[00:02:04] <atwilson> Hi all - I'm banging my head on a bug and figured I'd ask here to see if people have any ideas.
[00:02:26] *** happygrue_ has quit IRC
[00:02:59] <trungl-bot> Tree closed by vandebo at chromium dot org: Tree is closed (get permission for safer commits)
[00:03:26] <atwilson> I'm trying to make the windows version of Chrome outlive the last browser window. It's coming along fine, except when I close the last browser window, any desktop notifications are also closed. They close, because the IPC channel to the Renderer processes gets closed, and I haven't yet tracked down how this happens. Does this sound familiar to anyone (is there anything that somehow detects the last browser window closing and closes IPC connections)?
[00:03:36] <atwilson> It doesn't happen on the Mac, of course.
[00:03:37] *** fta has joined #chromium
[00:04:02] <vandebo> I don't understand why the testers aren't starting... so safer commits are welcome (closed only so that people don't arbitrary commit)
[00:04:11] <eroman> vandebo: can i do a commit?
[00:04:18] <jam2> vandebo: i have something only for pepper
[00:04:23] <jam2> green on all buildbots, and needed by vtl
[00:04:29] <vandebo> jam2: sounds good
[00:04:33] <jam2> thanks
[00:04:34] <vandebo> eroman: is it safe?
[00:04:53] <eroman> vandebo: it passes all the trybots... "should" be
[00:05:24] <vandebo> well, I mean how invasive is it... we need a new lkgr
[00:05:53] <eroman> vandebo: its cool, i can wait
[00:08:16] <kinuko> they started the builds
[00:08:30] <vandebo> like 20min later...
[00:09:12] *** fishd has joined #chromium
[00:09:12] *** ChanServ sets mode: +v fishd
[00:11:03] <kinuko> sigh extract build failure
[00:11:31] <vandebo> yea, I'm going to force that one
[00:11:51] <vandebo> maybe get an lkgr at 51426
[00:12:32] <thakis> vandebo: ok to submit http://codereview.chromium.org/2815042/show ?
[00:12:59] <vandebo> sure
[00:13:16] <feldstein> vandebo: ok to submit?  two line cocoa change http://codereview.chromium.org/2832045/show
[00:13:45] <vandebo> go ahead
[00:13:48] <feldstein> thx
[00:13:48] <kinuko> trungl-bot: lkgr
[00:13:49] <trungl-bot> kinuko: 51404
[00:14:13] <vandebo> ohh, didn't notice that update.  In better shape than I thought
[00:14:37] <kinuko> vandebo: :)
[00:15:03] *** kinuko has left #chromium
[00:15:06] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[00:15:12] *** kinuko has joined #chromium
[00:15:13] *** ChanServ sets mode: +v kinuko
[00:22:04] *** vithos has joined #chromium
[00:23:08] <dcheng> stuartmorgan: Feel like making arguments for/against https://bugs.webkit.org/show_bug.cgi?id=41476 ?
[00:23:10] <dcheng> Or anyone else really.
[00:23:11] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open (valgrind bots -> stuartmorgan)
[00:25:36] <stuartmorgan> dcheng: hyatt is probably right. When I filed the bug I hadn't yet discovered the distinction in where the drag started
[00:25:53] <stuartmorgan> dcheng: so I thought it was just a whitespace-stripping bug in the tab strip
[00:25:56] *** urbanape has quit IRC
[00:26:16] <dave_levin> stuartmorgan: suppresion text issuesee http://build.chromium.org/buildbot/waterfall.fyi/builders/Webkit%20Linux%20(valgrind%20webkit.org)/builds/7853/steps/memory%20test:%20test_shell/logs/stdio
[00:26:18] <thakis> git: 'pulll' is not a git command. See 'git --help'.
[00:26:26] <stuartmorgan> dave_levin: yeah, I know
[00:26:42] <stuartmorgan> dave_levin: I'm adding another suppression anyway, so I'm fixing it with that; cl shortly
[00:27:00] <stuartmorgan> dave_levin: Then I'll be learning to write a presubmit check :P
[00:27:15] *** eseidel has joined #chromium
[00:28:08] <dcheng> stuartmorgan: At the same time, it seems bizarre to exhibit two different behaviors for dragging the exact same bit of data.
[00:28:25] <dcheng> stuartmorgan: That being said, if you tend to agree with hyatt, I'm just going to close the Chromium issue as won't fix/by design.
[00:28:59] <stuartmorgan> dcheng: well, it would be nice if it worked
[00:29:50] <stuartmorgan> But not so much that I see having a protracted argument about whether it's better to handle broken data as worth my time
[00:30:01] *** eseidel has quit IRC
[00:30:02] <dcheng> stuartmorgan: Can you propose an alternate/safer algorithm to fix broken links?
[00:30:08] *** urbanape has joined #chromium
[00:31:27] <akalin> hey guys
[00:31:57] <akalin> anyone have a rough idea of how many people run chromium nightlies/continuous builds?
[00:32:27] <stuartmorgan> dcheng: nobody has objected to the algorithm have they?
[00:32:46] <stuartmorgan> dcheng: his argument is "the source is broken, not our problem"
[00:33:09] <dcheng> The algorithm is fairly dumb and can result in false positives.
[00:33:22] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Mac10.6 Tests (dbg)(1)" from 51433: feldstein at chromium dot org, thakis at chromium dot org)
[00:33:36] <thakis> gah!
[00:34:21] <feldstein> thakis: was that you?
[00:34:25] <vandebo> content settings
[00:34:26] <thakis> dunno
[00:34:28] <thakis> looking
[00:34:32] <thakis> i touched content settings
[00:34:47] <vandebo> ContentBlockedBubbleControllerTest.Init content_blocked_bubble_controller.mm(126)] Check failed: settingsType != CONTENT_SETTINGS_TYPE_NOTIFICATIONS (6 vs. 6)
[00:34:53] <thakis> vandebo: looking
[00:35:16] <stuartmorgan> dcheng: but that's not the objection
[00:35:32] <stuartmorgan> dcheng: as long as the objection is that it's not a bug, the algorithm isn't going to matter
[00:35:41] *** fta has quit IRC
[00:36:24] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open ("Mac10.6 Tests -> thakis at chromium dot org)
[00:38:49] <vandebo> thakis: revert ?
[00:38:56] <thakis> vandebo: i have a fix ready
[00:38:58] <dcheng> stuartmorgan: I don't really care strongly enough one way or another. If you have more compelling arguments than me, feel free to add them to the WebKit bug =)
[00:38:59] <vandebo> ok
[00:39:00] <thakis> vandebo: http://codereview.chromium.org/2805061/show
[00:39:19] <stuartmorgan> dcheng: I'll try one comment I guess
[00:39:27] <vandebo> thakis: if you say so - TBR it
[00:39:55] *** victorw has joined #chromium
[00:39:55] *** ChanServ sets mode: +v victorw
[00:41:54] <thakis> vandebo: done
[00:44:22] *** mrossetti has quit IRC
[00:44:22] *** chaser_ has quit IRC
[00:44:32] *** cbentzel has joined #chromium
[00:44:49] *** chaser_ has joined #chromium
[00:46:28] *** mrossetti_ has joined #chromium
[00:46:29] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Mac10.5 Tests" from 51437: ananta at chromium dot org (:iyengar), chron at chromium dot org, dtseng at chromium dot org, eroman at chromium dot org (:eroman), feldstein at chromium dot org, jam at chromium dot org (:jam2), jhawkins at chromium dot org, stuartmorgan at chromium dot org, thakis at chromium dot org)
[00:47:02] *** chronarion has joined #chromium
[00:47:04] *** ChanServ sets mode: +v chronarion
[00:47:08] *** urbanape has quit IRC
[00:47:21] *** MikeSmithW3C^ has quit IRC
[00:47:29] <trungl-bot> Tree opened by thakis at chromium dot org: Tree is open ("Mac10.[56] Tests -> thakis at chromium dot org)
[00:47:51] <rsesek> eglaysher_lin: ping
[00:47:55] <vandebo> er, yea, that's what I meant
[00:48:10] *** aroben has quit IRC
[00:48:21] <thakis> "/usr/bin/pump: line 499: 66983 Bus error PATH="$distcc_location:$PATH" "$@"" is probably a flake?
[00:49:34] *** fta has joined #chromium
[00:51:56] <thakis> yes
[00:51:59] *** MikeSmithW3C^ has joined #chromium
[00:53:40] *** Ruetobas has quit IRC
[00:54:27] <atwilson> estade: git blame seems to think you're a good person to ask about browser_shutdown::OnShutdownStarting() - is that true?
[00:54:46] <estade> atwilson: yes
[00:54:50] * cyberix coded a websocket console
[00:54:51] <cyberix> http://www.cs.helsinki.fi/u/twruottu/webcat.html
[00:55:48] <atwilson> estade: I'm changing the chrome shutdown code to try to make the behavior on different platforms be more consistent (which also allows me to keep chrome running after the last window is closed in Win/Linux).
[00:55:51] <Hixie> cyberix: yours is prettier than mine :-) http://software.hixie.ch/utilities/js/websocket/
[00:55:53] <thakis> cyberix: cool :-)
[00:56:13] *** Ruetobas has joined #chromium
[00:56:40] <cyberix> Hixie: Oh. I need to support different protocols. :-D
[00:56:53] <estade> atwilson: ok
[00:56:54] <eglaysher_lin> rellig: pong
[00:57:31] <atwilson> estade: OnShutdownStarting() is currently called from Browser::OnWindowClosing(), and one side-effect is it triggers a fast shutdown if this is the last window on win/linux.
[00:58:06] <atwilson> estade: I'd like to move that code into BrowserList instead.
[00:58:25] <atwilson> But this would mean that OnShutdownStarting() would not get invoked until the Browser destructor (not OnWindowClosing).
[00:58:54] <atwilson> Do you know whether that's a bad idea or not? I don't know how much additional latency (if any) that would incur.
[00:59:48] <estade> it would probably essentially defeat the purpose of the fast shutdown
[00:59:48] <atwilson> estade: the rationale is that I don't really know (without doing something ugly like peeking at the BrowserProcess ref count) whether closing the last window is going to cause the app to shutdown.
[00:59:56] <eglaysher_lin> rsesek: pong
[01:00:35] <rsesek> eglaysher_lin: is your todo in button_menu_item_model.h still relevant (re: groups?) I see an AddItemWithStringIdAndGroup() so I'm not sure about the baked-ness of the feature
[01:00:45] <estade> atwilson: hmm. it seems that you will still get fast shutdown via CloseAllTabs
[01:00:47] *** Venom_X has quit IRC
[01:00:47] <atwilson> estade: OK, I wasn't certain how long it takes for the Browser destructor to get called, but it sounds like it's delayed until the renderer processes all exit?
[01:01:05] <eglaysher_lin> rsesek: it's not complete yet. I need to hook it up in the GTK implementation (still don't really have an idea of how I'm going to do that)
[01:01:11] *** Venom_X has joined #chromium
[01:01:26] <eglaysher_lin> The users of the model don't do it either.
[01:01:42] <eglaysher_lin> But maybe the code is complete. It looks like it.
[01:01:46] <estade> atwilson: yea, the renderer processes would have to exit (or at least sometimes, depending on their unload handlers)
[01:01:47] <rsesek> eglaysher_lin: okay. I'm doing this for Mac. I think i may not use the XP-model and just override the display with the 2 special cases and handle everything myself
[01:01:58] <atwilson> estade: OK, if that's true, then what's the point of the code in OnShutdownStarting() calling FastShutdownIfPossible() on everything?
[01:02:00] <dcheng> Anyone here mind testing something on Win dev channel?
[01:02:00] <rsesek> because the layout is so specific
[01:02:02] <estade> but since closealltabs already does fast shutdown, it's probably ok to do what you said
[01:02:13] <estade> atwilson: I don't know. I was the last person to touch this code but not the original other
[01:02:26] <estade> probably it used to be necessary but lost its purpose
[01:02:34] <atwilson> estade: Do you know who else might know about this?
