithW3C^ has joined #chromium
[17:12:44] * trungl landed.
[17:13:09] <trungl> thakis_: get back in line! set a good example for the young 'uns.
[17:13:29] * thakis_ helps an old lady across the street
[17:13:32] <mnissler> please collect trungl's baggage on belt 3
[17:13:47] *** Omen_20 has joined #chromium
[17:14:04] *** stuartmorgan has joined #chromium
[17:14:04] *** ChanServ sets mode: +v stuartmorgan
[17:14:04] * trungl hopes thakis_ waited for the walk signal.
[17:14:42] *** [Nef] has quit IRC
[17:14:49] *** trungl-bot has joined #chromium
[17:14:49] *** ChanServ sets mode: +v trungl-bot
[17:15:03] * bauerb was grabbed by and old lady wanting to cross the street once.
[17:15:05] <trungl> 'morning, trungl-bot
[17:15:05] <trungl-bot> trungl: Good morning!
[17:15:15] <bauerb> an old lady*
[17:15:26] <trungl> trungl-bot: where have you been?
[17:15:30] *** phajdan-jr has joined #chromium
[17:15:30] *** ChanServ sets mode: +v phajdan-jr
[17:15:44] *** MikeSmithW3C has quit IRC
[17:16:10] <trungl> trungl-bot seems to have gone missing again
[17:16:28] *** seventh has quit IRC
[17:16:52] <mnissler> I guess trungl-bot is sitting on the ethernet cable between codf17 and the rest of the world...
[17:17:06] <trungl-bot> trungl: Error: "where" is not a valid command.
[17:17:06] *** trungl-bot has quit IRC
[17:17:16] *** trungl-bot has joined #chromium
[17:17:16] *** ChanServ sets mode: +v trungl-bot
[17:17:21] * trungl unceremoniously restarts trungl-bot.
[17:17:28] *** duffydack has joined #chromium
[17:17:37] <trungl> maybe it'll get a different irc node or something
[17:17:45] <thakis_> trungl-bot got inspired by the tree
[17:17:45] <trungl-bot> thakis_: Error: "got" is not a valid command.
[17:18:01] <thakis_> trungl-bot also needs to relax more
[17:18:01] <trungl-bot> thakis_: Error: "also" is not a valid command.
[17:18:20] <trungl> it had been a very long time since trungl-bot had taken a vacation
[17:18:44] *** Erkan_Yilmaz has joined #chromium
[17:19:18] <danno_> speaking of a well earned vacation, can I land a .grd change?
[17:19:23] * danno_ ducks for cover
[17:19:49] <trungl> danno_: take a number
[17:19:55] <bauerb> danno_: by well-earned vacation, you mean a forced one if you land it?
[17:19:59] <agl> danno_: probably not for a few hours at least
[17:22:52] <danno_> agl: after I stopped laughing, I realized you're serious.
[17:23:00] *** bweinstein has joined #chromium
[17:24:16] <agl> The Mac10.5 Perf(2) regression window contains four patches, all of white appear innocent
[17:24:41] <agl> It's only a slight slowdown as well, and across multiple actions.
[17:24:46] <agl> god knows
[17:25:01] *** wr| has joined #chromium
[17:25:11] * mnissler cheers for agl: green linux bots!
[17:25:18] <agl> Everything has been getting slower over time: http://build.chromium.org/buildbot/perf/mac-release-10.5/startup/report.html?history=150&rev=-1
[17:25:43] <mnissler> agl: reboot the bot?
[17:26:21] *** ph0nk has quit IRC
[17:26:30] <agl> mnissler: seems like a reasonable idea if I knew how
[17:26:33] <nsylvain> hey. i'm here now. Did anyone fix the git problem?
[17:26:50] <agl> nsylvain: I bodged it for now, but the git repo is still down
[17:26:59] <nsylvain> ok
[17:27:05] <agl> nsylvain: the machine doesn't reply to ssh, I think mmoss was going to try to reboot it
[17:27:24] <agl> thakis_: land
[17:27:28] <nsylvain> Ok
[17:27:32] <mnissler> agl: don't know either
[17:29:05] *** til_ has joined #chromium
[17:29:05] *** til has quit IRC
[17:29:05] *** til_ is now known as til
[17:29:23] <thakis_> agl: done
[17:29:27] *** tonyg-cr has quit IRC
[17:30:33] <agl> nsylvain: I think we would also like to reboot Mac10.5 Perf (2)
[17:32:40] <agl> who else is waiting to land?
[17:32:43] <rsesek> me
[17:32:55] *** phajdan-jr is now known as phajdan-jr|afk
[17:33:18] <agl> rsesek: ok, you're next up then but hold off for a moment.
[17:33:22] <rsesek> sure
[17:33:23] <rsesek> thanks
[17:34:55] <rsesek> thakis: your CL broke win?
[17:35:07] <thakis_> Error executing rc.exe (tool returned code: 1)
[17:35:29] <thakis_> that's probably not me
[17:35:29] *** duff has joined #chromium
[17:35:59] <thakis_> so it's too soon to know if i broke win
[17:36:16] <agl> thakis_: looks like a builder hiccup, no worries.
[17:36:20] *** tfarina has joined #chromium
[17:36:36] <tfarina> hey what is going on with the tree these days??
[17:37:06] * trungl will be back. If something breaks, it's not his fault.
[17:37:18] <agl> tfarina: we're manually rate limiting checkins
[17:37:26] * trungl is going to a more work-ish location.
[17:37:57] *** trungl has quit IRC
[17:38:03] *** tonyg-cr has joined #chromium
[17:38:03] *** ChanServ sets mode: +v tonyg-cr
[17:38:11] *** duffydack has quit IRC
[17:38:25] <trungl-bot> Tree closed by agl at chromium dot org: Tree is closed - rate limiting. Ask on IRC to land (desperately need someone with privileges to fix crostools repository. agl working on getting Linux builders going again.)
[17:38:35] *** ukai has quit IRC
[17:38:45] <jochen__> agl: can i get a ticket for a checkin please?
[17:38:53] <agl> rsesek: land
[17:38:53] <tfarina> agl: because of the redness of for someone other reason?
[17:39:01] <agl> jochen__: is it a GRD change?
[17:39:02] <tfarina> or*
[17:39:10] <rsesek> agl: k
[17:39:14] <jochen__> agl: i'll clobber the builders to make sure there's not much redness
[17:39:16] <jochen__> agl: yes
[17:39:27] <agl> tfarina: many builders were down and we don't want a crash landing while we fix them up
[17:40:13] <agl> Anyone else have GRD changes?
[17:40:40] <rsesek> landed
[17:40:45] <agl> Anyone else waiting to land?
[17:41:32] <pfeldman> podivilov: might want to do WebKit roll
[17:41:33] <tfarina> agl: I hope in the next days the situation will  be better and under control.
[17:41:35] <pfeldman> ?
[17:41:48] <pfeldman> podivilov: if yes, shout to agl
[17:42:03] <agl> jochen__: ok, you're next up then
[17:42:09] <jochen__> yay
[17:42:14] <jochen__> will you give me an explicit go?
[17:42:46] <agl> jochen__: yea
[17:42:55] *** mrossetti has joined #chromium
[17:43:07] <agl> jochen__: I forget. We have to wait for all the windows builders to start building a GRD change, then submit a whitespace change, right?
[17:43:11] *** johnny_g has joined #chromium
[17:43:23] <rsesek> correct
[17:43:25] *** zaspire has joined #chromium
[17:43:50] <pcgod> dev.chromium.org/developers/tree-sheriffs/sheriff-details-chromium#TOC-Landing-a-GRD-Resources-Change  (wait, why is there no http?)
[17:43:53] <jochen__> i just wait for builders to start, then commit and instant clobber
[17:44:37] *** ebraminio has quit IRC
[17:44:47] <agl> jochen__: ok, a few builders have picked up rsesek's change so land at will.
[17:45:58] *** Aria has joined #chromium
[17:47:05] *** motownavi_ has joined #chromium
[17:47:05] *** ChanServ sets mode: +v motownavi_
[17:47:32] *** jrmuizel has joined #chromium
[17:48:07] *** pfeldman has quit IRC
[17:48:18] *** jwalden has joined #chromium
[17:49:52] *** thakis_ has quit IRC
[17:50:34] <jochen__> git still didn't pick up the latest revisions...
[17:50:40] <rsesek> jochen__: git svn fetch
[17:50:45] <rsesek> the git repo is on a delay
[17:51:07] <danno_> agl: I also have a .grd change, is it too late to get on the train?
[17:51:10] <jochen__> now it does
[17:52:06] <nsylvain> do we even need the git repo stuff in the DEPS file?
[17:52:10] *** duff has left #chromium
[17:52:16] <nsylvain> we can just delete it until we fix it
[17:52:23] <agl> nsylvain: that's what I've done, basically
[17:52:25] *** dmaclach has joined #chromium
[17:52:27] <nsylvain> ok
[17:52:35] <agl> danno_: probably not too late, ask jochen__
[17:52:47] <jochen__> i landed mine
[17:52:53] <jochen__> what should i tell him?
[17:52:57] <jochen__> he sits right next to me
[17:53:00] *** fqian has joined #chromium
[17:53:01] <agl> nsylvain: I sshed into the bots, commented out the crostools from /b/slave.DEPS/DEPS and run gclient a couple of times, manually
[17:53:14] <nsylvain> great, thanks!
[17:53:14] <agl> jochen__: if you can get both in at the same time, all the better
[17:53:53] <agl> nsylvain: I don't really know what I'm doing. Hopefully the DEPS files is the same on every bot because I overwrote it on each.
[17:54:07] <agl> nsylvain: the list of bots that I touched: codg181 codf232 172.22.15.147 172.22.15.157 codf182 codf231 codg130 codg131 codf146 codf153 codf154 codf147 codg138 codf151 codf150 codf188 codg185 codg161 codf149 codf230
[17:54:35] <agl> nsylvain: when the git server is running again it should be easy to copy the old DEPS file to each of them. I think that's all that would be needed.
[17:54:36] <nsylvain> We're asking a tech to go look at the machine, our 3 normal ways of rebooting it are not working. (Ssh not answering, Can't login for remote KVM,  remote power reboot is broken (unrelated I think, it was working yesterday) )
[17:54:50] <nsylvain> DEPS file should be the same, yes
[17:55:17] <nsylvain> Might be easier to just run a svn revert on it
[17:56:08] <mmoss> agl, nsylvain: I can do that when we're ready
[17:56:12] <agl> nsylvain: I didn't know that the file was maintained in svn :)
[17:58:29] <nsylvain> mmoss: I'd wait a day or two.. The git server might die again!
[17:58:55] <mmoss> nsylvain: then we need to do something with the arm bots because they aren't updating now
[17:59:10] <mmoss> maybe agl should check his manual change into svn
[17:59:14] <nsylvain> yeah
[17:59:44] <agl> I'm happy to, although I don't know where the SVN is
[17:59:45] <nsylvain> (and the exact one, otherwise it will do a SVN conflict, and that will cause yet another set of problems)
[17:59:51] <nsylvain> i'll paste you the link
[18:00:21] *** dmaclach has quit IRC
[18:00:30] *** fusion44 has quit IRC
[18:00:37] *** dmaclach has joined #chromium
[18:02:47] <danno_> agl: i just landed my grd cl, too.