[01:02:39] <estade> ojan
[01:02:57] <atwilson> estade: tx
[01:03:53] *** Mauro__ has joined #chromium
[01:04:51] *** shepazu has joined #chromium
[01:04:59] *** super__ has quit IRC
[01:05:13] *** Mauro__ has left #chromium
[01:05:40] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "ipc_tests" on "Mac10.5 Tests (dbg)(1)" from 51444: arv at chromium dot org, thakis at chromium dot org)
[01:06:38] <vandebo> thakis: you lose
[01:06:55] <thakis> vandebo: ipc_tests
[01:06:58] <thakis> vandebo: different test
[01:07:54] <thakis> vandebo: only on 10.5 too
[01:07:56] *** MichealH has quit IRC
[01:08:21] <vandebo> sigh, but it could be anywhere from 51433
[01:08:39] <dcheng> If anyone on Win dev could try going to http://www.becuhomeloans.org/wizard/LoanConsultant.asp, could they let me know if they see "refinance" show up in the "purpose of the loan" drop down?
[01:08:49] <dcheng> For me, it appears everywhere except on my Win box which seems bizarre.
[01:09:14] <vandebo> no, ipc tests passed before
[01:11:15] *** tonikitoo has quit IRC
[01:11:51] <vandebo> arv: I assume this ipc test isn't you?
[01:12:21] <thakis> neither arv's nor mine cl look like likely causes
[01:12:45] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open (ipc test flake?)
[01:14:10] *** viro101 has joined #chromium
[01:14:18] <thakis> akalin: the ipc_tests faiure is in sync. does the failing test name ring a bell?
[01:14:30] <thakis> akalin: (SyncSocketTest.SanityTest)
[01:15:05] *** eglaysher_lin has quit IRC
[01:15:12] <akalin> no, doesn't sound familiar
[01:15:17] <akalin> was there a sync checkin recently?
[01:15:22] *** eglaysher has joined #chromium
[01:15:22] *** ChanServ sets mode: +v eglaysher
[01:15:30] <thakis> not really
[01:16:03] <thakis> akalin: it's not really sync code either now that i look at it :-P
[01:16:18] <akalin> oh
[01:16:24] *** MichealH has joined #chromium
[01:18:14] <vandebo> greened up on the next run
[01:18:36] <evmar> i'll give my change a 70% success probability
[01:18:48] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[01:19:10] *** tav_ has joined #chromium
[01:20:28] *** tav has quit IRC
[01:20:29] *** tav_ is now known as tav
[01:20:44] *** fta_ has joined #chromium
[01:22:24] <thakis> evmar: wee!
[01:22:32] *** fta has quit IRC
[01:24:21] <vandebo> eroman: it looks like there are som sporadic heapchecker network errors in relation to proxies and utf16.  Are you aware?
[01:24:38] <eroman> vandebo: no, lemme look
[01:25:28] *** bent-mozilla has quit IRC
[01:27:03] *** chronarion has left #chromium
[01:28:08] *** dantasse has joined #chromium
[01:28:48] <dantasse> heyo!  got a grd change.  should I try to land it now, in a bit, or tomorrow?
[01:29:09] *** happygrue_ has joined #chromium
[01:29:37] <dantasse> ach, wait, nm, still waiting on a review
[01:33:10] *** happygrue has quit IRC
[01:35:40] <dave_levin> tkent: ping
[01:35:44] <kinuko> dantasse: is your change only for grd or does that include other changes?
[01:36:01] *** rafaelw has left #chromium
[01:36:06] <dantasse> it has a couple other changes
[01:36:21] <dantasse> I could split it up, if that helps
[01:37:36] <vandebo> dave_levin: fast/js/comparison-operators-less.html is randomly failing on linux. Do I add a suppression just in webkit/tools/layout_tests/test_expectations.txt, or do I need to file it upstream as well?
[01:39:01] *** hinoka|away has quit IRC
[01:39:37] <dave_levin> vandebo: Feel free to put it in downstream and file a bug etc.
[01:40:09] <dave_levin> vandebo: to clean up the bot. I can put it in upstream and remove it downstream, etc.
[01:42:05] <thakis> andybons: whoa, we have grey close buttons now? nice!
[01:42:20] <thakis> i like the dark grey too
[01:42:24] <kinuko> dantasse:  chromium builders seem to be getting ready.  are you familiar with instructions for landing grd changes?
[01:42:57] <andybons> thakis: yeah. and darker chrome finally
[01:43:48] *** yutak_home has joined #chromium
[01:43:48] *** ChanServ sets mode: +v yutak_home
[01:44:14] *** Alystair has joined #chromium
[01:44:24] <andybons> thakis: just need to figure out a good way to make the hover states have a bit more depth like the standard 'x' does now with it's inner shadow.
[01:44:30] <andybons> its
[01:44:41] *** Venom_X has quit IRC
[01:45:15] *** mrossetti_ has quit IRC
[01:46:39] <dantasse> kinuko: sorry, i forgot it's still waiting on a code review
[01:46:52] <dantasse> i'll wait till tomorrow.
[01:47:04] *** markmentovai has quit IRC
[01:47:05] <kinuko> dantasse: ah ok :)
[01:47:48] <dantasse> (ps: I'm submitting a different change right now, don't worry if you see my name on the waterfall :)
[01:48:33] *** andybons has quit IRC
[01:49:04] <thakis> andybons: i can't stand the dev channel upper chrome any more now that i've played with my tot build a bit
[01:49:10] <kinuko> dantasse: thx for the headsup :)
[01:50:17] <mirandac> nshkrob: ping
[01:50:49] *** RT|Chatzilla has joined #chromium
[01:51:42] *** arv has joined #chromium
[01:51:42] *** ChanServ sets mode: +v arv
[01:51:47] *** eseidel has joined #chromium
[01:54:25] *** rsesek has quit IRC
[01:55:24] *** cbentzel has quit IRC
[01:56:11] *** RT|Chatzilla has quit IRC
[01:58:29] *** dpranke has quit IRC
[01:58:29] *** bmizerany has quit IRC
[01:58:29] *** agl has quit IRC
[01:58:30] *** lianj has quit IRC
[01:58:31] *** davidben has quit IRC
[01:58:31] *** peper has quit IRC
[01:58:31] *** nshkrob has quit IRC
[01:58:32] *** erk_ has quit IRC
[01:58:32] *** c_zahmad_ has quit IRC
[01:58:33] *** leavittx has quit IRC
[01:58:33] *** brettw has quit IRC
[01:58:33] *** oshima has quit IRC
[01:58:33] *** faitz has quit IRC
[01:58:34] *** mattijle has quit IRC
[01:58:34] *** nirnimesh has quit IRC
[01:58:35] *** rellig has quit IRC
[01:58:35] *** Derevko has quit IRC
[01:58:35] *** kcliu has quit IRC
[01:58:35] *** hasimo_ has quit IRC
[01:58:35] *** bryeung has quit IRC
[01:59:47] *** RT|Chatzilla has joined #chromium
[01:59:51] *** arv is now known as arv_afk
[02:00:24] <evmar> 15 pendign changes
[02:01:08] *** fta__ has joined #chromium
[02:01:12] <trungl-bot> Tree closed by evan at chromium dot org (:evmar): Tree is closed, letting bots catch up
[02:02:26] *** fta_ has quit IRC
[02:02:39] <kinuko> evmar: thanks...
[02:02:47] *** nshkrob has joined #chromium
[02:02:47] *** ChanServ sets mode: +v nshkrob
[02:04:15] *** erk__ has joined #chromium
[02:05:34] <pathorn> evmar: the only change I made was aliasing __malloc_usable_size as well
[02:06:19] <pathorn> should I try another build with exactly what's in your cl?
[02:07:30] <Alystair> did you guys see my fun bug involving transparent gifs: http://code.google.com/p/chromium/issues/detail?id=48054#c1
[02:07:32] <evmar> pathorn: ah, didn't know you were on here
[02:07:59] <evmar> pathorn: will claimed that they should be the same (?)
[02:08:06] <Alystair> basically transparent gif is rendering the background -1 darker, so #ffffff becomes #fefefe :)
[02:08:09] <pathorn> ah okay
[02:08:12] <evmar> i don't know what i'm doing at all, and i'm not encountering this problem, but i am ok to impl it
[02:08:35] <evmar> Alystair: interesting, probably a rounding error somewhere?
[02:09:47] <thakis> evmar: stop making conflicting edits :-P
[02:09:57] *** 13WAA3UUX has joined #chromium
[02:09:58] *** dpranke has joined #chromium
[02:09:58] *** bmizerany has joined #chromium
[02:09:58] *** agl has joined #chromium
[02:09:58] *** lianj has joined #chromium
[02:09:58] *** peper has joined #chromium
[02:09:58] *** c_zahmad_ has joined #chromium
[02:09:58] *** brettw has joined #chromium
[02:09:58] *** oshima has joined #chromium
[02:09:58] *** faitz has joined #chromium
[02:09:58] *** mattijle has joined #chromium
[02:09:58] *** nirnimesh has joined #chromium
[02:09:58] *** zelazny.freenode.net sets mode: +vvvv agl brettw oshima nirnimesh
[02:09:58] *** rellig has joined #chromium
[02:09:58] *** Derevko has joined #chromium
[02:09:58] *** kcliu has joined #chromium
[02:09:58] *** hasimo_ has joined #chromium
[02:09:58] *** bryeung has joined #chromium
[02:09:58] *** ChanServ sets mode: +v dpranke
[02:10:07] <Alystair> No idea, I'm not a developer yet, just an odd bug reporter
[02:10:07] <thakis> i said the same thing, ccd the same people, but also filled in area
[02:10:11] <Alystair> I only noticed it because my monitor has some really insane gamma issues, I can easily tell #fefefe from #ffffff :D
[02:10:12] <thakis> for naught!
[02:10:24] *** peper has quit IRC
[02:10:39] *** peper has joined #chromium
[02:10:56] *** arv_afk is now known as arv
[02:13:45] *** Ambiguity has quit IRC
[02:15:10] <evmar> thakis: i should just wait 30 seconds before doing antyhign
[02:16:26] <thakis> everyone should wait hash(employee_id) % 61 seconds
[02:17:31] <evmar> http://codereview.chromium.org/2850042 , finally killing of the last of the file_util functions that use wstring
[02:17:34] <thakis> "in the future, everyone will wait for 15 seconds"
[02:17:34] <evmar> *off
[02:18:32] *** c_zahmad__ has joined #chromium
[02:18:39] *** evmar is now known as evmar_afk
[02:18:48] <thakis> woot
[02:20:19] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[02:21:02] *** smparkes has quit IRC
[02:21:28] *** markmentovai has joined #chromium
[02:21:28] *** ChanServ sets mode: +v markmentovai
[02:22:03] *** dmaclach has quit IRC
[02:22:15] *** c_zahmad_ has quit IRC
[02:25:27] *** mattijle has quit IRC
[02:26:09] *** mattijle has joined #chromium
[02:30:14] *** ph0nk has joined #chromium
[02:30:29] *** ph0nk has quit IRC
[02:30:51] *** 13WAA3UUX has quit IRC
[02:31:06] *** trungl has quit IRC
[02:31:20] *** oshima1 has joined #chromium
[02:31:20] *** ChanServ sets mode: +v oshima1
[02:32:37] *** brettw_ has joined #chromium
[02:32:37] *** ChanServ sets mode: +v brettw_
[02:32:47] <cyberix> branding <3 ;-) http://www.cs.helsinki.fi/u/twruottu/dwst.html
[02:34:30] <cyberix> ...also support for setting the protocol
[02:35:08] *** oshima has quit IRC
[02:35:08] *** brettw has quit IRC
[02:35:08] *** agl has quit IRC
[02:35:27] *** agl has joined #chromium
[02:35:27] *** ChanServ sets mode: +v agl
[02:36:53] *** arv is now known as arv_afk
[02:38:27] *** tonyg-cr has quit IRC
[02:42:31] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "net_unittests" on "Modules Mac10.5" from 51468: levin at chromium dot org (:dave_levin), mpcomplete at chromium dot org)
[02:42:36] *** leavittx_ has joined #chromium
[02:42:49] *** fta_ has joined #chromium
[02:43:39] <dave_levin> I don't see how either of those changes would have caused that issue.
[02:43:45] <vandebo> nope
[02:43:47] *** fta__ has quit IRC
[02:43:47] *** jamesr has joined #chromium
[02:43:47] *** ChanServ sets mode: +v jamesr
[02:44:08] *** maro has quit IRC
[02:44:33] <rvargas> That's related to my change :(
[02:44:38] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[02:44:51] *** Martijnc has quit IRC
[02:44:52] *** maro has joined #chromium
[02:45:13] <vandebo> ahh, didn't see you had a recent change.  I just forwarded to you
[02:46:31] *** victorw has left #chromium
[02:47:11] *** erikkay has left #chromium
[02:50:51] <thakis> is there some test that uses both the ui and io threads?
[02:51:19] *** hbono has joined #chromium
[02:52:30] *** Dogmeat_ has quit IRC
[02:53:04] *** urbanape has joined #chromium
[02:56:34] *** Sarten-X has quit IRC
[02:57:30] <vandebo> rvargas: looks like disk cache tests are failing on the mac memory bot since your checkin: DiskCacheBackendTest.SetSize, DiskCacheBackendTest.NewEvictionSetSize
[02:57:41] <vandebo> you should fix or revert soonish
[02:59:15] <rvargas> yeah... I'm looking at that... Actually I want to see what the other memory bots say
[02:59:55] *** markmentovai has quit IRC
[03:00:08] <rvargas> I guess I'll revert again
[03:00:23] *** yutak_home has quit IRC
[03:00:51] <vandebo> there's a linux_valgrind try bot
[03:01:48] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Mac10.5 Tests" from 51468: albertb at chromium dot org, dantasse at chromium dot org, davidben at chromium dot org, dpranke at chromium dot org, eroman at chromium dot org (:eroman), gman at chromium dot org, jcivelli at chromium dot org, levin at chromium dot org (:dave_levin), mpcomplete at chromium dot org, rvargas at google dot com, stuartmorgan at chromium dot org, vandebo at chromium dot org)
[03:03:09] <thakis> yay, not me for a change
[03:04:10] <dpranke> my name is in the group but I really doubt it's my change (password mgr stuff)
[03:05:26] <kinuko> hmm
[03:05:35] <kinuko> Terminating app due to uncaught exception 'NSInternalInconsistencyException', reason: 'Error (1007) creating CGSWindow'
[03:07:46] <rvargas> vandebo: yeah... I tried that, and the mac one and they failed to build unit_tests (out of mem)
[03:08:02] <vandebo> the link problem, yea
[03:08:30] <rvargas> vandebo: And this time there is no race or leak detected... I reverted
[03:08:32] *** Dogmeat_ has joined #chromium
[03:09:07] *** fta_ has quit IRC
[03:09:08] *** dpranke has quit IRC
[03:09:08] *** bmizerany has quit IRC
[03:09:09] *** lianj has quit IRC
[03:09:11] *** faitz has quit IRC
[03:09:11] *** nirnimesh has quit IRC
[03:09:11] *** rellig has quit IRC
[03:09:12] *** Derevko has quit IRC
[03:09:12] *** kcliu has quit IRC
[03:09:12] *** hasimo_ has quit IRC
[03:09:12] *** bryeung has quit IRC
[03:10:15] *** cbentzel has joined #chromium
[03:12:49] <vandebo> kinuko: I wonder if there's something wrong with that bot
[03:13:08] <kinuko> vandebo: yeah I'm looking... many tests are failing only on the bot
[03:14:30] *** kenneth_reitz has joined #chromium
[03:15:27] <vandebo> kinuko: got the thestig to reboot that bot
[03:16:01] <kinuko> vandebo: hope it fixes problems
[03:16:29] <vandebo> stuartmorgan said that NSInternalInconsistencyException generally means that the window system is horked
[03:16:50] <kinuko> oh I see.
[03:17:23] *** eseidel has quit IRC
[03:17:57] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[03:18:56] *** leeight has joined #chromium
[03:19:00] *** leeight has left #chromium
[03:20:53] *** Martijnc has joined #chromium
[03:20:53] *** fta_ has joined #chromium
[03:20:53] *** dpranke has joined #chromium
[03:20:53] *** bmizerany has joined #chromium
[03:20:53] *** lianj has joined #chromium
[03:20:53] *** faitz has joined #chromium
[03:20:53] *** nirnimesh has joined #chromium
[03:20:53] *** rellig has joined #chromium
[03:20:53] *** Derevko has joined #chromium
[03:20:53] *** kcliu has joined #chromium
[03:20:53] *** hasimo_ has joined #chromium
[03:20:53] *** bryeung has joined #chromium
[03:20:53] *** zelazny.freenode.net sets mode: +vv dpranke nirnimesh
[03:21:26] *** bmizerany has quit IRC
[03:21:31] *** bmizerany has joined #chromium
[03:22:40] *** kenneth_reitz has quit IRC
[03:23:05] *** thakis_ has joined #chromium
[03:23:05] *** ChanServ sets mode: +v thakis_
[03:26:59] *** thakis__ has joined #chromium
[03:27:02] *** ChanServ sets mode: +v thakis__
[03:27:10] *** pathorn has quit IRC
[03:27:28] <kinuko> Cannot use V8 Proxy resolver in single process mode.
[03:27:42] <kinuko> is this change related to 51464?
[03:28:13] *** kinuko has left #chromium
[03:28:23] *** kinuko has joined #chromium
[03:28:23] *** ChanServ sets mode: +v kinuko
[03:29:02] <dave_levin> eroman: ^^
[03:29:43] <eroman> kinuko: dave_levin: No, that is a warning that we always output, shouldn't be related...
[03:29:47] *** thakis_ has quit IRC
[03:30:07] <kinuko> eroman: ah I see.  thanks
[03:31:43] *** tedoc2000 has quit IRC
[03:31:48] <kinuko> so some other change may have broken chrome_frame_net_tests
[03:32:42] *** btipling has joined #chromium
[03:33:44] <kinuko> dantasse: ping
[03:33:51] <vandebo> the reliability bot crashed with the disk cache change, so the chrome frame net test failure maybe silently related.  Let's see what it does after the revert
[03:34:22] <kinuko> vandebo: got it.  let's see...
[03:34:37] *** thakis__ has quit IRC
[03:35:47] *** rolandsteiner has joined #chromium
[03:37:36] *** bmizerany has quit IRC
[03:38:51] *** thakis is now known as thakis_afk
[03:39:10] <vandebo> kinuko: is that a real mac perf regression?
[03:41:02] <eroman> vandebo: the ie7 chromeframe looks happy with the revert
[03:41:42] *** barcon332 has joined #chromium
[03:45:29] <vandebo> erg: ping
[03:46:16] <kinuko> vandebo: I doubt it.
[03:46:45] <vandebo> 51450 seems to be making heapcheck bot unhappy
[03:46:48] <vandebo> I guess I'll revert
[03:51:01] *** atwilson has quit IRC
[03:51:01] *** atwilson_ has joined #chromium
[03:51:01] *** ChanServ sets mode: +v atwilson_
[03:51:01] *** atwilson_ is now known as atwilson
[03:51:59] *** rafaelw2 has joined #chromium
[03:54:07] *** Martijnc has quit IRC
[03:56:43] *** fta__ has joined #chromium
[03:57:16] *** fta_ has quit IRC
[03:57:20] <vandebo> kinuko: I'm heading out
[03:58:17] <kinuko> vandebo:  thanks for doing the major work
[03:58:47] <kinuko> vandebo: I'll be around for a while
[04:01:08] <vandebo> oh, you might want to checkout the chromeos arm bot
[04:02:35] *** csilv has left #chromium
[04:03:48] <kinuko> vandebo: chromeos arm bot?
[04:03:59] *** stuartmorgan has quit IRC
[04:04:05] *** btipling has left #chromium
[04:04:18] *** bobthemilkman has joined #chromium
[04:10:25] *** ironfroggy has joined #chromium
[04:10:35] <ironfroggy> What is the appropriate channel for extension development?
[04:11:58] *** [Nef] has quit IRC
[04:12:17] *** bweinstein has left #chromium
[04:14:27] *** [Nef] has joined #chromium
[04:23:37] *** tonyg-cr has joined #chromium
[04:23:37] *** ChanServ sets mode: +v tonyg-cr
[04:23:55] *** cbentzel has quit IRC
[04:25:21] *** rubenbb has quit IRC
[04:26:56] *** rubenbb has joined #chromium
[04:29:43] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Linux Perf (2)" from 51480: suzhe at chromium dot org)
[04:32:07] <kinuko> suzhe: are you around?
[04:36:00] <kinuko> (trying to catch him on chat)
[04:36:12] *** dpranke has quit IRC
[04:38:28] *** Ambiguity has joined #chromium
[04:39:51] <trungl-bot> Tree closed by kinuko at google dot com: Tree is closed (compile error -> suzhe, kinuko will revert)
[04:40:25] *** fta_ has joined #chromium
[04:41:16] *** fta__ has quit IRC
[04:42:36] *** chrisccoulson has quit IRC
[04:43:04] <vandebo> any troopers around? Mac release bot must be wedged
[04:43:19] <vandebo> kinuko: (I was mistaken about arm)
[04:45:55] <trungl-bot> Tree opened by kinuko at google dot com: Tree is open (revert in)
[04:47:14] *** cbentzel has joined #chromium
[04:48:54] *** tav has quit IRC
[04:50:56] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Linux" from 51480: suzhe at chromium dot org)
[04:50:58] *** satorux has joined #chromium
[04:51:24] <satorux> I'm seeing this: net/ocsp/nss_ocsp.cc:640: error: NULL used in arithmetic
[04:51:28] *** jschuh|home has joined #chromium
[04:51:56] <trungl-bot> Tree opened by kinuko at chromium dot org: Tree is open (revert in)
[04:52:52] *** cbentzel has quit IRC
[04:54:12] <kinuko> satorux: I chatted with suzhe, seems like nss_oscp.cc change was wrongly included in the patch.
[04:54:20] <satorux> ah kinuko's revert seems to fix this
[04:54:55] <kinuko> satorux: yeah I just reverted it after chatting with suzhe.  He might be happy if you can let him know about the error though :)
[04:55:03] <satorux> I pinged him
[04:55:20] <satorux> looks he accidentally included an unrelated change when he landed his change
[04:55:47] *** roc has quit IRC
[04:56:01] <kinuko> satorux: yeah, that's what he said on chat.
[04:57:58] *** roc has joined #chromium
[04:59:04] *** awolfson has quit IRC
[05:12:15] *** awolfson has joined #chromium
[05:14:55] *** alt173 has joined #chromium
[05:20:01] *** bradleymeck2 has joined #chromium
[05:20:17] *** eseidel has joined #chromium
[05:20:23] *** thakis has joined #chromium
[05:20:23] *** ChanServ sets mode: +v thakis
[05:21:47] *** fta__ has joined #chromium
[05:23:14] *** fta_ has quit IRC
[05:23:47] *** eseidel has quit IRC
[05:24:46] *** WeaponX has joined #chromium
[05:26:59] *** eseidel has joined #chromium
[05:27:06] *** WeaponX has left #chromium
[05:29:25] *** abarth has joined #chromium
[05:29:25] *** ChanServ sets mode: +v abarth
[05:30:42] *** alt173 has left #chromium
[05:31:19] *** rjb has joined #chromium
[05:40:46] *** barcon332 has quit IRC
[05:40:55] *** tav has joined #chromium
[05:41:52] <rjb> I got the message Program received signal SIGPIPE, Broken pipe. I have Ubuntu 10.04 Chromium
[05:42:33] *** DBO has quit IRC
[05:43:24] *** bradleymeck2 has quit IRC
[05:49:30] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Linux Builder x64" from 51489: brettw at chromium dot org)
[05:51:24] <kinuko> brettw_: ping
[05:51:31] <trungl-bot> Tree closed by brettw at chromium dot org: Tree is closed (Brett looing at 64-bit issue)
[05:51:56] <kinuko> brettw_: thx
[05:56:08] <tkent> I'm working on "WebKit (dbg)(2)" failure.
[05:56:54] <FullFlannelJacke> Any "extension" devs here?
[05:57:18] <FullFlannelJacke> I want to request someone write a copy of the Frefox "HTTPS EverywherE" extension for Chromium.
[05:57:35] <trungl-bot> Tree closed by tkent at chromium dot org: Tree is closed (Brett looing at 64-bit issue, WebKit (dbg)(2) -> tkent)
[05:59:39] <dave_levin> tkent: I thought I disabled that :( .. Oh on this bot it is IMAGE+TEXT but I only added IMAGE.
[06:02:06] <tkent> dave_levin: The test shows a CHECK error for debug build.  I already have a patch; http://codereview.chromium.org/2876036/show
[06:03:18] *** jamesr has quit IRC
[06:03:38] <trungl-bot> Tree closed by brettw at chromium dot org: Tree is closed (Brett hopefully fixed Linux64)
[06:04:25] *** tonyg-cr has quit IRC
[06:07:45] *** FullFlannelJacke has quit IRC
[06:08:14] *** michaeln has quit IRC
[06:14:39] *** rjb has left #chromium
[06:14:45] <trungl-bot> Tree opened by brettw at chromium dot org: ????TREE OPEN????