[18:02:54] *** fqian has quit IRC
[18:03:09] *** xji has joined #chromium
[18:03:09] *** ChanServ sets mode: +v xji
[18:03:38] <agl> danno_: ok
[18:06:41] *** bweinstein has quit IRC
[18:06:56] *** tonyg-cr has left #chromium
[18:07:08] *** tonyg-cr has joined #chromium
[18:07:08] *** ChanServ sets mode: +v tonyg-cr
[18:07:35] *** bweinstein has joined #chromium
[18:08:38] *** karpar has joined #chromium
[18:08:58] <agl> rebooting Mac10.5 Perf (2) appears to have fixed the startup tests regression.
[18:09:05] *** RT|Chatzilla has quit IRC
[18:09:20] *** xji has quit IRC
[18:13:23] *** bweinstein has quit IRC
[18:14:13] *** victorw has joined #chromium
[18:14:14] *** ChanServ sets mode: +v victorw
[18:15:06] *** markmentovai has joined #chromium
[18:15:06] *** ChanServ sets mode: +v markmentovai
[18:16:37] <trungl-bot> Tree closed by nsylvain at google dot com: Tree is closed - rate limiting. Ask on IRC to land  - Waiting for linux to cycle green
[18:16:45] *** j^ has left #chromium
[18:17:23] *** fta has joined #chromium
[18:18:32] <nsylvain> should we open the tree, or was there any other things to fix first
[18:19:48] <dmaclach> do we know what's happening with the Mac Tests?
[18:20:16] <agl> nsylvain: it's lunch time here, so I'm heading off.
[18:20:19] <agl> mmoss: you have the bridge
[18:20:22] <dmaclach> (got on duty a little while ago, and still trying to catch up)
[18:20:31] <mmoss> agl: thanks
[18:20:43] <agl> nsylvain: we're have two GRD changes checked in at the moment
[18:20:58] <agl> nsylvain: ( jochen__ and danno_'s GRD changes)
[18:21:01] <nsylvain> ok
[18:21:06] <agl> nsylvain: so we probably need to sync for them.
[18:21:10] <pinkerton> did the toolbar get darker?
[18:21:17] <rsesek> yes
[18:21:21] <rsesek> andybons did that
[18:21:28] <pinkerton> ok, just checking
[18:22:47] * pinkerton sees a busted folder path in the chrome folder
[18:22:57] <pinkerton> "broswer/chromeos/cors"
[18:22:58] *** huanr has joined #chromium
[18:22:58] *** ChanServ sets mode: +v huanr
[18:23:21] <pinkerton> xcode shows it as red, clearly because it doesn't exist
[18:23:50] <rsesek> rohitrao: are you bisecting that new plugin bug or should I?
[18:23:54] <huanr> mmoss: can I land  http://codereview.chromium.org/2946017/show
[18:24:01] *** kurrik has joined #chromium
[18:24:01] *** ChanServ sets mode: +v kurrik
[18:24:01] <jibot> kurrik is Arne Roomann-Kurrik and wants you to help define the next iteration of OpenSocial APIs ( http://rurl.org/jy2 )
[18:24:43] <rohitrao> rsesek: go for it.  I was just syncing back until I found a build that worked :)
[18:24:49] *** trungl has joined #chromium
[18:24:49] *** ChanServ sets mode: +v trungl
[18:25:10] <phajdan-jr|afk> I'd like to land http://codereview.chromium.org/3011001, please let me know when there's best time for that. Change is trivial.
[18:25:14] *** phajdan-jr|afk is now known as phajdan-jr
[18:25:14] <mmoss> huanr: yes
[18:26:04] <markmentovai> welcome back, pinkerton
[18:26:07] <pinkerton> thx
[18:26:17] *** Aria has quit IRC
[18:26:44] <danno_> mmoss: now that both builders have started on my grd change, should I commit a whitespace change?
[18:27:12] <mmoss> danno_: they should get kicked by huanr change
[18:27:57] <huanr> yes. I just landed
[18:29:32] *** madhav has joined #chromium
[18:29:35] *** andybons has joined #chromium
[18:29:35] *** ChanServ sets mode: +v andybons
[18:29:45] <andybons> pinkerton: pinnnnnnk
[18:29:51] <pinkerton> heya
[18:30:05] <pinkerton> i like the new toolbar look, just pulled down the latest build
[18:30:25] <andybons> pinkerton: thanks. i'm just following order ;)
[18:30:28] *** abarth|zZz is now known as abarth
[18:30:31] <pinkerton> andybons: did you get anywhere with the bezel/outline on the icons themselves?
[18:30:36] <andybons> pinkerton: i need some help from a vet on this
[18:30:45] <pinkerton> pixels and i rarely get along
[18:30:49] <madhav> any native client developer around
[18:30:51] <vandebo> Can I land this cleanup CL? http://codereview.chromium.org/3013001
[18:30:52] <madhav> ?
[18:31:19] *** andrix has quit IRC
[18:31:31] <andybons> pinkerton: i'd like to know why the bookmark bar's gradient seems to get lighter as you go down the y-axis
[18:31:34] <dmaclach> andybons: what's the issue?
[18:31:41] <andybons> so it looks convex
[18:32:06] <andybons> when it should just get darker like the toolbar does as it goes down
[18:32:43] <andybons> so i'm in bookmark_bar_toolbar_view.mm:35
[18:32:43] <pinkerton> how are you drawing the gradient? maybe you just need to reverse it?
[18:32:47] <pinkerton> looking
[18:32:52] *** andrix has joined #chromium
[18:33:31] <rohitrao> is the gradient not intentional?
[18:33:35] <rohitrao> i was wondering that yesterday
[18:33:48] <andybons> rohitrao: the gradient should not go from dark to light
[18:34:10] <rohitrao> it looked like it was darkest in the middle, and lighter on both the top and bottom
[18:34:11] <pinkerton> i was wondering about that myself
[18:34:14] <pinkerton> right
[18:34:23] <pinkerton> actually, i think it looks kinda neat that way :D
[18:34:29] <pinkerton> but i'm sure i'm in the minority
[18:34:37] <andybons> rohitrao: it should follow the convention of getting darker as it goes down the y-axis so that the light source seems like it's coming from the top of the screen.
[18:34:59] <pinkerton> avi is the one who best understands how all the theme gradients interact, you might try asking him
[18:35:00] <rohitrao> oh, i wasn't disagreeing :)
[18:35:03] <rohitrao> i bet we
[18:35:11] <rohitrao> we're just drawing the gradient from the wrong place
[18:35:20] *** sbyer has left #chromium
[18:35:31] <rohitrao> background_gradient_view.mm might be a reasonable place to look
[18:35:40] <rohitrao> I'm pretty sure that's where the toolbar drawing code is
[18:35:41] <stuartmorgan> rohitrao: I doubt it, since the colors match
[18:35:46] <andybons> rohitrao: i'm totally aware that this isn't a preference issue :). when i'm terse i'm just trying to not waste people's time, ya know? so i write short stuf
[18:35:53] <andybons> rohitrao: also screw you and your ideas.
[18:36:10] <andybons> rohitrao: yup. in that as well
[18:36:35] *** malavv has quit IRC
[18:36:44] <pinkerton> this doesn't work on 10.4, clearly you all suck
[18:36:51] <andybons> rohitrao: but we use that for the toolbar and tab strip as well. why would it suddenly flip?
[18:37:04] <andybons> I WANT PPC SUPPORT
[18:37:36] <pinkerton> maybe that view is flipped?
[18:37:36] <andybons> and why are we using #define's for constants in that file. belugh
[18:38:47] <pinkerton> wait, if it goes light->dark won't we then have two toolbars that go light->dark->light->dark?
[18:38:50] *** loislo has quit IRC
[18:38:55] <pinkerton> and won't that look really odd?
[18:39:03] <trungl> andybons: I want 68k support
[18:39:09] <stuartmorgan> pinkerton: that's why I'm saying this doesn't look like a flip accident
[18:39:10] *** bweinstein has joined #chromium
[18:39:14] <rohitrao> two toolbars?
[18:39:20] *** mirandac has left #chromium
[18:40:22] <stuartmorgan> It would have to go dark->darker, which starts to get into the potential favicon issues we had with Camino
[18:40:31] *** rniwa has joined #chromium
[18:40:40] *** fahadsadah has quit IRC
[18:40:40] *** rniwa has quit IRC
[18:40:51] <andybons> stuartmorgan: what favicon issues?
[18:40:55] *** joshia has joined #chromium
[18:40:55] *** ChanServ sets mode: +v joshia
[18:41:14] <stuartmorgan> andybons: basically, the darker the background the more favicons look like garbage
[18:41:25] <stuartmorgan> Because they are generally designed to show on a white background
[18:42:06] *** xji has joined #chromium
[18:42:06] *** ChanServ sets mode: +v xji
[18:42:15] <vandebo> mmos: Can I land this cleanup CL? http://codereview.chromium.org/3013001
[18:42:26] <vandebo> mmoss ^^
[18:42:31] <pinkerton> stuartmorgan: but we already have that problem with most themes, so it's sorta moot
[18:42:35] <andybons> stuartmorgan: gotcha. well let's address that point when we come to it
[18:42:53] <stuartmorgan> andybons: yeah, it might not be quite dark enough yet to be a problem
[18:43:09] <andybons> stuartmorgan: i'm trying to get favicons removed from the bookmark bar anyway ;) more space for actual content. but that's a while away.
[18:43:12] <stuartmorgan> pinkerton: Just because some themes look bad doesn't mean the default should :P
[18:43:14] <pinkerton> rsesek: is the "buttons in the wrench menu" landed, or is it behind a flag?
[18:43:20] <andybons> pinkerton: landed
[18:43:27] <andybons> bam. beat rsesek to the punch.
[18:43:40] <stuartmorgan> andybons: except that's an anti-feature for a lot of people
[18:43:42] <pinkerton> then how come i don't see it?
[18:43:46] <stuartmorgan> andybons: favicons = more space
[18:43:58] <mmoss> vandebo: win trybot failure is just flakey?
[18:44:03] <rsesek> pinkerton: it juuuust landed
[18:44:06] <pinkerton> ahhh ok
[18:44:11] <rsesek> 52485
[18:44:12] <andybons> stuartmorgan: so do you know why the gradient is like that and how i can experiment with making it look correct.
[18:44:12] <pinkerton> tree closure and all, i guess
[18:44:23] <pinkerton> ok, i'm running 52480
[18:44:36] <vandebo> mmoss: yes
[18:44:38] <joshia> mmoss, dmaclach, tyoshino: can I land chrome frame CL: http://codereview.chromium.org/2936010 ?
[18:44:38] <stuartmorgan> andybons: I don't know where the code is, no, I just don't think it was someone accidentally drawing upside down
[18:44:54] <mmoss> vandebo: ok
[18:45:37] <pinkerton> andybons: can you print out the NSGradient that's used to draw that bar and see what it's start/stop colors are?
[18:45:42] *** sbyer has joined #chromium
[18:45:42] *** ChanServ sets mode: +v sbyer
[18:45:54] <andybons> sure. hang on
[18:46:01] <pinkerton> andybons: and see if it's somehow different from the toolbar code?
[18:46:23] <pinkerton> (though both should be BackgroundGradientViews, iirc)
[18:46:53] *** maikmerten has joined #chromium
[18:47:15] *** shreyas_ has joined #chromium
[18:47:28] <mmoss> joshia: win trybot failures?
[18:47:56] *** dhollowa has left #chromium
[18:48:03] *** dhollowa has joined #chromium
[18:48:04] *** ChanServ sets mode: +v dhollowa
[18:48:33] <andybons> pinkerton: yeah they're both backgroundgradientviews
[18:48:34] <joshia> mmoss: not related, they are in net unit tests
[18:48:49] <mmoss> joshia: ok
[18:48:55] <pinkerton> andybons: if you apply a theme, does it draw properly across both toolbars?