[06:19:12] *** gonzojive has joined #chromium
[06:20:13] <gonzojive> is there an api for chromium so I can use it as a library?  I'd like to do screen scraping with a full-fledge web browser as the crawling agent
[06:24:26] *** temp01 has quit IRC
[06:25:24] *** FullFlannelJacke has joined #chromium
[06:25:24] *** temp01 has joined #chromium
[06:26:06] *** dave_levin has quit IRC
[06:27:22] *** rafaelw2 has quit IRC
[06:27:29] <tkent> gonzojive: test_shell in Chromium tree and DumpRenderTee in WebKit tree can be used for such purpose though they are applications, not libraries.
[06:29:17] <gonzojive> tkent: ideally I am looking to do more than screen scrape: execute arbitrary javascript code, perform javascript / page rendering unit tests, and other cool things chromium to do.   it's essential to work with a browser as a library
[06:31:33] *** dmaclach has joined #chromium
[06:31:56] <trungl-bot> Tree opened by tkent at chromium dot org: ????TREE OPEN???? (WebKit (dbg)(2) -> tkent)
[06:33:50] *** eseidel has quit IRC
[06:35:57] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Mac10.5 Tests" from 51493: brettw at chromium dot org, hbono at chromium dot org, tim at chromium dot org (:timsteele))
[06:36:15] *** dmaclach_ has joined #chromium
[06:36:58] <trungl-bot> Tree opened by brettw at chromium dot org: Tree is open (Brett should have fixed Windows compile)
[06:39:33] *** dmaclach has quit IRC
[06:39:34] *** dmaclach_ is now known as dmaclach
[06:43:39] *** fta_ has joined #chromium
[06:43:51] *** MikeSmithXX has joined #chromium
[06:45:15] *** fta__ has quit IRC
[06:47:00] *** MikeSmithW3C^ has quit IRC
[06:51:16] *** lolsuper_ has joined #chromium
[06:56:34] *** Beetny` has joined #chromium
[06:57:09] *** dmaclach_ has joined #chromium
[06:58:07] *** fta_ is now known as fta
[06:59:02] *** Beetny has quit IRC
[06:59:49] *** mitchell687 has joined #chromium
[07:00:42] *** dmaclach has quit IRC
[07:00:43] *** dmaclach_ is now known as dmaclach
[07:05:09] *** dmaclach has quit IRC
[07:06:52] *** dmaclach has joined #chromium
[07:20:25] *** dmaclach has quit IRC
[07:28:17] *** eroman has quit IRC
[07:28:48] *** Beetny has joined #chromium
[07:30:43] *** Beetny` has quit IRC
[07:30:54] *** thakis has quit IRC
[07:33:20] *** zloidemon has joined #chromium
[07:42:51] *** thakis has joined #chromium
[07:42:51] *** ChanServ sets mode: +v thakis
[07:42:59] <thakis> can i land a grd change?
[07:43:26] <thakis> kinuko: ^
[07:43:29] *** abarth has quit IRC
[07:44:56] *** Alystair has quit IRC
[07:44:59] <kinuko> thakis: oh, yeah.  the tree looks good now.
[07:45:57] <kinuko> thakis: do you want me to close the tree or will you do that?
[07:46:07] <thakis> do you think that's necessary?
[07:46:27] <thakis> doesn't look like there's a lot of commit traffic
[07:46:57] <kinuko> thakis: right.  probably it'll be ok.
[07:47:37] <jochen__> morning
[07:48:07] <thakis> kinuko: this will probably turn the tree red, but it should go green after the next cl that goes in once r51503 is done compiling on the windows builders
[07:48:10] <thakis> kinuko: thanks!
[07:48:13] <thakis> jochen__: morning
[07:48:19] <thakis> good night everyone
[07:48:21] *** thakis has quit IRC
[07:52:43] <trungl-bot> Tree opened by kinuko at chromium dot org: Tree is open (grd change is in; win bots will turn red after r51503)
[07:52:51] *** rolandsteiner has quit IRC
[07:57:05] *** dmaclach has joined #chromium
[08:06:47] *** temp01 is now known as away01
[08:06:55] *** dmaclach has quit IRC
[08:07:53] *** aboodman has quit IRC
[08:09:51] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium XP" from 51503: hamaji at chromium dot org, rsimha at chromium dot org, thakis at chromium dot org, tkent at chromium dot org)
[08:13:23] <tkent> This failure looks a tool problem.
[08:14:44] *** DBO has joined #chromium
[08:16:37] <kinuko> hmm
[08:16:58] <hamaji> tkent: any idea to fix this? maybe clober build?
[08:16:58] *** viro101 has quit IRC
[08:17:17] <kinuko> iirc last time it got fixed by itself on the next build
[08:17:47] *** aboodman has joined #chromium
[08:17:47] *** ChanServ sets mode: +v aboodman
[08:18:01] <hamaji> kinuko: i see. let's just wait for a while
[08:18:12] <tkent> agree
[08:18:36] *** rsimha has joined #chromium
[08:20:12] *** gavin has quit IRC
[08:20:31] *** gavin has joined #chromium
[08:20:31] *** gavin has joined #chromium
[08:20:57] <trungl-bot> Tree closed by kinuko at chromium dot org: Tree is closed (compile error -> tool problem? waiting the next build)
[08:23:23] *** [Nef] has quit IRC
[08:26:46] *** viro101 has joined #chromium
[08:28:02] *** General1337 has joined #chromium
[08:28:14] *** rsimha has quit IRC
[08:28:25] *** rsimha has joined #chromium
[08:29:39] *** trungl has joined #chromium
[08:29:39] *** ChanServ sets mode: +v trungl
[08:29:55] *** MichealH|Laptop has joined #chromium
[08:30:08] <trungl-bot> Tree opened by kinuko at chromium dot org: Tree is open (chromium xp compile error -> waiting next build)
[08:31:14] *** General13372 has quit IRC
[08:34:53] *** aboodman has quit IRC
[08:35:28] *** aboodman has joined #chromium
[08:35:28] *** ChanServ sets mode: +v aboodman
[08:35:29] *** BryanWB has joined #chromium
[08:37:13] *** rsimha has quit IRC
[08:37:47] <kinuko> ooh tree is turning red...
[08:41:18] <trungl-bot> Tree closed by hamaji at chromium dot org: Tree is closed (many red)
[08:41:36] <hamaji> i closed the tree
[08:43:44] <kinuko> hamaji: thanks - I should have done that
[08:45:17] <hamaji> tkent: is the redness relevant to your change?
[08:46:18] <tkent> hamaji: yes.  My change contains a grd change for Linux. I thought it didn't affect the Windows grd issue, but actually it did.
[08:46:32] <tkent> hamaji: These redness should be clearerd by 51504
[08:47:36] *** Erkan_Yilmaz has quit IRC
[08:59:46] *** MichealH|Laptop has quit IRC
[09:02:29] <trungl-bot> Tree opened by kinuko at chromium dot org: Tree is open (should cycle green)
[09:10:16] <leiz> arg, looks like I need to kick the mac bot again
[09:11:29] *** hayato has quit IRC
[09:11:38] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "update" on "Chrome Frame Tests (ie8)")
[09:13:36] <leiz> the mac bot looks sick so I rebooted it
[09:14:35] <leiz> it's starting the next run, let's hope it turns green.
[09:20:10] *** phajdan-jr has joined #chromium
[09:20:10] *** ChanServ sets mode: +v phajdan-jr
[09:22:45] <trungl-bot> Tree opened by tkent at chromium dot org: Tree is open
[09:26:28] *** star-affinity has joined #chromium
[09:30:16] <jochen__> another grd change...
[09:31:37] <leiz> arg
[09:32:47] <leiz> I'll roll deps for gyp another day...
[09:33:29] *** [Nef] has joined #chromium
[09:38:21] *** eseidel has joined #chromium
[09:47:58] *** pfeldman has joined #chromium
[09:51:21] *** roc has quit IRC
[09:53:15] *** buenol has joined #chromium
[09:56:15] <buenol> hi,two questions (sorry) :
[09:56:43] <buenol> _ why <video> have no fullscreen and is buggy/unintuitive ?
[09:57:46] <leiz> see /topic
[09:58:02] <buenol> _ is the sync bookmarks feature planed to have fonctions like "revision" or "cancel sync" (to restore bookmarks)
[09:58:35] <buenol> (it's dev questions)
[09:59:14] <leiz> no idea, you can file a feature request on crbug.com though
[10:00:30] <buenol> okay, i will do for "sync bookmarks" but "full-screen video" seems blocked like "rss subscription"
[10:01:56] <leiz> if you don't think full-screen video is working correctly, you can file a bug for that too
[10:03:37] <buenol> yep it's http://code.google.com/p/chromium/issues/detail?id=16735
[10:04:18] <leiz> well, there ya go. asking here won't make it happen faster
[10:04:38] <buenol> (it's what i do)
[10:05:01] <leiz> which is why we tell you to read /topic
[10:06:16] <leiz> reliability bot's shell step should turn green on the next run
[10:06:26] *** jschuh has quit IRC
[10:06:39] *** rubenbb has quit IRC
[10:08:55] *** hwennborg has joined #chromium
[10:09:24] *** zloidemon has quit IRC
[10:09:31] <trungl-bot> Tree opened by thestig at chromium dot org (:leiz): Tree is open (Chromium Mac still needs a bit of attention)
[10:09:46] <leiz> g'nite #chromium!
[10:10:08] <buenol> (nice new scrollbar)
[10:10:15] *** rubenbb has joined #chromium
[10:19:52] *** urbanape has quit IRC
[10:20:21] *** tittiatcoke has joined #chromium
[10:24:51] *** rubenbb has quit IRC
[10:27:01] *** buenol has quit IRC
[10:28:15] *** ROBOd has joined #chromium
[10:29:49] *** hrna has joined #chromium
[10:30:10] *** Dykam has quit IRC
[10:32:07] *** satish_ has joined #chromium
[10:32:39] <satish_> yurys: Good morning, things going ok?
[10:32:46] *** rubenbb has joined #chromium
[10:45:09] *** GeekShadow has joined #chromium
[10:45:32] *** FullFlannelJacke has quit IRC
[10:46:45] *** FullFlannelJacke has joined #chromium
[10:50:09] *** bauerb has joined #chromium
[10:50:09] *** ChanServ sets mode: +v bauerb
[10:51:09] *** GeekShadow has quit IRC
[10:55:10] *** leeight has joined #chromium
[10:55:13] *** leeight has left #chromium
[10:58:03] *** chrisccoulson has joined #chromium
[11:01:04] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Vista Tests (dbg)(1)" from 51508: hamaji at chromium dot org, jochen at chromium dot org (:jochen__))
[11:02:42] *** urbanape has joined #chromium
[11:05:05] <trungl-bot> Tree opened by satish at chromium dot org: Tree is open (looking at Vista Tests)
[11:05:29] <jochen__> doesn't look relatred to my change :/
[11:07:02] *** ukai__ has joined #chromium
[11:07:03] *** ChanServ sets mode: +v ukai__
[11:07:11] <satish_> yes it could be a flaky run, let's see.. I've forced another build
[11:09:45] *** ukai_ has quit IRC
[11:13:09] *** rolandsteiner has joined #chromium
[11:20:17] <jochen__> it's green again in the latest revision
[11:22:55] <satish_> jochen__: yes, it was just flaky, not your CL :)
[11:28:05] *** tkent has quit IRC
[11:28:14] <trungl-bot> Tree closed by bauerb at chromium dot org: Tree is closed (bauerb landing a GRD change)
[11:29:26] <satish_> bauerb: Will you also clobber the windows builders after landing the change?
[11:29:55] <bauerb> satish: i wasn't planning on doing it, because i've been told that this doesn't help things
[11:30:20] <satish_> bauerb: So planning to just wait for 2 windows builds to happen and cycle green?
[11:30:34] <bauerb> satish_: i'm checking in a whitespace change
[11:31:04] <bauerb> satish_: http://dev.chromium.org/developers/tree-sheriffs/sheriff-details-chromium, section "Landing a GRD (Resources) Change"
[11:32:11] <satish_> bauerb: ok, let's see :)
[11:33:26] <bauerb> satish_: if i screw things up, we still can clobber to our heart's desire ;)
[11:34:11] <satish_> bauerb: yes of course, hopefully it doesn't go as far as that :D
[11:36:27] *** hbono has quit IRC
[11:44:02] *** urbanape has quit IRC
[11:51:30] *** roc has joined #chromium
[11:55:13] *** lianj_ has joined #chromium
[11:58:18] *** lianj has quit IRC
[11:58:42] *** urbanape has joined #chromium
[12:04:52] *** hbono has joined #chromium
[12:08:15] <bauerb> satish_: i'm going for lunch. would you mind reopening the tree if it looks good?
[12:10:25] <satish_> bauerb: sure
[12:10:29] <bauerb> thanks
[12:11:05] <satish_> bauerb: as I see now a whole bunch of windows builders are failing, will wait until they cycle green
[12:23:59] *** hbono has quit IRC
[12:27:15] *** bobthemilkman has quit IRC
[12:28:54] *** urbanape has quit IRC
[12:36:07] *** rubenbb has quit IRC
[12:36:29] *** hebz0rl has joined #chromium
[12:37:52] <trungl-bot> Tree opened by satish at chromium dot org: Tree is open
[12:40:05] *** zaspire has joined #chromium
[12:43:08] *** shepazu has quit IRC
[12:44:55] <trungl-bot> Tree opened by satish at chromium dot org: Tree is open (looking at Chrome Frame Tests)
[12:46:37] *** MikeSmithW3C has joined #chromium
[12:47:31] *** MikeSmithXX has quit IRC
[12:48:00] *** eseidel has quit IRC
[12:52:46] *** andybons has joined #chromium
[12:54:55] *** Martijnc has joined #chromium
[13:01:52] *** rubenbb has joined #chromium
[13:05:21] *** urbanape has joined #chromium
[13:08:56] *** phajdan-jr is now known as phajdan-jr|afk
[13:11:23] *** BryanWB has quit IRC
[13:12:38] *** cbentzel has joined #chromium
[13:14:23] *** ukai__ has quit IRC
[13:21:12] *** awolfson has quit IRC
[13:24:08] *** hebz0rl has quit IRC
[13:29:38] *** Mattias has joined #chromium
[13:30:27] <Mattias> How do I set which file manager should open up when I choose "open folder" on downloaded files?
[13:30:31] *** Dykam has joined #chromium
[13:30:38] <Mattias> right now it opens another browser instead..
[13:30:58] *** wr| has quit IRC
[13:31:06] <Mattias> I'm on latest chromium-dev
[13:31:52] *** urbanape has quit IRC
[13:32:55] *** thomasvl has joined #chromium
[13:32:55] *** ChanServ sets mode: +v thomasvl
[13:36:22] *** awolfson has joined #chromium
[13:38:20] <maruel> Mattias: see topic
[13:38:46] <satish_> Anyone from chromeFrame here knows why NoInterferenceTest.FLAKY_JavascriptWindowOpen keeps crashing and leaving the builder red repeatedly?
[13:39:54] *** ph0nk2 has quit IRC
[13:40:52] <thomasvl> fyi bouncing mac bot to help a slowed down cycle time
[13:46:48] <thomasvl> maruel: you looking at vista perf (dbg)
[13:46:53] <thomasvl> gclient steps failing
[13:47:34] *** rolandsteiner has quit IRC
[13:47:42] *** rolandsteiner has joined #chromium
[13:49:36] *** happygrue_ has quit IRC
[13:51:54] *** rolandsteiner has quit IRC
[13:57:53] *** tonikitoo has joined #chromium
[13:59:08] *** barcon332 has joined #chromium
[13:59:47] *** urbanape has joined #chromium
[14:01:34] *** Mattias has left #chromium
[14:03:08] <jochen__> landing another grd change... (bots clobbered)
[14:04:37] *** tonikitoo has quit IRC
[14:04:47] <jochen__> gnah... bulach was faster than me :/
[14:05:13] <bulach> jochen__: ops, sorry.. :)
[14:05:29] <jochen__> nvm
[14:05:49] <jochen__> i'll commit anyway and just close the tree
[14:05:56] *** cbentzel has quit IRC
[14:06:51] <trungl-bot> Tree closed by jochen at chromium dot org (:jochen__): Tree is closed (grd change from jochen)
[14:06:59] <bulach> jochen__: sounds good.. perhaps closing the tree when clobering? at non-MTV hours shouldn't be a big deal.. sorry anyway, I should've read the msg here..
[14:09:32] <jochen__> yeah
[14:09:43] <jochen__> i hope someone will finally fix the gyp thingy...
[14:10:29] <bulach> jochen__: +infinite for fixing the gyp.. :)
[14:11:59] <satish_> I'm disabling the chrome frame "NoInterferenceTest.FLAKY_JavascriptWindowOpen" test (when the tree reopens) as it keeps crashing and leaves the cr-frame builder red.
[14:12:33] *** andybons has quit IRC
[14:20:02] *** rolandsteiner has joined #chromium
[14:20:20] *** rolandsteiner has left #chromium
[14:24:22] <thomasvl> jochen__: yt?
[14:25:02] <thomasvl> jochen__: looks like you clobbered before your submit, so that whole build cycle is for not.  cause it doesn't have you change.  http://build.chromium.org/buildbot/waterfall/waterfall?builder=Chromium%20Builder
[14:26:41] <thomasvl> jochen__: the windows build failure looks like my settings change, reverting my change
[14:26:44] <jochen__> it's ok to clobber before you commit. if nobody commits in between :)
[14:26:46] <jochen__> yeah
[14:28:40] *** rubenbb has quit IRC
[14:32:57] *** urbanape has quit IRC
[14:36:06] <thomasvl> jochen__: you still have to make sure you commit before the bot starts the clobber
[14:37:34] *** awolfson has quit IRC
[14:43:19] <jochen__> yes
[14:43:31] <jochen__> but if you check the "clobber" flag, the next build will be clobbered
[14:43:47] <jochen__> so you can just wait until the build for the last revision started, then clobber, then commit
[14:45:27] *** AaronMT has joined #chromium
[14:46:40] *** rubenbb has joined #chromium
[14:56:04] *** leeight has joined #chromium
[14:56:07] *** leeight has left #chromium
[15:02:33] *** cbentzel has joined #chromium
[15:12:21] <thomasvl> jochen__: really?  you're hitting a force build button, so it's also triggering a build?
[15:14:32] <yurys> is anyone looking at Vista Perf(dbg) ?
[15:18:18] <jochen__> hum, actually dunno
[15:18:21] <jochen__> maybe you're right
[15:21:36] *** cbentzel has quit IRC
[15:22:36] <jochen__> i got another grd change
[15:24:06] <jochen__> this time i clobbered right after committing :)
[15:28:03] <satish_> jochen__: are all the GRD changes in now? Should I start monitoring and open the tree as they cycle green?
[15:28:52] <trungl-bot> Tree opened by jochen at chromium dot org (:jochen__): Tree is open (Move along. There's nothing to see here)
[15:28:53] <satish_> yurys: I remember someone above asked maruel if he can look at the builder.
[15:28:59] <satish_> jochen__: thanks.
[15:29:00] <jochen__> i think nothing should happen from my changes
[15:29:49] <mnissler> satish_: Hey, I have another grd change...
[15:30:22] *** rubenbb has quit IRC
[15:30:32] <mnissler> satish_: If you don't mind, I'll land that now ;)
[15:31:19] <satish_> jochen__: did at least one of the builders cycle green after your GRD changes?
[15:31:27] <satish_> mnissler: ummm.. waiting for jochen__'s reply
[15:31:51] <jochen__> i'm sitting right next to mnissler btw
[15:32:01] <jochen__> for the first one, all are green
[15:32:15] <jochen__> for the second one, the builders just started
[15:32:21] <satish_> ok :) mnissler, go ahead. I'll keep watching.
[15:32:32] <mnissler> ok, closing the tree...
[15:32:39] *** tittiatcoke has quit IRC
[15:33:31] *** awolfson has joined #chromium
[15:33:55] <trungl-bot> Tree closed by mnissler at chromium dot org: Tree is closed (mnissler: grd change)
[15:34:12] *** tonikitoo has joined #chromium
[15:37:26] <ironfroggy> What is the appropriate channel for extension development?
[15:37:39] <jochen__> the chromium-extensions mailing list
[15:41:06] *** aerajpoapap has joined #chromium
[15:41:57] <ironfroggy> no irc channel for it? alright, i'll check that out
[15:42:47] *** rafaelw2 has joined #chromium
[15:44:27] *** trungl_ has joined #chromium
[15:44:27] *** ChanServ sets mode: +v trungl_
[15:47:16] *** cbentzel has joined #chromium
[15:47:38] *** rafaelw2 has quit IRC
[15:48:08] *** trungl has quit IRC
[15:48:08] *** trungl_ is now known as trungl
[15:50:39] *** kenneth_reitz has joined #chromium
[15:55:26] *** rubenbb has joined #chromium
[15:56:43] *** Zaba has quit IRC
[16:04:51] *** skanduk has joined #chromium
[16:07:18] *** Zaba has joined #chromium
[16:09:12] *** Erkan_Yilmaz has joined #chromium
[16:09:46] *** aerajpoapap has quit IRC
[16:11:21] *** Beetny has quit IRC
[16:16:31] *** skanduk has quit IRC
[16:17:38] *** Ambiguity has quit IRC
[16:20:54] *** urbanape has joined #chromium
[16:22:14] *** rsesek has joined #chromium
[16:22:15] *** ChanServ sets mode: +v rsesek
[16:23:24] *** rohitrao has joined #chromium
[16:23:24] *** ChanServ sets mode: +v rohitrao
[16:24:23] *** tedoc2000 has joined #chromium
[16:25:35] *** dr_win has quit IRC
[16:27:33] <trungl> 'morning, Chromium.
[16:27:33] <trungl-bot> trungl: Good morning!
[16:28:09] *** tonyg-cr has joined #chromium
[16:28:10] *** ChanServ sets mode: +v tonyg-cr
[16:28:33] *** fta has joined #chromium
[16:33:46] <satish_> mnissler: looks like some builds are turning green now?
[16:34:08] <mnissler> satish_: The linux chromium OS builder hits a crash in unit_tests...
[16:34:35] <mnissler> satish_: Never saw that before on the trybots or locally. But it's possible I caused it.
[16:35:01] <mnissler> satish_: I'm in the process of trying to reproduce locally. Might need to roll back.
[16:35:35] <rsesek> mnissler: if you do have to roll back, leave the grd change in and just revert your code
[16:35:46] <rsesek> that way things will roll green faster
[16:36:02] <satish_> mnissler: er.. I thought the GRD change was the only one in?
[16:36:34] <satish_> mnissler: sorry I spoke too soon, i guess you have grd + code, so rsesek suggests to remove the code while leaving the grd changes in
[16:36:36] *** yakattak has joined #chromium
[16:36:38] <mnissler> satish_: no, it's a code change including grd changes
[16:36:57] <yakattak> quick question about mac os x snapshots, why are they so small all of the sudden?
[16:37:00] <mnissler> satish_, rsesek: ok
[16:37:13] <satish_> mnissler: sounds good then, any reason you think this crash has something to do with your change? since it is within OnWindowStateEvent which may not be using resourcse?
[16:37:16] <rsesek> yakattak: oh weird
[16:37:28] <rsesek> yakattak: beacuse they're broken
[16:37:58] <mnissler> satish_: I touched WidgetGtk, so I'm probably causing it. I'll revert.
[16:38:02] <yakattak> rsesek: they've been broken for like the past 15 revs atleast
[16:38:14] <satish_> mnissler: ok
[16:38:20] <rsesek> yakattak: do you have the rev# where it broke? I'm looking at the waterfall to see what's up
[16:38:36] <yakattak> i'm on the last one where it worked correctly, 1 sec
[16:38:46] <yakattak> worked correctly on 51362
[16:38:52] <yakattak> then after that it started getting screwed up
[16:38:56] *** beinghuman has joined #chromium
[16:39:05] <beinghuman> http://i.imgur.com/gY5mN.png
[16:39:13] <beinghuman> probably an issue with webkit, but I'm curious about it
[16:40:41] <rsesek> thomasvl: ping
[16:40:48] <beinghuman> for instance, why is chrome rendering these border-styles so ugly when opera and FF do it "right" imo
[16:40:50] <thomasvl> rsesek: pong
[16:40:59] <rsesek> thomasvl: yakattak discovered the build archive is broken
[16:41:05] <thomasvl> already fixed
[16:41:16] <thomasvl> thread in buzz also
[16:41:31] <rsesek> http://build.chromium.org/buildbot/waterfall/builders/Chromium%20Mac/builds/6512/steps/archive_build/logs/stdio is the latest run and it still happened ? is it super recently fixed?