[18:48:58] <joshia> mmoss: thx!
[18:49:22] *** estade_laptop has quit IRC
[18:50:16] <andybons> pinkerton: yup. but most themes don't have gradients. actually, do any?
[18:50:50] <pinkerton> no, just making sure something else isn't wrong at a more fundamental level, i guess
[18:51:05] <andybons> pinkerton: no this is an issue that's been there since i started
[18:51:35] <andybons> pinkerton: it just becomes more visually obvious when the activated state's toolbar gradient is a bit darker
[18:51:39] <pinkerton> ok
[18:51:39] <mmoss> dmaclach, tyoshino, huanr: I'll fix the Chromium Linux Builder x64 update failure
[18:52:03] <andybons> pinkerton: brb. phone screen
[18:52:37] *** fahadsadah has joined #chromium
[18:53:16] <rohitrao> andybons: oh, this has been broken forever?
[18:53:24] <andybons> rohitrao: correct.
[18:53:30] <andybons> rohitrao: not new.
[18:53:31] *** davirtavares has quit IRC
[18:53:50] <rohitrao> andybons: ok, I was wondering how we managed to break it recently without changing any drawing code :)
[18:53:54] *** malavv has joined #chromium
[18:54:13] <satish_> Hello chromium! Is there a way (in browser process code) to make a POST request to a URL and stream the post data in the request body as multiple chunks over time? The use case is to send recorded audio from the browser to a remote server as the audio gets recorded, instead of waiting for it to stop and send a single chunk.
[18:55:11] <andybons> rohitrao: nah. always broken. like my brain.
[18:55:41] *** jshin has joined #chromium
[18:55:41] *** ChanServ sets mode: +v jshin
[18:57:07] <thakis> japhet: can you r+ https://bugs.webkit.org/show_bug.cgi?id=42214 ? it has an lgtm from agl, one of the skia heroes
[18:58:39] <stuartmorgan> satish_: if this is native code, why not use sockets or similar, where you have persistent connection?
[18:59:29] *** stoyan has joined #chromium
[18:59:30] *** ChanServ sets mode: +v stoyan
[18:59:40] *** jmcantrell has joined #chromium
[19:00:15] <satish_> stuartmorgan: It is native code so yes sockets are an option. I was looking at http and chunked transfer encoding first
[19:00:44] <jmcantrell> i'm seeing an issue with the mac dev version. can anyone tell me if this is a known issue?
[19:00:58] *** aroben has joined #chromium
[19:00:58] <jibot> aroben is a Safari/WebKit engineer and is an Aerobed and is pippity-poppity and the enforcer of null-checking, period
[19:01:06] <aroben> jibot:!
[19:01:14] <dmaclach> jmcantrell: what's the issue?
[19:02:08] <phajdan-jr> dmaclach: I've seen people that asked for landing later than me have landed their patches. Can I land as well?
[19:02:20] *** jamesr has joined #chromium
[19:02:20] *** ChanServ sets mode: +v jamesr
[19:02:20] <stoyan> Sherrifs: May I land http://codereview.chromium.org/2893015/show?  Two and a half ChromeFrame related lines.
[19:02:26] <phajdan-jr> dmaclach: http://codereview.chromium.org/3011001. Change is trivial.
[19:02:37] <jmcantrell> if you have bookmarks in the bookmarks bar that have no name, the sides of the icons are cut off. i only see this in the mac version, not windows or linux.
[19:03:30] <dmaclach> phajdan-jr: we were just waiting on the linux bots to go green
[19:04:06] <pinkerton> jmcantrell: can you file a bug?
[19:04:28] <jmcantrell> pinkerton: sure. i was just going to see if it was already a known issue
[19:04:45] <phajdan-jr> dmaclach: okay, does it mean I can land or should wait more? The tree status still says you're rate limiting.
[19:04:58] <pinkerton> jmcantrell: don't think it is, it doesn't ring a bell. maybe search the bug db first to be safe
[19:05:25] <dmaclach> phajdan-jr: go ahead.
[19:05:29] *** bent-mozilla has quit IRC
[19:05:32] *** Aliv3 has joined #chromium
[19:05:35] <Aliv3> ey
[19:05:40] <Aliv3> hello everyone
[19:05:47] <satish_> stuartmorgan: just found out can't use plain sockets since server only supports http
[19:05:50] <phajdan-jr> dmaclach: thanks, committing right now
[19:05:54] <Aliv3> where can I download a full .ISO copy of chromium os?
[19:06:11] <stuartmorgan> Aliv3: wrong channel; see topic
[19:06:37] <Aliv3> ooo
[19:06:38] <Aliv3> mmk
[19:06:50] *** Aliv3 has left #chromium
[19:07:58] <agl> mmoss: I'm on the x64 builder failure
[19:08:04] <dmaclach> thanks agl
[19:08:18] <mmoss> agl: from huanr?
[19:08:21] <agl> oh, actually looks like somebody has already fixed it
[19:08:23] <mmoss> already fixed
[19:08:33] <agl> mmoss: cool, no worries then.
[19:08:51] <mmoss> agl: you can take a break now, sir :)
[19:08:55] <joshia> mmoss: one more chroem frame patch:  http://codereview.chromium.org/2893015/show OK to go in?
[19:10:07] *** fta has quit IRC
[19:10:14] <mmoss> joshia: ok
[19:10:24] *** hebz0rl has quit IRC
[19:10:42] *** mabbo has quit IRC
[19:11:01] <phajdan-jr> mmoss: OK to land http://codereview.chromium.org/2985012/show ?
[19:11:03] *** janm has quit IRC
[19:11:17] <mmoss> phajdan-jr: lg
[19:11:49] <jmcantrell> http://code.google.com/p/chromium/issues/detail?id=49165
[19:12:39] *** janm has joined #chromium
[19:13:17] *** timsteele has joined #chromium
[19:13:17] <jibot> timsteele is a Chromium Googler.
[19:13:20] <phajdan-jr> mmoss: thanks, landed
[19:13:28] *** csilv has joined #chromium
[19:13:28] *** ChanServ sets mode: +v csilv
[19:13:47] *** davirtavares has joined #chromium
[19:15:58] *** thakis_ has joined #chromium
[19:15:58] *** ChanServ sets mode: +v thakis_
[19:16:42] *** jmcantrell has quit IRC
[19:17:54] <eglaysher> mmoss: may I commit http://codereview.chromium.org/2965015 ? All the ui_test failures appear to be of the form (server_.get() != NULL) and net_unittests.Connect failed with a bunch of patches on the windows builder last night.
[19:18:26] <eglaysher> (Also, the patch is only shuffling header definitions. It compiling should == working.)
[19:18:36] <timsteele> g' morning... I've got a change that touches a GRD that needs landing... any eta on when that might be possible?  (trybots had no issues with it).
[19:18:51] <chase> pkasting, pathorn: ping
[19:19:44] *** thakis_ has quit IRC
[19:19:44] <mmoss> eglaysher: yeah, now is probably a good time, I'll hold off on more until that one works through a bit
[19:20:06] <agl> mmoss: I've a comment only change (http://codereview.chromium.org/2805082/show) that I'd like to land
[19:20:21] <mmoss> agl: quick, before eglaysher
[19:20:30] *** duffydack has joined #chromium
[19:20:47] *** gavinp2 has joined #chromium
[19:20:47] *** ChanServ sets mode: +v gavinp2
[19:21:03] <gavinp2> I just want to make sure: every linux buildserver having "update scripts failed", that's OK, right?
[19:21:26] <mmoss> gavinp2: they're clearing out
[19:21:52] <mmoss> gavinp2: a couple are still bitching, due to an apparent gclient anomoly, but the core failure is fixed
[19:22:02] <gavinp2> awesome mmoss, thanks
[19:22:10] <mmoss> I'm manually clearing the ones that are still unhappy
[19:22:12] <rohitrao> shess: for this popup-resizing-animation bug, what's actually broken?
[19:22:13] *** madhav has left #chromium
[19:22:14] <gavinp2> so perhaps try will work more reproducably in the near future
[19:22:30] <gavinp2> i'm pretty sure some builds failed testing because of bad partial updates, although i'm far from sure.
[19:22:36] <rohitrao> shess: I see a slight lag before the popup resizes, but I don't see it getting completely stuck at the wrong size
[19:22:59] <mmoss> gavinp2: probably unrelated -- the failure on the main waterfall had to to with crostools, which most things don't use
[19:22:59] *** karpar has quit IRC
[19:23:17] <shess> rohitrao: hmm.  Could this be 10.5?  What I see is that if I make the window bigger, the popup stays smaller until I release the mouse button.
[19:23:22] <gavinp2> mmoss, oh.  hrm.  then i guess i'll run more tests locally to be sure
[19:23:24] <shess> [I mean 10.5-only]
[19:23:24] <gavinp2> this is interesting
[19:24:04] *** akalin has quit IRC
[19:24:04] <rohitrao> shess: and this is only if you start resizing while an animation is running?
[19:24:05] <shess> rohitrao: it does _not_ happen for my in the beta channel, though.
[19:24:07] *** akalin has joined #chromium
[19:24:09] <agl> mmoss: landing now
[19:24:22] <mmoss> agl: ok
[19:24:57] <shess> rohitrao: type to get the popup up, then resize to make the main window wider, and the popup does not follow along.  I don't know if there's an animation running (probably not), but the existing code is coded as if there is potentially an animation running.
[19:25:03] <agl> mmoss: opps, that might not have been quick enough. The git tree hadn't caught up so it looked like erg hadn't committed.
[19:25:07] <timsteele> mmoss:  i
[19:25:15] <rohitrao> shess: ok, I'm not seeing that then.  Let me try 10.5 :)
[19:25:17] <timsteele> bah. hit enter too soon,
[19:25:25] *** dmaclach has quit IRC
[19:25:42] *** dmaclach has joined #chromium
[19:25:42] *** ChanServ sets mode: +v dmaclach
[19:25:44] <timsteele> mmoss: guessing Im out of luck for a while huh :(
[19:25:46] <pkasting> chase: pong
[19:25:52] *** ebraminio has joined #chromium
[19:25:59] <mmoss> timsteele: yeah, let's wait a bit
[19:26:03] *** dmaclach has quit IRC
[19:26:19] <mmoss> not that I don't trust erg, but it is the 4th attempt :)
[19:26:43] <mmoss> and things could use a moment to catch up anyhow
[19:26:45] *** dmaclach has joined #chromium
[19:26:45] *** ChanServ sets mode: +v dmaclach
[19:26:50] <eglaysher> ;_;
[19:27:12] <pkasting> mmoss: I have two changes I'd like to land, one of which is a grd change
[19:27:21] <shess> rohitrao: BTW, I consider this a polish bug, so would be perfectly fine not worrying about it until later.
[19:27:33] *** chrisccoulson has quit IRC
[19:27:37] *** mabbo has joined #chromium
[19:27:40] <rohitrao> k
[19:27:42] <shess> [I mean WRT my other code review :-).]
[19:28:05] *** rafaelw has joined #chromium
[19:28:06] *** ChanServ sets mode: +v rafaelw
[19:28:09] <mmoss> pkasting: doh
[19:28:30] *** hrna has joined #chromium
[19:28:49] <mmoss> calling all grd changes, batch 'em up with pkasting if ya got 'em
[19:28:50] <pkasting> mmoss: I think landing it without closing the tree would be faster, as long as I watch the aftereffects carefully so I can distinguish real bustage from fake bustage
[19:29:33] <mmoss> pkasting: we can land while it's throttling, since red is still clearing, but let's give erg's a chance to run a moment
[19:30:05] <pkasting> mmoss: Sure.  What about my other change?  Do you want me to land it first/simultaneously/later?