[16:41:52] <thomasvl> y
[16:41:55] *** desu has joined #chromium
[16:42:01] <rsesek> awesome. yakattak^
[16:42:18] <thomasvl> rsesek: http://src.chromium.org/viewvc/chrome?view=rev&revision=51528
[16:42:20] *** andybons has joined #chromium
[16:42:29] <rsesek> ah weird
[16:42:36] <yakattak> so next rev it should work?
[16:42:41] <rsesek> seems like it
[16:42:42] <thomasvl> from the ide it gets it right
[16:42:58] <thomasvl> but from command line, it sometimes fails to run most of the steps on the chrome target
[16:43:14] <rsesek> I wonder what the issue is; I wonder if it parallelizes dependencies or something and then doesn't realize it
[16:43:31] <yakattak> i'll wait for the next build then
[16:43:41] <thomasvl> no, it flat out doens't run some of the steps
[16:43:48] <rsesek> hm
[16:43:56] <thomasvl> minor things: copies, compiles, links
[16:44:01] <thomasvl> no deps in the loop
[16:44:12] <thomasvl> cause it runs other steps, just not all of them.
[16:44:20] <yakattak> however, for those of you on mac os x and want easy way to update: https://code.google.com/p/chromiumupdaterformac/ check that out
[16:44:22] <rsesek> did you file a radar?
[16:44:30] <thomasvl> y
[16:44:34] <rsesek> cool
[16:44:54] <thomasvl> already had a radar about it not honoring the parallel flag in the project
[16:45:03] <thomasvl> i'm starting to think they don't honor it because it's busted
[16:45:05] <rsesek> yakattak: maybe add a sanity check to make sure the .app is > 10M?
[16:45:14] <yakattak> heh, i should have
[16:45:17] <rsesek> :D
[16:45:44] <yakattak> it unzipped like 3 files and on top of that downloaded the zip really fast and i was like wtf, and i checked out the snapshots and they were in fact, exremely tiny
[16:45:52] <mnissler> satish_, rsesek: I guess I have to revert the .grd changes too, since I deleted constants... The build will break with the old code and .grd deletions...
[16:46:01] <rsesek> that's unfortunate
[16:46:14] <satish_> mnissler: ok, fingers crossed.
[16:46:16] <beinghuman> could anyone point me to a suggestion box
[16:46:18] <beinghuman> or something
[16:46:35] <beinghuman> so I can request that chrome make borderstyles like i posted look more like opera and FF... it just makes more sense from a design perspective
[16:46:42] <rsesek> beinghuman: new.crbug.com
[16:46:44] *** andybons has quit IRC
[16:46:47] <beinghuman> should I take my concerns to webkit?
[16:47:00] <rsesek> beinghuman: if Safari renders the same way, it's most likely a WebKit bug
[16:47:05] <thomasvl> yakattak: you should also check the build that snapshot came from, noting says it is good.  you need to see how it did on unittests to decided if it's crap or not
[16:47:29] <yakattak> hmm
[16:47:37] <yakattak> this is an open source project you know :P
[16:47:43] <yakattak> (talking about the script)
[16:47:49] <rsesek> that's the reason why I don't autoupdate; I like to make sure that the build I'm pulling is good
[16:47:55] *** tonikitoo has quit IRC
[16:48:20] <beinghuman> if I were to show this design glitch to webkit.. how long would it take for it to be integrated into chromium... and then integrated into chrome?
[16:48:20] *** yutak_home has joined #chromium
[16:48:20] *** ChanServ sets mode: +v yutak_home
[16:48:56] <rsesek> beinghuman: webkit's tip of tree is integrated daily (sometimes multiple times). and then a few months after that it'll be in chrome stable
[16:49:28] <beinghuman> alright
[16:50:29] <thomasvl> yakattak: this also is open source, nothing says ever spin is good. ;)
[16:50:37] *** MichealH|Laptop has joined #chromium
[16:51:04] <yakattak> thomasvl: that's not what i mean ha, my main language is objc, not python; in other words i'm not sure i could add the checking of the snapshot details and whatnot
[16:51:49] <thomasvl> beinghuman: chrome has 3 channels (dev, beta, stable), how long it takes also depends on which channel you are talking about.  months is usually the case for stable, but beta and dev can be much shorter.
[16:51:50] *** fta has quit IRC
[16:53:03] *** fta has joined #chromium
[16:54:20] <mnissler> satish_: revert is in, I'll land another whitespace eventually. Sorry for the hassle.
[16:54:53] <satish_> mnissler: no problem, since most folks are sleeping anyway
[16:55:38] *** beinghuman has quit IRC
[17:02:32] <mnissler> satish_: looks good, I guess we can reopen after we get some green blocks on the windows builders. Next build will resolve any leftovers from the grd revert.
[17:03:03] <satish_> mnissler: yes, will look for greenery. If you spot any that I miss, please ping me.
[17:03:57] <mnissler> satish_: will do. thanks for your help.
[17:07:47] *** thakis has joined #chromium
[17:07:47] *** ChanServ sets mode: +v thakis
[17:09:14] *** mrossetti has joined #chromium
[17:09:57] *** rubenbb has quit IRC
[17:10:03] *** fta_ has joined #chromium
[17:11:05] *** fta has quit IRC
[17:12:35] <satish_> mnissler: I'm planning to submit http://codereview.chromium.org/2810045/show to fix the "chrome frame tests (ie8)" builder as this test has been crashing all day. Do you prefer I wait until the build cycles green after your revert?
[17:14:02] <mnissler> satish_: no, go ahead, the tree is in exactly the state before I started the confusion (apart from the whitespace file)
[17:14:02] *** andybons has joined #chromium
[17:14:10] *** fta_ is now known as fta
[17:14:49] <satish_> mnissler: ok done, thanks.
[17:16:36] *** Venom_X has joined #chromium
[17:17:04] <trungl-bot> Tree opened by mnissler at chromium dot org: Tree is open (should cycle green)
[17:17:26] *** thakis_ has joined #chromium
[17:17:27] *** ChanServ sets mode: +v thakis_
[17:17:37] <mnissler> satish_: I need to leave in 10 minutes, are you ok with that?
[17:18:33] <satish_> sure go ahead. As you said your change is not in today, so it should only be jochen's previous grd change right?
[17:19:17] <satish_> mnissler: I'll check those if the builds continue to have issues.
[17:19:31] <satish_> mnissler: hopefully they come back to green with this cycle.
[17:19:43] <mnissler> satish_: yes. my stuff is definitely out.
[17:20:22] *** tonyg-cr has quit IRC
[17:20:56] *** thakis has quit IRC
[17:20:56] *** thakis_ is now known as thakis
[17:22:23] *** andrix has joined #chromium
[17:22:25] *** Buglouse has joined #chromium
[17:23:18] <mnissler> satish_: I'm off. Have a good weekend!
[17:23:30] *** rubenbb has joined #chromium
[17:25:21] <satish_> mnissler: thanks, have fun. ciao.
[17:30:18] <satish_> maruel: can anything be done with the 'chrome frame tests' builder? looks like a previous instance is running and files can't be deleted.
[17:30:47] *** pfeldman_ has joined #chromium
[17:32:16] *** andybons has quit IRC
[17:34:04] *** pfeldman has quit IRC
[17:34:04] *** pfeldman_ is now known as pfeldman
[17:34:28] *** fta_ has joined #chromium
[17:35:14] *** dmaclach has joined #chromium
[17:35:49] *** fta has quit IRC
[17:37:21] <satish_> pamg: maruel: thomasvl: anyone around? Vista Perf (dbg) builder has been failing repeatedly complaining lack of disk space.. needs a restart?
[17:37:38] * pamg will look.
[17:38:46] <ananta> satish: The cf builders should cycle green
[17:39:34] <pamg> I'm about to lose network connectivity, so it would be best if someone else took this on.
[17:39:53] <pamg> It's Friday evening here and IT is doing some network reconfiguration.
[17:40:27] <ananta> satish:- The chrome frame tests builder was in a wierd state with FF crashing everytime on startup. It should work fine now
[17:40:31] <pamg> I don't see disk-space issues on Vista Perf (dbg), but a locked directory and leftover cruft. It needs an svn cleanup
[17:41:09] *** bers has joined #chromium
[17:41:12] <pamg> And moving build.dead back to build so it doesn't have to do a clean checkout yet again
[17:41:17] <satish_> ananta: thanks
[17:41:50] <satish_> pamg: sorry, this may be a newbie question :) is this something I can do or should ask a buildbot trooper?
[17:42:01] <pamg> satish_: You need a trooper.
[17:42:29] <pamg> I could do it if I knew I'd have network for at least 10 more minutes, but I don't want to drop out in the middle and leave an even bigger mess.
[17:42:32] <satish_> pamg: ok, i'll keep pinging the troopers every few minutes and see who can help.
[17:42:42] *** rubenbb has quit IRC
[17:43:26] <satish_> ananta: I disabled a test which was crashing due to that FF hang, if you think that is irrelevant now please feel free to revert it.
[17:43:48] <ananta> ok
[17:45:20] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Linux Builder (dbg-shlib)" from 51535: avi at chromium dot org (:motownavi))
[17:45:23] *** AryehGregor has quit IRC
[17:45:39] <satish_> motownavi: Looking at the build break?
[17:46:21] <trungl-bot> Tree closed by satish at chromium dot org: Tree is closed (compile error -> avi at chromium dot org (:motownavi))
[17:46:43] <motownavi> agh
[17:46:44] <motownavi> i suck
[17:47:14] <motownavi> suggested parens?
[17:47:21] <motownavi> will quick fix, revert if it doesn't take
[17:48:50] *** rubenbb has joined #chromium
[17:49:23] <motownavi> fix in
[17:49:45] <satish_> ok checking
[17:50:24] <trungl-bot> Tree closed by avi at chromium dot org (:motownavi): Tree is closed (fix in)
[17:51:17] <satish_> motownavi: sorry I just reopened the tree, do you want it to remain closed until it cycles green?
[17:51:25] <trungl-bot> Tree opened by satish at chromium dot org: Tree is open (should cycle green)
[17:51:32] *** duffydack has joined #chromium
[17:51:37] <motownavi> nah, open is fine
[17:51:44] <motownavi> I'll watch the red
[17:52:23] *** RT|Chatzilla has quit IRC
[17:52:54] *** duffydack has left #chromium
[17:55:27] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Linux Perf (2)" from 51535: avi at chromium dot org (:motownavi), brettw at chromium dot org, cbentzel at chromium dot org, xiyuan at chromium dot org, yurys at chromium dot org)
[17:55:45] <agl> (just a second auto-close from the same error)
[17:55:54] <satish_> yes, opening again
[17:56:27] <trungl-bot> Tree opened by agl at chromium dot org: Tree is open (should cycle green)
[17:56:33] <satish_> nsylvian: maruel: thomasvl: leiz: anyone around? Vista Perf (dbg) builder has been failing repeatedly complaining lack of disk space.. needs a restart?
[17:56:52] *** urbanape has quit IRC
[17:58:19] <motownavi> Why does not putting parens around a && expression within || expressions give an error? :(
[17:58:51] *** thakis_ has joined #chromium
[17:58:51] *** ChanServ sets mode: +v thakis_
[17:58:55] *** thakis_ has quit IRC
[17:59:20] <satish_> motownavi: I think it is a warning, and the build treats warnings as errors.
[17:59:26] * motownavi sighs
[17:59:33] <agl> motownavi: it's a pretty dumb warning in that case, I agree.
[17:59:45] <motownavi> but it passed the buildbot!
[17:59:50] <satish_> motownavi: yes that's the bad part
[17:59:55] <satish_> perhaps you can bring it up in chromium-dev ?
[18:00:10] <motownavi> indeed I will.
[18:00:43] <agl> guess what, I just got bitten by the same thing!
[18:00:46] <agl> reverting
[18:01:09] <satish_> agl: the recent build breaks your job? :)
[18:02:12] *** thakis has quit IRC
[18:02:20] <agl> satish_: stupid, false-positive warnings which don't happen locally or on the try bots
[18:02:36] <satish_> agl: let me close the tree, too many things going in parallel now
[18:03:00] *** sberry has joined #chromium
[18:03:25] <agl> satish_: the lunch train is happening here now, but the revert is in so should all go green.
[18:03:53] <trungl-bot> Tree closed by satish at chromium dot org: Tree is closed (waiting for fixes to come in and cycle green)
[18:04:22] <thomasvl> warning probably only trips in release builds
[18:05:56] *** maikmerten has joined #chromium
[18:06:07] <thomasvl> satthere are a few win bots with gclient exceptions now, so I'd rather pass it off to one of the folks that know those bots better then i do
[18:06:55] *** dmazzoni has joined #chromium
[18:07:07] <satish_> thomasvl: I thought I saw only Vista Perf (dbg) to come up with gclient exception.. which other ones?
[18:07:19] *** [Nef] has quit IRC
[18:07:41] <thomasvl> satish_: xptests dbg 4, vista tests dbg 1
[18:07:41] *** bauerb is now known as bauerb_afk
[18:07:55] <trungl-bot> Tree opened by satish at chromium dot org: Tree is open (should cycle green)
[18:08:10] <thomasvl> satish_: see the purple in the waterfall summary line
[18:08:29] <satish_> thomasvl: yes. anyone you have in mind to look at this?
[18:08:42] <thomasvl> satish_: nsylvain or maruel
[18:09:08] <satish_> ok
[18:10:56] *** henriknj has joined #chromium
[18:12:17] *** AryehGregor has joined #chromium
[18:13:45] <satish_> vandebo: kinuko: you there?
[18:18:15] *** mnissler has quit IRC
[18:19:00] *** kenneth_reitz has quit IRC
[18:19:30] *** fta__ has joined #chromium
[18:20:51] *** fta_ has quit IRC
[18:20:57] *** pfeldman has quit IRC
[18:21:50] <vandebo> morning
[18:24:33] <nshkrob>
[18:25:03] *** jochen__ has quit IRC
[18:25:12] *** pamg has quit IRC
[18:25:14] *** danno_ has quit IRC
[18:25:36] *** bauerb_afk has quit IRC
[18:27:58] <satish_> vandebo: g'morning.
[18:28:08] <satish_> vandebo: ready for some sheriffing? :)
[18:28:24] <vandebo> lots of red :-(
[18:29:09] <satish_> vandebo: some windows builders seem to be having gclient/svn issues, need to find a trooper and fix the build bots. Unfortunately the only suggested names are maruel and nsylvain, both are unreachable
[18:29:15] <satish_> vandebo: so need to keep pinging them.
[18:29:31] <satish_> vandebo: the rest should cycle into green as the fixes were in.
[18:29:46] <vandebo> you don't expect the purple bots to cycle green?
[18:30:03] <satish_> vandebo: no, those are the ones with gclient issues and need maruel/nsylvain to look
[18:30:15] <satish_> vandebo: it was just 1 bot an hour ago, and now 4 bots :(
[18:30:35] <satish_> maruel: nsylvain: you there?
[18:31:08] <vandebo> satish_: fyi, the troopers are all listed on the sheriffing page
[18:31:27] <satish_> vandebo: yes, I asked pamg who couldn't do anything as she was out of network.
[18:31:40] <satish_> vandebo: and thomasvl said he is not that familiar with windows bots
[18:31:47] <satish_> vandebo: he suggested maruel or nsylvain
[18:33:20] <trungl-bot> Tree closed by vandebo at chromium dot org: Tree is closed (sick windows bots, need a trooper)
[18:34:09] *** trungl_ has joined #chromium
[18:34:09] *** ChanServ sets mode: +v trungl_
[18:34:58] <satish_> vandebo: cool, more linux bots turning green
[18:35:11] <satish_> maruel is OOO, says " try nsylvain as Im at the zoo"
[18:35:18] <satish_> vandebo: no idea what that means :D
[18:35:28] *** trungl_ has quit IRC
[18:35:51] *** pathorn has joined #chromium
[18:35:51] *** ChanServ sets mode: +v pathorn
[18:36:04] *** hinoka|away has joined #chromium
[18:36:04] *** ChanServ sets mode: +v hinoka|away
[18:36:54] *** bers has quit IRC
[18:37:13] <satish_> vandebo: looks like the non-windows issues are going down. I'm leaving in 10 minutes, if you are ok?
[18:37:16] <vandebo> I suspect he's at the zoo
[18:37:22] <vandebo> yup
[18:37:44] *** trungl has quit IRC
[18:37:46] <satish_> the real zoo? :) I thought he meant it as a code word for the irc channel :)
[18:38:52] *** urbanape has joined #chromium
[18:39:16] *** davidben has joined #chromium
[18:39:16] *** ChanServ sets mode: +v davidben
[18:39:38] *** satish_ has quit IRC
[18:49:58] *** MikeSmithW3C has quit IRC
[18:50:14] *** trungl has joined #chromium
[18:50:14] *** ChanServ sets mode: +v trungl
[18:51:22] *** bent-mozilla has joined #chromium
[18:53:45] <thomasvl> vandebo: looks like two are recovering on their own
[18:54:21] <vandebo> is there something wrong on the backend.  Xp tests4 recovered and then errored again
[18:54:42] *** MikeSmithW3C has joined #chromium
[18:56:14] *** urbanape has quit IRC
[18:56:35] <thomasvl> bounce one of the bots to just try to clear this
[18:57:23] <vandebo> hmm?
[18:58:03] <thomasvl> vandebo: vista perf dbg
[18:58:08] <thomasvl> it's been failing the loggest
[18:58:24] <thomasvl> xp tests4 made it through and is running tests at the moment
[18:58:30] <thomasvl> if it happens again to it, i guess we bounce it again
[18:58:45] <rohitrao> what does NOTREACHED() do in a release build?
[18:59:00] <rohitrao> does it still log the error message, or does it get compiled away completely?
[18:59:45] <vandebo> thomasvl: bounce=reboot?  Did you? I don't have access (as far as I know)
[19:00:08] *** fta__ is now known as fta
[19:00:08] <thomasvl> vandebo: yes, i did
[19:00:43] <nsylvain> satorux: pong
[19:00:45] *** bweinstein has joined #chromium
[19:00:49] *** star-affinity has quit IRC
[19:01:55] <thomasvl> nsylvain: there are a few bots flirting with gclient errors
[19:01:59] <nsylvain> looking
[19:02:03] <thomasvl> i just bounced one hoping it would help
[19:02:15] <thomasvl> the others seem to be correcting, but they had trouble a few runs
[19:02:50] <rohitrao> does anyone else in here run debug mac builds?
[19:03:03] <aboodman> does anyone know how to update the symbols in WebCore.order?
[19:03:14] <rohitrao> mine are all corrupting the text database shortly after launch
[19:03:30] <mirandac> rohitrao: known problem; see http://code.google.com/p/chromium/issues/detail?id=47059
[19:04:00] <mirandac> rohitrao: I just changed fts3 back to fts2 to fix for now
[19:04:50] *** fta_ has joined #chromium
[19:05:09] *** GeekShadow has joined #chromium
[19:05:34] *** fta has quit IRC
[19:05:56] <rohitrao> mirandac: ah, so I'm not going crazy :) how do I change back to fts2?
[19:06:16] * rohitrao greps for fts3
[19:06:17] *** csilv has joined #chromium
[19:06:17] *** ChanServ sets mode: +v csilv
[19:06:17] <mirandac> in text_database.cc, there's a single line -- a 1-char change :-)
[19:06:44] <rohitrao> hehe, thanks
[19:06:56] *** Clockmens has joined #chromium
[19:07:05] <mirandac> np, I was going crazy, too :-)
[19:07:06] <Clockmens> Hi
[19:07:17] <trungl> stupid 2-bit changes
[19:07:24] <Clockmens> I got a huge pain inn
[19:07:44] <Clockmens> Chromium OS' file manager is blank for me?
[19:07:53] <Clockmens> I go chrome://filemanager and its blank page
[19:07:57] <trungl> see channel topic
[19:08:41] <Clockmens> :|
[19:09:19] *** fta_ has quit IRC
[19:09:38] *** fta has joined #chromium
[19:09:55] <Clockmens> How do I list connected drives in terminal?
[19:10:36] <leiz> see /topic, you're in the wrong channel
[19:10:45] <Clockmens> It;'s a general linux question
[19:10:58] <trungl> then it's an even more wrong channel
[19:11:02] <leiz> and this isn't #linux-techsupport
[19:11:06] <Clockmens> Tits
[19:11:18] <Clockmens> So what is this
[19:11:27] <Clockmens> #tell-people-to-go-away?
[19:11:35] *** bent-mozilla has quit IRC
[19:11:39] *** ChanServ sets mode: +o trungl
[19:11:45] *** Clockmens was kicked by trungl (Clockmens)
[19:11:47] <nsylvain> thomasvl: code.google.com is just too slow. Try "svn ls http://rlz.googlecode.com/svn"
[19:11:55] *** yakattak has left #chromium
[19:12:00] *** ChanServ sets mode: +o vandebo
[19:12:01] *** ChanServ sets mode: -o trungl
[19:12:02] <nsylvain> It takes like 10-30 secs just to do "ls"
[19:12:05] *** MichealH has quit IRC
[19:12:14] <thomasvl> nsylvain: can we stick in a version number in deps to avoid the problem?
[19:12:20] <thomasvl> s/version/revision/
[19:12:22] <nsylvain> we have a version number
[19:12:25] <trungl> "googlecode svn slow, film at 11"
[19:12:26] <nsylvain> but now that we do clobber...
[19:12:32] <thomasvl> ugh
[19:12:44] *** vandebo sets mode: -o vandebo
[19:12:57] <nsylvain> i guess native client got updated? forcing us to update from code.google.com, and it made the bots fail. And now we need to fetch everything from code.google.com
[19:13:14] *** MichealH|Laptop has quit IRC
[19:13:25] *** MichealH has joined #chromium
[19:13:32] <nsylvain> googleurl is the one we updated
[19:13:57] *** tittiatcoke has joined #chromium
[19:14:59] <nsylvain> i can't replicate the slowness now
[19:15:05] <nsylvain> at least on my machine
[19:15:09] <kinuko> vandebo: (morning, I'm joining)
[19:15:33] <trungl> maybe their svn has recovered
[19:16:46] <leiz> i couple of win bots are really low on space, I'm clearing them out
[19:18:09] <rohitrao> thomasvl: you mentioned a way yesterday to only run unit_tests on a trybot?
[19:18:25] <vandebo> rohitrao: -t:unit_tests
[19:18:35] <rohitrao> vandebo: thanks!
[19:18:58] <trungl> sweet
[19:19:00] <vandebo> you can even run specific tests -t:test_binary:gtest_filter
[19:19:11] <vandebo> repeat for additional test binaries
[19:19:22] <vandebo> (though only the ones the bot normally runs)
[19:19:52] <thomasvl> rohitrao: there's a tips and tricks page on dev.chromium.org that talks about it
[19:20:46] <vandebo> should we let a commit through and see if the code.google.com is better now?
[19:21:11] <nsylvain> thomasvl: chromium mac has been really sick lately. Any idea?
[19:21:24] <nsylvain> browser tests has been hung for 4 hours
[19:21:45] *** maikmerten_ has joined #chromium
[19:21:45] <thomasvl> nsylvain: i just looked and didn't see anything on console, let me check again
[19:22:19] <nsylvain> thomasvl: i guess it's '[Mac] Use Core Graphics to draw the close button used within tabs, infobars, the download shelf and other places. Delete the original pdfs. This is in line with the new top chrome refresh.'
[19:22:24] <nsylvain> from andybons
[19:22:52] <nsylvain> actually
[19:22:58] <thomasvl> nsylvain: is that getting removed?
[19:23:00] <nsylvain> It might be 'Catch OOMs in purgeable memory.' from Avi
[19:23:02] <nsylvain> no idea
[19:23:05] *** maikmerten_ has quit IRC
[19:23:23] <thomasvl> nsylvain: i did bounce it this morning to help general slowdown
[19:23:39] <nsylvain> All the tests are timing out, since yeterday morning
[19:23:57] <vandebo> chomium mac has been bounced three times in the last 24 hours, def something wrong
[19:24:11] <nsylvain> there is that error :
[19:24:11] <nsylvain> LaunchApp: failed to execvp:
[19:24:12] <nsylvain> 2010-07-01 10:30:24.409 browser_tests[87268:20b] Could not find image named 'close_bar'.