[19:30:58] *** satish_ has quit IRC
[19:31:00] <mmoss> pkasting: I think together, since they're both yours. I'll leave it to you to distinguish which breaks what
[19:32:03] *** leiz has quit IRC
[19:32:23] *** chrisccoulson has joined #chromium
[19:32:54] *** isaacvv has joined #chromium
[19:33:17] *** rsimha has joined #chromium
[19:33:33] *** jamesr has quit IRC
[19:34:29] <timsteele> mmoss, pkasting: I'll be joining you on the grd bandwagon ride.
[19:34:54] <mmoss> timsteele: cool, maybe we can just do one today
[19:36:05] <stuartmorgan> rohitrao: do we know yet whether we are pulling the plug on CA Flash for M6?
[19:36:22] <dmaclach> mmoss: optimist ;-)
[19:36:25] <mmoss> timsteele, pkasting: ok, looks like windows builders are going. submit grd now?
[19:36:36] <rohitrao> stuartmorgan: I'm trying to make something work, but it's not looking good :(
[19:36:55] <davidben> I've got a grd change I'd like to get in as well.
[19:37:01] <rohitrao> stuartmorgan: turning on layers for everything is our best bet, but that breaks drawing for pretty much every view
[19:37:08] <mmoss> davidben: ok
[19:37:12] <stuartmorgan> rohitrao: seems pretty late in the game; I say we just call it
[19:37:15] <pkasting> timsteele: I'm just syncing my repo
[19:37:26] *** apatrick_ has left #chromium
[19:38:13] <thakis> i don't like layers for everything. i believe i hadn't said that yet for today
[19:39:10] <timsteele> pkasting: i'm synced up, are you changing generated_resoruces? if so Ill let you go first and I'll sync it up
[19:39:25] <pkasting> timsteele: No, I'm changing theme_resources
[19:39:33] <pkasting> Still syncing
[19:39:46] *** huanr has quit IRC
[19:40:11] <rohitrao> thakis: what else do we do? :)
[19:40:33] * pinkerton blames vtl for his valgrind bug
[19:40:34] <thakis> we test if child windows work
[19:40:55] *** ebraminio has quit IRC
[19:41:17] <thakis> ojan: ping
[19:41:25] * trungl is glad it's vtl's fault and not his.
[19:41:45] <rohitrao> thakis: http://codereview.chromium.org/2944017/show .  The biggest problem is keeping the browser window main
[19:41:51] *** Barkhorn has joined #chromium
[19:42:05] <davidben> pkasting: So, how do these end up working? (Or should I wait until a different batch?) Never actually gotten a chance to do a grd change yet. I am changing a couple strings in generated_resources.
[19:42:08] <rohitrao> thakis: and then there are a whole bunch of random other details
[19:42:25] <pkasting> davidben: Are you r+ and ready to land?
[19:42:43] * dmaclach watches rohitrao's descent into madness
[19:42:50] *** legion13 has joined #chromium
[19:42:56] <davidben> r+?
[19:43:00] *** Barkhorn has left #chromium
[19:43:24] <pkasting> davidben: You have a signed-off review
[19:43:26] *** legion13 has left #chromium
[19:43:34] <rohitrao> dmaclach: bah
[19:43:50] *** legion13 has joined #chromium
[19:43:52] <davidben> Yeah. Wan-teh reviewed this nearly 3 weeks ago, but the tree keeps going crazy. :-) http://codereview.chromium.org/2859026/show
[19:43:54] <dmaclach> rohitrao: make sure to test expose and spaces. Both of which can cause issues with the multiple window thing. Also, if you are floating the child window to a different layer, make sure alerts can't get caught behind it.
[19:44:11] *** lalalalilili has left #chromium
[19:44:19] <timsteele> pkasting: btw, your fix on the gaia login html... did you happen to check if the throbber now shows up on the _next_ screen in the setup flow (the 'customize chosen data types') screen?
[19:44:58] <timsteele> 'cause there's one there as well (choose_datatypes.html), and it was still working while gaia_login.html was busted
[19:45:16] <pkasting> timsteele: It was working because I had failed to change the path in it
[19:45:31] <pkasting> timsteele: It now works like gaia_login.html
[19:46:02] <pkasting> timsteele: I verified that the inliner did the right thing in it, I didn't actually hit OK on a sync setup to see the image appear
[19:46:17] <timsteele> oh, pff, I forgot that file was part of the patch.
[19:46:18] <timsteele> okay
[19:46:19] <timsteele> cool!
[19:46:22] <pkasting> timsteele: But since the image was inlined the only way it could not work is if I broke something different :)
[19:46:32] <timsteele> heh
[19:46:40] <pkasting> davidben: Is your checkout synced?
[19:46:41] <jochen__> btw, i think the SelfDeletePluginGetUrl test is flaky
[19:47:25] <davidben> pkasting: Yup. Just rebased a moment ago.
[19:47:48] <pkasting> timsteele: How about davidben land and then you sync and land as well?
[19:48:54] <pinkerton> trungl: small cl to you for a valgrind issue
[19:49:10] <trungl> pinkerton: looking
[19:49:20] <timsteele> pkasting: okay
[19:49:34] <pkasting> davidben: Go ahead and land, and post here as soon as you do
[19:50:02] <davidben> pkasting: Done.
[19:50:20] <pkasting> I'm also landed
[19:51:25] *** eroman has joined #chromium
[19:51:25] *** ChanServ sets mode: +v eroman
[19:51:26] <jibot> eroman is eric roman and works on chromium
[19:51:29] <jrg> dmaclach mmoss: would like to submit small Mac-only CL (http://codereview.chromium.org/2958016).  Is that OK?
[19:51:34] *** maikmerten_ has joined #chromium
[19:51:36] <timsteele> done
[19:51:52] * pinkerton wonders why jrg is emboldened today
[19:51:54] *** fusion44 has joined #chromium
[19:51:58] * jrg wonders as well
[19:52:06] <mmoss> jrg: bunch of grd changes just landed, let's wait a bit
[19:52:11] <jrg> mmoss: OK
[19:52:15] <pkasting> timsteele: You missed the train
[19:52:17] <jrg> pinkerton: welcome back
[19:52:19] <pinkerton> thx
[19:52:45] <eroman> i am having problems building Release on mac -- getting errors from the sync protobuffer stuff. anyone have a solution?
[19:53:12] <eroman> (actually same thing in debug; cant build on mac)
[19:53:13] <timsteele> geez :( I was ready to sync up generated_resources, but then your change had sync conflicts so it took an extra second
[19:53:20] <timsteele> well not conflicts
[19:53:27] <timsteele> more out of date files
[19:53:49] <phajdan-jr> eroman: is it a very recent thing? My build from about 10:30 on Mac is fine (Debug).
[19:53:54] <eglaysher> arm failed?
[19:53:57] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed - rate limiting + grd changes + linux cycling green. Ask on IRC to land.
[19:54:13] *** maikmerten has quit IRC
[19:54:25] * phajdan-jr suggests removing "ask on IRC to land" from the status. It's probably not going to happen.
[19:54:40] <eroman> phajdan-jr: I had this issue yesterday too, and synced again today. I recently went to 10.6 as well in case that has an impact
[19:54:47] *** Viclame has left #chromium
[19:54:47] <pinkerton> trungl: yes, you'd think the compiler would have noticed that
[19:55:10] <phajdan-jr> eroman: not sure then, it just looks really weird :(
[19:55:15] *** shreyas_ has quit IRC
[19:55:26] <trungl> thakis: would llvm have noticed this bug: http://codereview.chromium.org/3006002/show ?
[19:55:33] <dmaclach> eroman, I just did a sync and build on 10.6 without any problems
[19:55:35] <eglaysher> I am comitting a one line fix to the chrome arm compile.
[19:55:51] <ojan> dear sheriffs, i'd like permission to land http://codereview.chromium.org/2878030/show
[19:56:03] <ojan> it's very safe and last passed the try bots
[19:56:28] <ojan> dmaclach, mmoss, tyoshino: ^^^^^
[19:56:32] <pkasting> ojan: Seems fine to me
[19:56:41] *** fqian has joined #chromium
[19:56:55] <thakis> trungl: not the compiler. maybe the dataflow analysis part
[19:56:59] <trungl-bot> Tree closed by erg at chromium dot org (:eglaysher): Tree is closed - rate limiting + grd changes + linux cycling green. Ask on IRC to land. + chrome_arm -> erg
[19:56:59] <ojan> pkasting: are you acting sheriff?
[19:57:01] <thakis> (but that doesn't work with c++ yet)
[19:57:06] <eroman> phajdan-jr: dmaclach: ok good to know, I will try nuking my build output again see if it helps.
[19:57:06] <fusion44> hey guys, is there an official native client channel?
[19:57:12] <thakis> ojan: ping
[19:57:20] <ojan> thakis: what's up?
[19:57:31] <thakis> ojan: can you r+ https://bugs.webkit.org/show_bug.cgi?id=42214 ? it has an LG from one of the skia heroes
[19:57:31] <dmaclach> ojan: can you hold off until we cycle a bit? I assume this isn't blocking.
[19:58:04] <ojan> dmaclach: it's not blocking yet. eventually it'll block the webkit merge. but that's still a good ways out. just let me know when.
[19:58:11] <dmaclach> sounds good. thanks.
[19:58:22] <ojan> thakis: looking
[19:58:46] <thakis> trungl: (i even checked for you!)
[20:00:00] <trungl> thakis: you mean that you checked that the compiler doesn't catch it?
[20:00:01] <trungl-bot> Tree closed by erg at chromium dot org (:eglaysher): Tree is closed - rate limiting + grd changes + linux cycling green. Ask on IRC to land. + chrome arm builder should cycle green
[20:00:09] <thakis> trungl: yes!
[20:00:15] * trungl knows nothing about how llvm is split up
[20:00:30] <trungl> thakis: why are you yelling so much today?!
[20:00:36] *** loislo has joined #chromium
[20:00:36] *** davemoore_ has joined #chromium
[20:00:36] *** ChanServ sets mode: +v davemoore_
[20:00:51] <pinkerton> shouldn't it know that you're using the variable you're declaring to initialize the variable you're declaring?
[20:00:53] <thakis> trungl: i'm not yelling, i try to pretend that i'm happy & excited
[20:01:00] *** shreyas_ has joined #chromium
[20:01:28] <pinkerton> i mean, it can tell that from the parse tree, no crazy data analysis needed.
[20:01:34] <thakis> pinkerton: it needs a control flow graph for this, and hence it's done by the analyzer
[20:01:46] <thakis> pinkerton: it could've been done conditionally (?:)
[20:01:54] <thakis> so it's a dataflow analysis
[20:02:14] <thakis> some compilers (javac) do some dataflow analyses, but c++ compilers usually don't
[20:02:21] <thakis> clang has an extra pass that does
[20:02:25] <thakis> which doesn't support c++ yet
[20:02:30] <davemoore_> it looks like erg's change to remove histogram.h from message_loop.h broken the cros build
[20:03:02] <davemoore_> it's a trivial fix
[20:03:32] <eroman> Any help appreciated with build error I am getting linker error in protoc"
[20:03:34] <rohitrao> that's not necessarily a bug, is it?
[20:03:35] <eroman> http://pastebin.com/3GzF4inn
[20:03:51] <eroman> (while building on vanilla mac checkout)
[20:04:25] <dmaclach> eroman: hooboy
[20:04:38] <dmaclach> why are you linking in libgcov?
[20:04:48] <dmaclach> the royal "you" ;-)
[20:05:22] <davemoore_> May I land this: http://codereview.chromium.org/2806053
[20:05:30] <eroman> dmaclach: dunno. like i said, this is a vanilla checkout on mac os 10.6, and i regenerated project files + deleted xcodebuild several times already. building the "Chrome" target from xcode
[20:05:31] <dmaclach> info about the gcov mess on 10.6 here: http://code.google.com/p/google-toolbox-for-mac/wiki/SnowLeopardGCov
[20:05:38] <dmaclach> eroman ^^^
[20:05:42] <davemoore_> to fix the cros breakage (anyone trying to build for cros is broken right now)
[20:06:17] <eroman> dmaclach: wait no i am an idiot!
[20:06:31] <eroman> dmaclach: I had set the gyp variable coverage = 1
[20:06:41] <eroman> but didnt realize it
[20:06:59] <dmaclach> eroman: np. If you want to do coverage, the link above will explain the joys of getting it set up
[20:07:15] <eroman> dmaclach: thankyou! I am sure that was the issue
[20:09:03] <willchan> mmoss, dmaclach, tyoshino: requesting permission to land http://codereview.chromium.org/2994003/show
[20:09:07] *** duffydack has left #chromium
[20:09:26] <mmoss> jrg, ojan: good to go
[20:09:33] <mmoss> willchan: next
[20:09:43] <willchan> ok, let me know
[20:09:50] <ojan> mmoss: thx
[20:10:10] <mmoss> willchan: can it wait for the win trybot to finish?
[20:10:37] *** janm has quit IRC
[20:10:40] <davemoore_> never mind...elliot fixed it already
[20:11:15] <willchan> mmoss: sure.  note the only diff between the previous run is a merge, but i'm happy to wait.  just wanted to get clearance first.
[20:11:54] *** apavlov has quit IRC
[20:13:08] *** jshin has quit IRC
[20:15:19] *** eseidel has joined #chromium
[20:15:25] *** bandu has joined #chromium
[20:15:41] <thakis> ojan: addressed comments at https://bugs.webkit.org/show_bug.cgi?id=42214
[20:15:52] <akalin> are windows trybots backed up today, too?
[20:15:58] *** bent-mozilla has joined #chromium
[20:16:31] *** bandu is now known as coyo
[20:18:59] *** legion13 has quit IRC
[20:20:26] *** cpu has quit IRC
[20:20:33] *** jshin has joined #chromium
[20:20:33] *** ChanServ sets mode: +v jshin
[20:21:55] <pkasting> The biggest things people could do to help the trybot backlog are to open nsylvain's email about which parts of the runs have gotten orders of magnitude slower, and fix those
[20:22:07] <akalin> okay
[20:22:20] <pkasting> IMO this will have an even larger effect than flipping the svn upload default from "try" to "no try"
[20:22:38] <pkasting> Because not only will it help the trybots, it will help the buildbots, and engineers' own test runs
[20:23:14] <pkasting> It sounded like there were multiple steps that have gone from 20 seconds to 10 minutes.  That's a huge savings if we can cut it back down.
[20:23:22] <jochen__> who's currently scheduling commits?
[20:23:39] <dmaclach> however adding --no_try appropriately to CL uploads is the easiest thing you can do while those tests are being fixed up.
[20:24:22] <jochen__> dmaclach: you're a sheriff :)
[20:24:51] <jochen__> dmaclach: please revert 52486 i think that's causing a lot of the trouble
[20:24:56] <dmaclach> jochen__: that's what my calendar told me ;-)
[20:25:11] <dmaclach> jochen__: looking
[20:25:46] *** dave_levin has joined #chromium
[20:26:30] <dmaclach> jochen__:  you blaming the windows redness on that?
[20:27:12] <jochen__> some of the tests failing are the once i introduced in this cl
[20:27:14] *** pdelgallego has quit IRC
[20:27:25] <jochen__> i can't tell why so better revert it
[20:27:32] *** Buglouse has quit IRC
[20:28:03] <jochen__> + it's 8:30pm here... i'm going home now
[20:29:24] *** janm has joined #chromium
[20:29:25] <dmaclach> jochen__:  drovering
[20:29:31] <pkasting> Vista unit test failure looks to be an expected transient from the .grd changes
[20:29:40] <dmaclach> thanks pkasting
[20:30:02] *** pinkerton has quit IRC
[20:31:50] *** jamesr has joined #chromium
[20:31:50] *** ChanServ sets mode: +v jamesr
[20:33:44] *** pkasting has left #chromium
[20:34:05] <rvargas> mmoss, dmaclach, tyoshino: requesting permission to land http://codereview.chromium.org/2891022/show
[20:34:18] *** novas0x2a has joined #chromium
[20:34:24] *** MikeW has joined #chromium
[20:34:28] *** pkasting has joined #chromium
[20:34:29] *** ChanServ sets mode: +v pkasting
[20:34:29] <jibot> pkasting is a well-known rabble-rouser and works on Google Chrome
[20:34:38] <pkasting> Why the hell is jibot in this channel
[20:35:11] <selckin> they wanted to race jibot vs trungl-bot
[20:35:16] <pkasting> XP unit test failure is identical to vista
[20:35:29] *** jorlow has quit IRC
[20:35:31] <trungl> what is jibot?
[20:35:37] <trungl> who is jibot?
[20:36:04] <MikeW> Hi all. Anyone have any ideas why Chrome 5.0.375.99 beta on Windows 7 would (On first load of the binary on every reboot) load the window frame, then spend between 30 and 90 seconds hitting my HDD hard before it's usable?
[20:36:07] *** gfeher has joined #chromium
[20:36:09] <dmaclach> rvargas: I just reverted jochen__ 's CL 52486. Can you hold off a bit?
[20:36:28] <rvargas> sure, just let me know
[20:36:56] <dmaclach> rvargas: will do
[20:37:40] <estade> dmaclach: can I land http://codereview.chromium.org/3003001/show along with rvargas
[20:37:48] <MikeW> I think I'll get Process Explorer and try and see what the release builds are touching on the disk
[20:39:22] *** malavv_ has joined #chromium
[20:39:27] *** malavv_ has quit IRC
[20:39:31] *** MikeW has left #chromium
[20:42:09] *** shreyas_ has quit IRC
[20:42:29] <mpcomplete> has anyone heard of AddRemoveAutocompleteTextFieldEditorTest ? my patch seems to be failing it on the trybot, but it doesn't seem to actually exist anywhere
[20:42:46] <alyssad> mmoss, dmaclach, tyoshino: requesting to land: http://codereview.chromium.org/2923010/show (whenever possible)
[20:42:53] *** bauerb has quit IRC
[20:43:31] *** sanjeevr has joined #chromium
[20:43:43] <pkasting> Interactive UI test failures are also .grd fallout
[20:44:46] <thakis> mpcomplete: sounds like something shess might know
[20:45:26] <dmaclach> alyssad: noted... we'll let you know
[20:45:55] <alyssad> thanks!
[20:46:05] <thakis> mpcomplete: it's in chrome/browser/cocoa/autocomplete_text_field_editor_unittest.mm (search for TEST_VIEW)
[20:46:18] <thakis> mpcomplete: (TEST_VIEW adds an AddRemove test)
[20:46:25] *** asdsad has joined #chromium
[20:46:39] <rohitrao> thakis: you beat me by 3 seconds :P
[20:46:50] <mpcomplete> thakis: ah.. thanks
[20:47:18] <mmoss> rvargas, estade, alyssad: go ahead, then I think we'll stop for lunch and make sure the grd redness clears out and give things a chance to clear up
[20:47:30] <thakis> jamesr: can you reply on https://bugs.webkit.org/show_bug.cgi?id=42214 please?
[20:47:47] <mmoss> it would be nice to at least see the release builders all green at one point today :(
[20:48:25] *** tonyg-cr has quit IRC
[20:48:39] *** lzheng has joined #chromium
[20:48:39] *** ChanServ sets mode: +v lzheng
[20:48:45] <rvargas> mmoss: thanks
[20:49:22] <jamesr> thakis: sure
[20:50:25] *** trungl has quit IRC
[20:51:18] <jamesr> thakis: do you have access to a SL box?
[20:51:28] <thakis> jamesr: nope
[20:51:32] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed - waiting for green.
[20:51:42] *** michaeln has joined #chromium
[20:51:42] *** ChanServ sets mode: +v michaeln
[20:51:49] *** trungl has joined #chromium
[20:51:49] *** ChanServ sets mode: +v trungl
[20:52:02] *** ananta_ has joined #chromium
[20:52:15] *** motownavi_ has quit IRC
[20:52:55] *** bmizerany has joined #chromium
[20:54:20] <nirnimesh> I deleted some dirs from my git repository. what's the command to get it back?
[20:54:35] *** lzheng has quit IRC
[20:54:39] <tfarina> perphas git reset --hard?
[20:54:55] <nirnimesh> I didn't commit
[20:55:16] <tfarina> git checkout .
[20:56:00] <nirnimesh> nope. that doesn't do it
[20:56:29] <selckin> it should
[20:56:46] <pkasting> XP/Vista tests definitely going green.  Not sure what Mac 10.5 browser test failure is.
[20:58:30] *** asdsad has quit IRC
[20:59:48] <dmaclach> pkasting: Mac builds have errors in different places. I'll keep an eye on them, but currently betting on flake.
[20:59:58] *** rajivts has quit IRC
[21:00:15] <pkasting> dmaclach: Is your revert of jochen's change hopefully going to clean up the Chromium XP UI test failure?
[21:00:20] <markmentovai> having a hard time opening the buildbot console again. great.
[21:00:25] <dmaclach> pkasting: that's the hope
[21:00:31] <pkasting> K
[21:01:33] *** podivilov has quit IRC
[21:01:51] *** jshin has quit IRC
[21:03:40] *** reiz has joined #chromium
[21:03:41] *** ChanServ sets mode: +v reiz
[21:03:51] <willchan> mmoss: can i land now?  win trybot passed
[21:03:59] <pkasting> nsylvain: Reliability bot may be in trouble?: http://build.chromium.org/buildbot/waterfall/builders/Chromium%20Reliability/builds/11041/steps/shell/logs/stdio
[21:04:14] <mmoss> willchan: ok
[21:04:17] <willchan> thx
[21:04:41] <mmoss> pkasting, nsylvain: I got it
[21:05:11] <mmoss> pkasting: there are actually a bunch like that. buildbot doesn't like the gclient warning for some reason
[21:06:26] *** lzheng has joined #chromium
[21:06:26] *** ChanServ sets mode: +v lzheng
[21:09:59] *** gman2 has joined #chromium
[21:09:59] *** ChanServ sets mode: +v gman2
[21:10:28] *** rniwa has joined #chromium
[21:10:29] *** ChanServ sets mode: +v rniwa
[21:11:47] *** JakeSays has quit IRC
[21:11:54] *** coyo|pingout has joined #chromium
[21:12:59] *** JakeSays has joined #chromium
[21:13:16] *** pdelgallego has joined #chromium
[21:13:23] <estade> tony^work: can you post your inspector targets patch to rietveld?
[21:13:35] <estade> and land it and all that
[21:13:46] *** apatrick_ has joined #chromium
[21:13:46] *** ChanServ sets mode: +v apatrick_
[21:14:03] <estade> compiling chrome = automatically clear all console scrollback :[
[21:14:31] <awong> estade: > /tmp/log 2>&1
[21:14:41] *** coyo has quit IRC
[21:17:14] *** Erkan_Yilmaz has quit IRC
[21:18:46] <pkasting> mmoss: Yeah, I just got that same problem on my linux trybot run.  Is there something we can do to fix all these?
[21:19:04] <mmoss> pkasting: working through them still
[21:19:11] <mmoss> haven't gotten to trybots yet
[21:21:16] *** pinkerton has joined #chromium
[21:21:16] *** ChanServ sets mode: +v pinkerton
[21:22:36] *** pinkerton has quit IRC
[21:22:44] *** pinkerton has joined #chromium
[21:22:44] *** ChanServ sets mode: +v pinkerton
[21:23:46] <pinkerton> has BrowserTest.ThirtyFourTabs gotten flaky on mac lately?
[21:25:03] <jrg> pinkerton: is it timing out?
[21:25:05] <pinkerton> yes
[21:25:22] <pinkerton> just happened on a trybot
[21:25:23] <jrg> pinkerton: I changed timeout from 30 to 45 seconds since I saw it taking ~31 seconds.
[21:25:34] <pinkerton> trybot is probably overloaded, eh?
[21:25:43] *** dob1 has left #chromium
[21:25:48] <jrg> pinkerton: perhaps.  I don't think the right answer is to make it >45 seconds.
[21:26:02] <pinkerton> me either, since clearly it doesn't always work
[21:26:10] <pinkerton> i just want to make sure i didn't break anything
[21:26:34] <jrg> pinkerton: you did not.  But if it stays flaky enough we should FLAKY_ it.
[21:26:43] <pinkerton> we might as well then
[21:26:45] <jrg> uh, re-FLAKY_ it.
[21:26:52] <pinkerton> for trybots, at least
[21:28:04] *** eseidel has quit IRC
[21:29:00] <pkasting> dmaclach: Heh, your revert of jochen's change is itself a .grd change that's going to take a while to cycle
[21:29:20] <dmaclach> pkasting: oi... didn't notice.
[21:29:28] <pinkerton> ha
[21:31:04] *** Erkan_Yilmaz has joined #chromium
[21:31:50] *** sanjeevr has quit IRC
[21:32:37] *** JodaZ has quit IRC
[21:32:54] *** cuba33ci has quit IRC
[21:33:27] *** mirandac has joined #chromium
[21:33:27] *** ChanServ sets mode: +v mirandac
[21:34:01] *** cuba33ci has joined #chromium
[21:34:03] *** estes has joined #chromium
[21:34:54] *** JodaZ has joined #chromium
[21:36:23] *** dumi has joined #chromium
[21:36:23] *** ChanServ sets mode: +v dumi
[21:36:49] *** maikmerten_ has quit IRC
[21:37:50] *** bmizerany has quit IRC
[21:38:24] *** mabbo has quit IRC
[21:39:25] *** eseidel has joined #chromium
[21:40:27] <phajdan-jr> I know we try to preserve svn metadata when copying or moving files (svn cp/mv). Now what should I do if I split a file (approx in half)? Any recommendations?
[21:40:27] *** MichealH has quit IRC
[21:40:39] *** [Nef] has joined #chromium
[21:40:46] *** bmizerany has joined #chromium
[21:42:24] <mmoss> anyone with a simple change?
[21:43:29] <shess> mpcomplete: I saw one of those, too (that test failing).  it greened on the next pass with no changes.
[21:43:48] <pinkerton> mmoss: i do!
[21:43:53] <mmoss> pinkerton: go
[21:43:57] <pinkerton> sweet
[21:44:38] <shess> trungl: compiler doesn't even catch: NSRect tabFrame; CGFloat y = NSMaxY(tabFrame); tabFrame = NSMakeRect(0, y, ...);
[21:44:53] *** davirtavares has quit IRC
[21:44:58] *** MichealH has joined #chromium
[21:45:10] <shess> trungl: count me disgusted.
[21:45:38] <trungl> what's the definition of NSMaxY(), anyway?
[21:45:49] <oshima> jrg, pinterton: it's actually crashing: http://build.chromium.org/buildbot/try-server/builders/mac/builds/38228/steps/browser_tests/logs/stdio
[21:45:56] *** mabbo has joined #chromium
[21:46:00] <oshima> jrg, pinkerton:
[21:46:19] <pinkerton> oshima: what's actually crashing?
[21:46:26] <shess> trungl: AFAICT, it was the obvious.
[21:46:30] *** gfeher has quit IRC
[21:46:44] <jrg> oshima: yep, looks like BrowserTest.ThirtyFourTabs is crashing.  That is a new regression I think.
[21:46:47] <shess> trungl: is not catching when I write it without NSMaxY(), either.
[21:46:51] <oshima> pinkerton: look at log
[21:47:07] <pinkerton> it's crashing across all trybots?
[21:47:11] <jrg> oshima: do you need help with this?
[21:47:14] <pinkerton> or on our testers too?
[21:47:45] <oshima> last two browser_tests failures are due to this crash.
[21:47:57] <trungl> "NS_INLINE CGFloat NSMaxY(NSRect aRect) { ... }" fwiw
[21:48:00] <pinkerton> if that's the case, how did our last known good build get updated to include this?
[21:48:05] <pinkerton> since it's on the trybots
[21:48:09] <awong> evnarm: if I compile something if -fvisibility=hidden, should no symbols show up in "nm -DC libblah.so"?
[21:48:40] <awong> evmar: ^^
[21:48:59] <shess> trungl: Compiler doesn't catch things if I write the entire block long-hand.
[21:49:02] <oshima> pinkerton: i'm just saying that they're crash but not timeouts
[21:49:06] <shess> trungl: that doesn't feel right to me.
[21:49:20] <eglaysher> mmoss: you were asking for simple changes. Is http://codereview.chromium.org/3013002 simple enough for you?
[21:49:36] <eglaysher> (hasn't passed trybot, though I can't imagine it being a problem)
[21:49:36] <trungl> shess: maybe it's not smart about structs
[21:50:02] <trungl> (since a struct could have a constructor)
[21:50:06] <mmoss> eglaysher: you can race pinkerton
[21:50:27] <shess> trungl: not true, and I'm also seeing it with basic types.
[21:50:34] <trungl> d'oh
[21:50:41] <trungl> ok, I'm out of hypotheses
[21:50:47] <trungl> "gcc is dumb"
[21:50:50] <thakis> awong: sounds right
[21:51:07] <shess> trungl: Need -O.
[21:51:13] <mmoss> pkasting: fyi, I cleaned all the Linux trybots I was able to access, let me know if you still see that error on new try runs
[21:51:17] <shess> trungl: Can't tell uninitialized w/o -O.
[21:51:20] <pinkerton> oshima: so just to be clear, if my trybot timed out on that test, i shouldn't worry because it's happening everywhere?
[21:51:40] <trungl> shess: of course, I've seen this sort of thing before
[21:51:44] <trungl> but it's rather stupid
[21:52:08] <shess> trungl: well, probably just a compiler pass which never happens unless -O1 or greater.  *sigh*
[21:52:24] <trungl> shess: but it's pretty horrible
[21:52:36] <thakis> trungl / shess: you talk like young'uns
[21:52:37] <shess> trungl: would make sense (to me) to run the trybots with -g -O1.
[21:52:40] <thakis> be happy you have a compiler at all
[21:52:55] <thakis> or use groovy or something like that
[21:53:16] <oshima> pinkerton: i've seen it's timing out recently when i'm sheriffing, so it's not your fault. We need to fix it though.
[21:53:24] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed - throttling a little longer, Linux Mem going green, ask on IRC
[21:53:25] <mirandac> mmoss:  may I http://codereview.chromium.org/2934011/show ?
[21:53:26] <trungl> "without -O, it'll compile a bit faster and the disassembly will be a bit more straightforward ... but we'll also skip giving you a pile of warnings"
[21:53:26] <pinkerton> k, just checking before i land :)
[21:53:43] <trungl> thakis: pshaw, compilers are for wusses; I'd be happy with just a macro assembler
[21:54:09] *** jshin has joined #chromium
[21:54:09] *** ChanServ sets mode: +v jshin
[21:54:23] <shess> I think no -O is reasonable for developers, but I'm not sure how much value it adds to trybots.  I mean, who gdb's anything off the trybots?
[21:54:32] <shess> [unless they're debugging the trybots.]
[21:54:32] <pinkerton> mmoss: landed
[21:54:43] <mmoss> pkasting: thx
[21:54:48] <mmoss> mirandac: can it wait for win trybot to finish?
[21:54:58] <mirandac> mmoss: sure thing.
[21:54:59] <shess> and it would make the trybots X% faster to run the tests :-).
[21:55:26] <awong> thakis: hmmm...maybe I'll try a clean build...  It's still showing up for me. :(
[21:55:45] *** Gina1 has joined #chromium
[21:55:54] <trungl> shess: it's very bad to not get warnings/errors on the trybots that you'd get in the release build
[21:55:59] *** Sarten-X has quit IRC
[21:56:22] *** Gina1 has quit IRC
[21:56:41] *** sanjeevr has joined #chromium
[21:56:45] <oshima> jrg: should i file a bug for this crash?
[21:57:10] <jrg> oshima: try this.  disable test completely, file bug, assign to me, and I'll look into it.
[21:57:10] *** janm has quit IRC
[21:57:23] <jrg> oshima: it's not my area but I feel Mao-tastic today.
[21:57:49] <oshima> jrg: will do. thanks.
[21:57:49] *** Gina has quit IRC
[21:58:39] *** janm has joined #chromium
[21:59:42] <markmentovai> shess, trungl: debug and release are complementary, as we use them. neither will ever fully catch everything that the other does.
[21:59:53] <jrg> mmoss: can I land mac-only CL?  Is small.  mac trybot green.    http://codereview.chromium.org/2958016
[22:00:54] <shess> markmentovai: not suggesting to remove debug.  Trung had a CL bounce because of a trivial uninitialized variable which is not warned if not -O1 or greater.  I was positing that trybots could do debug builds with -O1 without losing much.
[22:01:10] <markmentovai> lose a little speed
[22:01:14] <markmentovai> :)
[22:01:19] <trungl> well, presumably it was pink's CL
[22:01:25] <shess> markmentovai: on the build - but would gain on running the tests.
[22:01:53] <trungl> markmentovai: I'm just worried that without -O, a nontrivial class of warnings aren't generated
[22:02:01] <trungl> (or is that detected?)
[22:02:05] <trungl> noticed?
[22:02:33] <markmentovai> maybe the green tree people have some sense of how often this is responsible for tree closures. nsylvain
[22:03:01] <jrg> who is throttling the the tree now?  dmaclach mmoss tyoshino
[22:03:18] <pinkerton> trungl: but you reviewed my crappy code, so i can shift all blame to you
[22:03:31] <dmaclach> jrg go ahead
[22:03:35] <jrg> dmaclach: thx
[22:03:58] <jrg> dmaclach: done r52515
[22:04:39] * trungl 's broad shoulders accepts all blame.
[22:05:23] <pinkerton> mrossetti: ping?
[22:05:34] <mrossetti> pinketron: pong
[22:05:48] <pinkerton> did you see my email yesterday about the thumbnail store?
[22:05:48] <mrossetti> heh, 'pinketron' --> pinkerton
[22:06:02] <mrossetti> Yes.
[22:06:18] <mrossetti> Don't know anything about thumbnail store.
[22:06:23] <mrossetti> Only know about history.
[22:06:28] <mrossetti> Despite the rumors.
[22:06:30] <pinkerton> didn't you write code to not back it up with time machine?
[22:06:45] <pinkerton> or are we backing it up still?
[22:07:00] <trungl> mrossetti: I like the new moniker
[22:07:10] *** sbyer has quit IRC
[22:07:22] <rsesek> pinketron is the pink version of trung-bot
[22:07:41] <pinkerton> there used to be a pink-bot, with video and a pink tshirt.
[22:07:46] *** sanjeevr has quit IRC
[22:07:52] <pinkerton> it was all the rage in mtnview
[22:08:18] <mrossetti> pinkerton: I did exclude some files from TM but didn't recall if thumbnails was part of it.  Lemme go take a look at the code and pay the proper attention to your email.  Eh.
[22:08:25] <pinkerton> :)
[22:08:28] <pinkerton> sorry :)
[22:08:37] <rohitrao> pinkerton, stuartmorgan: I think I have an easy fix for the findbar, but not for fullscreen
[22:08:44] <pinkerton> yay?
[22:08:53] *** chrisccoulson__ has joined #chromium
[22:08:59] <rohitrao> I don't know if that's enough to leave flash on
[22:09:01] <mattm_g> dmaclach: mmoss:  I have four small changes to land (all green) http://codereview.chromium.org/3015001/show http://codereview.chromium.org/2802035/show http://codereview.chromium.org/2881014/show http://codereview.chromium.org/2891023/show
[22:09:04] <rohitrao> I bet people will complain about fullscreen :)
[22:09:23] <pinkerton> rohitrao: maybe look at the stats to see how many people use it?
[22:09:33] <pinkerton> everyone uses flash and it would be great to get the acceleration
[22:09:34] <akalin> sherriffs: can I check in http://codereview.chromium.org/2928012/show ?  local build on windows works, trybots pass for other platforms, the change for google-cache-invalidation-api is tiny
[22:09:35] *** sbyer has joined #chromium
[22:09:35] *** ChanServ sets mode: +v sbyer
[22:09:43] * dmaclach got rid of the pink-bot's pink google tshirt
[22:09:56] <rohitrao> ooh, usermetrics
[22:09:56] <pinkerton> :( i rather liked pinkbot
[22:10:28] <akalin> what is pinkbot
[22:10:30] *** sanjeevr has joined #chromium
[22:10:44] <trungl> we should get pinkerton one of those new telepresence robots
[22:10:52] *** Dykam has quit IRC
[22:11:03] <mmoss> mattm_g: ok
[22:11:26] *** Dykam has joined #chromium
[22:11:33] *** jcastro has quit IRC
[22:11:53] <pinkerton> akalin: back when we were working on google desktop, the mtnview folks set up an old imac in their office that i could VC into from home and we just left it connected all day long so we could talk any time
[22:12:07] <dmaclach> would someone be kind enough to mark DispatchSimple as flaky?
[22:12:09] *** chrisccoulson has quit IRC
[22:12:13] <sanjeevr> mmoss, dmaclach: Lost IRC connection, trying again. I have a Pepper v2 change in webkit/glue. Trybots green. http://codereview.chromium.org/2819047/show. OK to land?
[22:12:22] <akalin> i see
[22:12:25] *** jcastro has joined #chromium
[22:12:32] *** lalalalilili has joined #chromium
[22:13:09] <dmaclach> akalin, I think we still have the pinkponyomg aim account that we used for it ;-)
[22:13:36] <pinkerton> :D
[22:13:46] <mmoss> sanjeevr: ok
[22:13:54] <sanjeevr> mmoss: Thx
[22:14:04] <akalin> mmoss, dmaclach: ^^
[22:14:12] <akalin> you may have not seen my previous message
[22:14:27] <akalin> re. http://codereview.chromium.org/2928012/show
[22:14:37] *** Erkan_Yilmaz_ has joined #chromium
[22:15:00] <dmaclach> mmoss your call re: akalin's cl
[22:15:01] *** eseidel has quit IRC
[22:15:05] <mmoss> akalin: ok
[22:15:27] *** sbyer has quit IRC
[22:15:36] <akalin> mmoss: yay, thanks
[22:15:59] *** hrna has quit IRC
[22:16:00] *** Erkan_Yilmaz__ has joined #chromium
[22:16:13] *** dumi has quit IRC
[22:17:31] <ojan> anyone around who understands our SVG implementation?
[22:17:47] <ojan> or SVG at all for that matter
[22:17:48] *** fta has joined #chromium
[22:17:54] *** Erkan_Yilmaz has quit IRC
[22:17:55] <pinkerton> rsesek: did you track down the regression window for flash?
[22:18:00] <willchan> no one understands the webkit SVG implementation
[22:18:07] <rsesek> pinkerton: ja someone's on it already
[22:18:14] <pinkerton> rsesek: hotness
[22:18:48] <dmaclach> ojan, once upon a time I did, but it's been a couple of years now.
[22:19:26] *** Erkan_Yilmaz_ has quit IRC
[22:19:27] <ananta> dmaclach: Can I land this ChromeFrame CL? http://codereview.chromium.org/3014001/show
[22:20:12] <rsesek> pinkerton: crbug.com/49158
[22:21:02] <pinkerton> more hotness
[22:21:12] *** sbyer has joined #chromium
[22:21:12] *** ChanServ sets mode: +v sbyer
[22:21:32] <sanjeevr> mmoss: Committed as 52520
[22:21:49] <dmaclach> ananta: can you wait a few minutes? That should clean up some red in the tree
[22:22:08] <mmoss> ananta: and give the win trybot a bit more time
[22:23:19] *** willchan has quit IRC
[22:23:39] *** eseidel has joined #chromium
[22:24:04] <alyssad> mmoss: requesting to land (whenever possible) http://codereview.chromium.org/2903008/show
[22:24:34] *** Kunalagon has joined #chromium
[22:25:42] *** tonyg-cr has joined #chromium
[22:25:42] *** ChanServ sets mode: +v tonyg-cr
[22:25:54] *** andrix has quit IRC
[22:27:10] <ananta> mmoss: The previous incarnation of the win try bot passed http://build.chromium.org/buildbot/try-server/builders/win/builds/40207
[22:27:43] <ananta> mmoss: The current run is just a chrome frame test html file added to the cl. This does not run in the win try runs anyway
[22:27:52] *** chrisccoulson__ is now known as chrisccoulson
[22:28:10] <ananta> dmaclach: ok. let me know when it is ok to commit then
[22:29:03] *** Dykam has quit IRC
[22:29:08] <awong> gyp question: in common.gypi, there is a 'cflags! = ["-fvisibility=hidden"]' for shread libraries.  Is there a way to override this and readd -fvisibility=hidden in a lower target? The exclusion seems to take precedence over everything else.
[22:29:46] <markmentovai> you can try 'cflags/': [['include', '^-fvisibility=hidden$']]
[22:31:00] <dmaclach> ananta/alyssad: go ahead
[22:31:00] *** sbyer has quit IRC
[22:31:06] *** mnissler_ has joined #chromium
[22:32:02] <markmentovai> sheriffs: i would like to break your tree in the most disruptive way possible. what?s the wait on something like that?
[22:32:13] <alyssad> thanks!
[22:32:25] <markmentovai> if needed, i?m available to help make otherwise-functioning bots turn purple.
[22:33:05] <ananta> dmaclach: thanks
[22:33:28] <awong> markmentovai: whoa...that crazy gyp syntax worked.  Thanks!
[22:33:35] <markmentovai> awong: no prob
[22:36:33] *** zaspire has quit IRC
[22:37:46] <dmaclach> markmentovai: what do you want do to?
[22:38:03] <rsesek> what's the best way to get the Browser* that a TabContents* is a part of?
[22:38:13] <markmentovai> dmaclach: http://codereview.chromium.org/3032001 - but if the plan is to reopen the tree at some point, i can just wait like a civilized mofo.
[22:38:23] <mmoss> maybe he can take down an upstream source control system
[22:38:37] <markmentovai> i can do that too. i'm flexible.
[22:38:51] *** monreal has quit IRC
[22:39:03] *** novas0x2a has quit IRC
[22:39:33] *** pinkerton has left #chromium
[22:39:45] *** phajdan-jr is now known as phajdan-jr|afk
[22:39:45] *** jcastro has quit IRC
[22:40:12] *** jcastro has joined #chromium
[22:40:35] *** pinkerton has joined #chromium
[22:40:35] *** ChanServ sets mode: +v pinkerton
[22:40:52] *** roc has quit IRC
[22:40:55] <mirandac> sheriffs: http://codereview.chromium.org/2934011/show green on all bots, may I proceed?
[22:43:00] <dmaclach> markmentovai: go ahead... break away
[22:44:36] <dmaclach> mirandac: can you wait a minute or two? That should clear up the bits of breakage on windows side. Mento got to go ahead because he most likely will only break the Mac ;-)
[22:44:41] <markmentovai> dmaclach: thanks. the next thing i want to do is to back out all changes ever committed on the weekend, because people shouldn't be working on the weekend. there will be merge conflicts, but i intend to simply leave the conflict markers in place and check in anyway. can i land this now?
[22:45:16] * dmaclach notes that it isn't after 5 even in NY yet
[22:45:24] <markmentovai> it's thursday though
[22:45:35] <mirandac> dmaclach: sure thing.  mento's changes sound violent and extreme, I don't want to get in his way ;-)
[22:48:55] *** mirandac is now known as mirandac_walking
[22:49:05] *** mirandac_walking is now known as miranda_afkwdog
[22:49:29] *** tommi has quit IRC
[22:53:55] *** jcivelli has joined #chromium
[22:53:55] *** ChanServ sets mode: +v jcivelli
[22:54:24] *** tommi has joined #chromium
[22:55:33] <mmoss> miranda_afkwdog: good to go when you're back
[22:56:53] *** pathorn has joined #chromium
[22:56:53] *** ChanServ sets mode: +v pathorn
[22:57:59] *** Erkan_Yilmaz_ has joined #chromium
[22:58:07] <mmoss> be gentle people
[22:58:17] <trungl-bot> Tree opened by mmoss at chromium dot org: Tree is open
[22:58:34] <akalin> MOAR CHANGELISTS
[22:59:16] <thakis> :-D
[22:59:21] *** roc has joined #chromium
[22:59:38] <akalin> i want to check in a change that switches the sync protos from proto_lite to proto
[22:59:50] <akalin> it seemed to break on the windows trybot, but local clean build seems fine
[22:59:59] <akalin> so if i check this in, it may require a clobber of the windows bots
[23:00:12] <akalin> mmoss, dmaclach: http://codereview.chromium.org/2951016/show
[23:00:15] <akalin> how should I proceed?
[23:00:38] *** Kunalagon has quit IRC
[23:01:13] *** Erkan_Yilmaz__ has quit IRC
[23:01:22] <pinkerton> are we going to mark DispatchSimple as flaky?
[23:01:27] <pinkerton> it seems to be horking up a lot
[23:01:36] <dmaclach> pinkerton: I was just working on it
[23:01:38] <pinkerton> k
[23:02:33] <markmentovai> akalin: if the tools have disrupted you, DON'T check in. instead, continue to be disrupted. exception: if you think you can cause greater disruption to the team, DO check in. but be sure you have disruption readability, or approval of someone else who does.
[23:02:38] *** sbyer has joined #chromium
[23:02:38] *** ChanServ sets mode: +v sbyer
[23:02:43] <dmaclach> akalin: I assume the break on the trybot was test related?
[23:02:50] <akalin> dmaclach: it was some sort of link error
[23:03:08] <akalin> i tried sending another (clobbered) trybot job, but that may take a while
[23:03:10] <akalin> i can wait, i guess
[23:03:21] <mmoss> akalin: can you send it to a windows trybot with a clobber command, and see if that indeed works?
[23:03:25] <akalin> markmentovai: i can't tell if you're joking o_O
[23:03:29] <pinkerton> hrm, i'm now in a state where i have to cmd-w twice to close a tab
[23:03:31] <dmaclach> if that's ok... we'd love to have the tree officially open for a few minutes :)
[23:03:33] <akalin> mmoss: okay, i did.  i'll wait.
[23:03:33] <mmoss> I'd hate to clobber if not necessary
[23:03:36] <markmentovai> pinkerton: i've seen that
[23:03:58] *** MichealH has quit IRC
[23:04:00] <rohitrao> get symbols and debug it!
[23:04:19] <thakis> markmentovai: tell a joke, so that akalin has a reference point
[23:04:41] *** abarth has quit IRC
[23:04:42] <thakis> we won't even make fun of you if the joke's about ld
[23:05:00] <markmentovai> akalin: example: my explanation was disruptive, so i should be permitted to check it in.
[23:05:38] <markmentovai> thakis: i can't really do the toothpaste shtick in this #channel.
[23:07:24] *** ROBOd has quit IRC
[23:07:26] <pinkerton> hrm, first cmd-w made the renderer go away but the second one was needed to close the tab
[23:07:36] <pinkerton> wonder if we have a retain cycle somewhere
[23:08:13] *** sbyer has quit IRC
[23:08:40] <markmentovai> pinkerton: once you?re in this state, you?ll probably stay there? hook up a debugger?
[23:08:49] *** abarth has joined #chromium
[23:08:49] *** ChanServ sets mode: +v abarth
[23:10:43] <thakis> i love it when "confused by earlier errors, bailing out" is the only error gcc is reporting
[23:10:48] *** sbyer has joined #chromium
[23:10:48] *** ChanServ sets mode: +v sbyer
[23:10:48] <thakis> easily confused
[23:11:04] <pinkerton> markmentovai: nightly release build
[23:11:18] <pinkerton> no symbols, no way to fake it, right?
[23:11:19] <markmentovai> gcc's objc parser sucks. there's that joke you were asking for.
[23:11:23] <pinkerton> we've been down this path before :)
[23:11:26] <markmentovai> pinkerton: nightly? no
[23:11:32] <markmentovai> don't run those?
[23:11:46] <pinkerton> i think it might have to do with dragging tabs out into another window
[23:11:46] <markmentovai> dev channel, we've got more symbols for those than we know what to do with.
[23:11:55] <pinkerton> maybe
[23:12:36] <rohitrao> pinkerton: the first cmd-w gives you a sad tab?
[23:13:34] *** MikeSmithW3C has joined #chromium
[23:13:47] <pinkerton> rohitrao: no, the content area is white
[23:13:59] *** agl has quit IRC
[23:14:41] <rohitrao> huh, I wonder what's happening there
[23:14:48] <pinkerton> ditto
[23:14:58] *** pfeldman has joined #chromium
[23:15:10] <rohitrao> I don't think we ever remove the web view, we always replace it
[23:15:17] <rohitrao> so white seems odd
[23:15:46] <rohitrao> can you tell if the render processes are actually dying, or if they're just painting white?
[23:15:46] *** til has quit IRC
[23:15:49] <pinkerton> didn't do it that time
[23:15:58] <pinkerton> the tab was still alive after the first cmd-w
[23:16:02] *** til has joined #chromium
[23:16:06] <pinkerton> grr
[23:16:28] *** MikeSmithW3C^ has quit IRC
[23:16:50] *** devinus has joined #chromium
[23:16:56] <jcivelli> After Windows last update 'git cl dcommit' fails with the infamous "fatal error: unable to remap"
[23:17:03] <pinkerton> dmaclach: i'm not used to looking the bug field for a *new* bug :)
[23:17:10] <jcivelli> I did the rebaseall magic and it wouldn't help
[23:17:19] <devinus> why does chromium silently fail when you try to do an xhr with no Access-Control-Allow-Origin header?
[23:17:23] <sky_> jcivelli: doing a git pull then rebase seemed to fix it for me.
[23:17:32] <sky_> but it's very finicky, I keep getting it too.
[23:17:44] <sky_> although wait, I haven't rebooted yet... On no.
[23:17:48] <jcivelli> sky_: thanks, I'll try it
[23:18:40] <pinkerton> hm, it's not just cmd-w, it's also clicking the close box
[23:19:16] <rohitrao> pinkerton: I think that makes sense? They all get funneled through Browser
[23:19:34] <pinkerton> yeah, that makes me feel a little better about the cmd-w handling
[23:19:40] <pinkerton> but only just a little
[23:20:17] *** loislo has quit IRC
[23:21:22] <pinkerton> ok i'm out, l8r all
[23:21:25] *** pinkerton has quit IRC
[23:21:28] *** jcivelli has quit IRC
[23:21:57] <victorw> The Windows chromium webkit build fails with this error afterI roll chromium DEPS from 51736 to 52483, anyone why know what's the issue? gfx.lib(rect.obj) : error LNK2005: "public: void __thiscall gfx::Rect::set_width(int)" (?set_width@Rect@gfx@@QAEXH@Z) already defined in glue.lib(webplugin_delegate_impl_win.obj) 39>gfx.lib\WebKitBuildSlave\chromium-win-release\build\WebKit\chromium\Release\DumpRenderTree.exe : fatal error LNK1169: one or
[23:22:10] <victorw> http://build.webkit.org/builders/Chromium%20Win%20Release/builds/9028/steps/compile-webkit/logs/stdio
[23:25:11] *** mnissler_ has quit IRC
[23:26:48] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Mac10.6 Tests" from 52526: csilv at chromium dot org, johnnyg at chromium dot org (:johnny_g), mark at chromium dot org (:markmentovai))
[23:26:52] <markmentovai> NOT IT
[23:27:07] *** tedoc2000 has quit IRC
[23:28:37] <johnny_g> ditto :)
[23:28:56] <csilv> looking...
[23:29:17] *** tedoc2000 has joined #chromium
[23:29:51] * markmentovai suspects flake.
[23:30:35] <markmentovai> 10.5 rel and dbg was ok; 10.6 dbg was ok
[23:30:50] <csilv> i suspect flake as well
[23:31:01] <markmentovai> then it's settled
[23:31:45] <csilv> yeah, DispatchSimpler which was subsequently marked as FLAKEY.
[23:31:50] <trungl-bot> Tree opened by mark at chromium dot org (:markmentovai): Open
[23:32:26] *** pfeldman_ has joined #chromium
[23:32:35] <eglaysher> chromeos failure something to do with Notifications?
[23:32:47] <eglaysher> johnny_g: ^
[23:33:08] <johnny_g> eglaysher: checking
[23:33:38] <johnny_g> eglaysher: do you have a link?
[23:34:05] <mmoss> johnny_g: http://buildbot.jail.google.com/buildbot/chromium/builders/Linux%20Builder%20(ChromiumOS)/builds/9112/steps/compile/logs/stdio
[23:34:14] <mmoss> chrome/browser/chromeos/notifications/desktop_notifications_unittest.cc: In member function 'int chromeos::MockBalloonCollection::UppermostVerticalPosition()':
[23:34:21] <mmoss> chrome/browser/chromeos/notifications/desktop_notifications_unittest.cc:68: error: 'class Balloon' has no member named 'position'
[23:34:29] <johnny_g> oh, that sounds likely
[23:36:03] *** pfeldman has quit IRC
[23:36:03] *** pfeldman_ is now known as pfeldman
[23:36:55] <johnny_g> fixing
[23:38:25] <akalin> tree is grey as butt
[23:39:10] *** Erkan_Yilmaz__ has joined #chromium
[23:40:08] <mmoss> akalin: I don't want to know the basis of that expression, but I agree
[23:40:33] <mmoss> I'm trying to leave it open since it was closed so long
[23:40:46] <dmaclach> mmoss: want to switch back to throttle mode, or should we just respectfully ask our submitters to be kind and see if that sticks? ;-)
[23:40:49] <mmoss> if all the running tests scroll off my screen, I'll close
[23:41:08] <mmoss> I have no faith in our submitters
[23:41:12] <akalin> mmoss: i think we're in some sort of death spiral :(
[23:41:17] <mmoss> look where asking them to be gentle got us
[23:41:21] <dmaclach> mmoss: how big is your monitor? I'm looking at it on my laptop, and we're pretty darn close
[23:42:14] *** Erkan_Yilmaz_ has quit IRC
[23:42:31] <mmoss> dmaclach: 3 more commits
[23:42:34] <mmoss> maybe sooner
[23:42:42] <ananta> dmaclach: The ChromeFrame builders should turn green in the next couple of runs. Please ignore the redness for now
[23:42:52] <dmaclach> ananta: thanks
[23:43:02] *** Gina has joined #chromium
[23:43:36] *** jcivelli has joined #chromium
[23:43:36] *** ChanServ sets mode: +v jcivelli
[23:46:43] <tfarina> whoray, finally the tree is open!!
[23:47:21] <johnny_g> chromeos compiles again.  sorry about that.
[23:47:31] <mmoss> tfarina: not for long
[23:47:52] <tfarina> mmoss: why not?
[23:48:18] <eglaysher> win bots are 15 patches behind, I'd guess
[23:48:27] <mmoss> is grey your favorite color?
[23:48:30] *** dumi has joined #chromium
[23:48:30] *** ChanServ sets mode: +v dumi
[23:50:04] <dumi> anybody an expert on running layout tests as ui_tests?
[23:50:18] <trungl-bot> Tree closed by mmoss at chromium dot org: Tree is closed. Waiting to catch up.
[23:51:07] *** Gina has quit IRC
[23:51:21] <jcivelli> My build fails linking on Mac (libwecore.a can't map file), I know there were some emails sent about this, but can't find what the resolution was.
[23:51:58] *** [Nef] has quit IRC
[23:52:07] <awong> jcivelli: stop using 10.5?
[23:52:14] <akalin> try turning off svg
[23:52:34] *** viro101 has joined #chromium
[23:53:03] <dhollowa> GYP_DEFINES='enable_svg=0' gclient sync; GYP_DEFINES='enable_svg=0' gclient runhooks
[23:53:17] <jcivelli> I'll give it a try, thanks!
[23:53:59] <tony^work> estade: I will land the inspector resources fix
[23:54:07] <estade> thx
[23:54:08] <tony^work> sorry, tkent posted a correct version later
[23:54:16] <tony^work> I will land his patch
[23:54:51] <awong> markmentovai: does my latest revision to the ppapi change pass the grammar check? :P http://codereview.chromium.org/3018003/show
[23:55:03] *** markmentovai is now known as mcHammer
[23:55:30] <mcHammer> nickserv isn't gonna let me get away with this, but lgtm
[23:55:38] *** mcHammer is now known as markmentovai
[23:56:04] *** hicham has joined #chromium
[23:56:07] <awong> thankyou!
[23:56:20] *** hicham has left #chromium
[23:57:25] *** knighthawk has joined #chromium
[23:58:39] *** Viclame has joined #chromium
[23:59:29] *** tfarina has quit IRC
[23:59:45] *** Gina has joined #chromium





top