[19:24:12] <nsylvain> 2010-07-01 10:30:24.497 browser_tests[87268:20b] Could not find image named 'close_bar_p'.
[19:24:28] *** maikmerten has quit IRC
[19:24:38] <leiz> i even rebooted it last night
[19:26:12] <nsylvain> anyone knows andybons?
[19:26:26] <thomasvl> nsylvain: dump those in a bug, i'll fix them later (i've fixed them once, they are non critical)
[19:27:09] <nsylvain> thomasvl: oh ok. so i guess that's not why the tests are not working
[19:27:18] <thomasvl> nsylvain: i know
[19:27:26] <thomasvl> it's a bad build, the zip is tiny
[19:27:30] <thomasvl> so there is no real app to launch
[19:27:36] *** trungl has quit IRC
[19:27:39] <thomasvl> so we must be hitting a timeout on app launch in every test
[19:28:20] <thomasvl> nsylvain: just stop the build or bounce it so it done the next build (with the fix for the bad app bundles) and it will be happy again.
[19:28:34] <thomasvl> s/done/does/
[19:28:38] <nsylvain> ok
[19:30:09] <kinuko> vandebo: googlecode looks ok now.  I think it's ok to open
[19:30:14] *** cbentzel has quit IRC
[19:30:50] *** fishd has quit IRC
[19:31:19] <vandebo> what bout the mac bots?
[19:32:13] <kinuko> new build has been started on chromium mac
[19:32:44] <vandebo> anyone have a commit - I'd rather see the win bots finish a sync before opening
[19:34:06] *** evmar_afk is now known as evmar
[19:34:07] <evmar> me me
[19:34:09] <evmar> linux-only
[19:34:12] <evmar> ready to go
[19:34:13] <vandebo> sounds good
[19:35:08] <evmar> done
[19:36:04] <nsylvain> once Chromium Builder is done building, i think we can reopen the tree. I believe this was the main bot causing really problems
[19:36:08] <trungl-bot> Tree closed by vandebo at chromium dot org: Tree is closed (win bot troubles - waiting for a good sync)
[19:36:25] <nsylvain> I think we can deal with nacl ans chrome frame losing coverage while they sync
[19:38:24] <kinuko> sounds good.  let's see
[19:39:03] *** dave_levin has joined #chromium
[19:41:07] *** eroman has joined #chromium
[19:41:07] *** ChanServ sets mode: +v eroman
[19:42:03] *** shreyas_ has joined #chromium
[19:42:14] *** maikmerten has joined #chromium
[19:42:57] *** mnissler has joined #chromium
[19:42:57] *** ChanServ sets mode: +v mnissler
[19:43:10] *** shreyas_ has quit IRC
[19:43:28] *** jochen__ has joined #chromium
[19:43:28] *** ChanServ sets mode: +v jochen__
[19:45:13] *** johnny_g has joined #chromium
[19:45:13] *** ChanServ sets mode: +v johnny_g
[19:46:28] *** fta has quit IRC
[19:47:20] <vandebo> why hasn't the chomium builder started yet?
[19:48:00] <kinuko> it has started but not shown on waterfall
[19:48:39] *** tonyg-cr has joined #chromium
[19:48:39] *** ChanServ sets mode: +v tonyg-cr
[19:49:56] *** andrix has quit IRC
[19:50:03] <vandebo> looks like it did the gclient step just fine...
[19:50:03] *** bitreader has joined #chromium
[19:50:33] <vandebo> open?
[19:51:17] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[19:52:49] *** tonikitoo has joined #chromium
[19:53:12] *** rafaelw has joined #chromium
[19:53:12] *** ChanServ sets mode: +v rafaelw
[19:58:25] *** thakis has joined #chromium
[19:58:25] *** ChanServ sets mode: +v thakis
[20:00:23] *** mrossetti has quit IRC
[20:01:56] *** Precea[BNC] is now known as Precea
[20:02:07] *** Venom_X has quit IRC
[20:04:02] *** thomasvl has quit IRC
[20:04:31] *** dpranke has joined #chromium
[20:04:31] *** ChanServ sets mode: +v dpranke
[20:04:56] *** dglazkov has joined #chromium
[20:04:56] *** ChanServ sets mode: +v dglazkov
[20:05:04] <dglazkov> good morning, Chromium!
[20:05:05] <trungl-bot> dglazkov: Good morning!
[20:06:28] *** zloidemon has joined #chromium
[20:09:15] *** jochen__ has quit IRC
[20:10:20] *** away01 is now known as temp01
[20:10:53] <thakis> dglazkov!
[20:12:46] *** Vanderhuge has joined #chromium
[20:13:41] *** victorw has joined #chromium
[20:13:41] *** ChanServ sets mode: +v victorw
[20:15:22] *** thakis has quit IRC
[20:21:04] *** jochen__ has joined #chromium
[20:21:04] *** ChanServ sets mode: +v jochen__
[20:23:50] *** henriknj has quit IRC
[20:23:58] *** jianli has quit IRC
[20:24:36] *** cbentzel has joined #chromium
[20:25:57] <dantasse> hello chromers! vandebo, satish, kinuko: i've got a grd change to land, when is convenient for you?
[20:26:31] <vandebo> how about during the lunch (pst) hour?
[20:26:49] <dantasse> actually, i've got to meet someone for lunch
[20:27:06] <dantasse> (at noon) could do it right before lunch maybe?
[20:27:07] <vandebo> then it's not a good time to start such a change?
[20:27:25] <dantasse> ok, after lunch then
[20:27:29] <vandebo> you were planning to stick around to make sure your change didn't have problems, weren't you?
[20:27:34] *** General13372 has joined #chromium
[20:28:00] <dantasse> oh yeah, but i figured 1/2 hr would be enough
[20:28:36] *** thakis has joined #chromium
[20:28:36] *** ChanServ sets mode: +v thakis
[20:28:39] <vandebo> grd land procedure takes about an hour
[20:29:05] <dantasse> oh ok... yeah, let's do it after lunch then
[20:30:07] <thakis> vandebo / dantasse: what about landing it over lunch when there's not many checkins, but keeping the tree open?
[20:30:13] <thakis> (if it's a grd-only change)
[20:30:44] *** General1337 has quit IRC
[20:30:45] <rsesek> if you're just adding strings, you could just land the grd part
[20:30:46] <dantasse> it's a grd plus a little extra code
[20:30:48] *** andybons has joined #chromium
[20:30:53] <dantasse> i could split it up though
[20:31:06] <dantasse> and just land the string over lunch
[20:31:16] <rsesek> rohitrao: ping
[20:31:22] <kinuko> that sounds good to me.
[20:31:37] <dantasse> ok, thanks, i'll let you know when it's ready
[20:32:19] <kinuko> ok.  we may stop you if we have too many pending builds though
[20:32:28] <dantasse> ok
[20:34:10] *** fantasticulous has joined #chromium
[20:34:18] <vandebo> hmmm - I wonder if it would make sense to put grd files in a different repo - then we could periodically (over lunch, at night), roll the grd rev
[20:34:39] <rsesek> what if you're deleting a string?
[20:34:46] <rsesek> then it wouldn't compile
[20:35:17] *** mrossetti has joined #chromium
[20:35:22] <vandebo> oh, I have no idea what grd files do or how they interact with the build ;-)
[20:35:27] <rohitrao> rsesek: hello
[20:35:49] <rsesek> rohitrao: re http://code.google.com/p/chromium/issues/detail?id=35374 ? is this bug more about the browser crash or the dead tabs
[20:36:26] <rsesek> rohitrao: I have an instance of chromium where I always get dead tabs when trying to restore (any profile, release or debug); but they don't crash the browser when you close them
[20:36:44] <rohitrao> rsesek: I've only ever seen the crash
[20:36:48] <rsesek> :(
[20:36:49] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "compile" on "Chromium Builder (dbg)" from 51540: mpcomplete at chromium dot org)
[20:36:58] <rohitrao> although it doesn't actually appear to be a crash, cause gdb stayed happt
[20:37:10] <rsesek> okay this sounds like a different issue
[20:37:20] <rsesek> something is getting wedged and not causing a load
[20:37:20] *** MichealH has quit IRC
[20:38:00] *** andybons has quit IRC
[20:38:23] <rohitrao> rsesek: do all the tabs wedge? or just a few?
[20:38:42] <rsesek> rohitrao: anything you restore will be wedged; reloading the tab will cause a full load; all new tabs work fine
[20:38:50] <rsesek> only happens on my local build, not waterfall builds
[20:39:52] <rohitrao> that sounds different then
[20:39:52] <rohitrao> this issue has always been "most restore fine, one or two spin forever, closing them causes crash"
[20:40:34] <vandebo> dantasse: it's going to take past noon to clear up this bot problem, sorry
[20:40:41] <dantasse> vandebo: ok, np
[20:41:07] <vandebo> familiar to anyone? cl : Command line error D8027 : cannot execute 'C:\Program Files (x86)\Microsoft Visual Studio 8\VC\bin\c2.dll'
[20:42:12] <vandebo> not sure if it'll cycle clean
[20:42:27] <vandebo> did someone change something on that bot?
[20:42:42] <rohitrao> I've seen that on trybots before, I think
[20:43:12] <leiz> it's just VC barfing I think
[20:43:18] *** thakis has quit IRC
[20:43:23] <leiz> usually goes away on the next run
[20:43:43] <vandebo> ok thanks
[20:43:54] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[20:56:03] *** zaspire has quit IRC
[20:57:03] *** [Nef] has joined #chromium
[20:58:08] *** zaspire has joined #chromium
[20:59:38] <kinuko> builders look ok now
[21:00:02] <kinuko> guess dantasse is already out for lunch though
[21:00:07] <dantasse> nah, i'm here
[21:00:11] <dantasse> friend is running late
[21:00:14] <kinuko> oh you're here
[21:00:17] <kinuko> there
[21:00:36] <kinuko> I think it's ready to land grd change
[21:00:44] <kinuko> vandebo: how do you think?
[21:00:44] <dantasse> ok, thanks
[21:00:54] <kinuko> dantasse: wait a min
[21:00:58] <dantasse> sure
[21:01:04] <trungl-bot> Unknown tree status set by vandebo at chromium dot org: Tree is close (grd change)
[21:01:10] <kinuko> vandebo: thx
[21:01:21] <kinuko> it'll be probably ok not closing the tree
[21:01:23] <kinuko> but anyway
[21:01:34] <rsesek> woo chromium mac can't link again!
[21:02:10] <trungl-bot> Tree closed by vandebo at chromium dot org: Tree is closed (grd change)
[21:02:12] <vandebo> seems that the tree will be mostly closed over the us holiday than
[21:02:19] *** zaspire has quit IRC
[21:02:48] <dantasse> kinuko: ready to go?
[21:03:04] <vandebo> yup
[21:03:20] <kinuko> a new change has just submitted but it's small so hope it'll be ok
[21:03:35] <kinuko> ah now it's ready
[21:03:51] <kinuko> dantasse: go ahead
[21:04:39] <vandebo> darn - that'll add 20min to the cycle
[21:05:04] *** fantasticulous has quit IRC
[21:05:21] <kinuko> sigh.
[21:05:36] <dantasse> kinuko: ok, done
[21:05:46] <dantasse> thanks!
[21:06:39] *** zaspire has joined #chromium
[21:08:10] *** tonyg-cr has quit IRC
[21:08:22] *** temp01 is now known as away01
[21:14:02] <evmar> mac folks, i have code like this
[21:14:03] <evmar>     result->assign(UTF8ToWide([[save_panel filename] UTF8String]));
[21:14:18] <evmar> result is a wstring.   what is the proper conversion when result is now a FilePath?
[21:14:23] <evmar> just drop the UTF8ToWide?
[21:14:40] <rsesek> UTF8String returns a const char*
[21:15:51] <evmar> what does [foo filename] return?
[21:15:56] <rsesek> an NSString*
[21:16:02] <evmar> i see
[21:16:09] <evmar> ok, my change is probably ok then
[21:16:18] <rsesek> I think so (I don't know anything about FilePaths)
[21:17:01] <bitreader> The Internet is serious business.
[21:24:52] <kinuko> vandebo: I have a tiny cl, it'd be a good time to land it to trigger another build?
[21:26:09] <vandebo> most of the windows bots will fail this round - is it a mac or linux only change?
[21:26:54] <kinuko> no it's not.
[21:27:46] *** Alystair has joined #chromium
[21:28:02] <kinuko> well- yes it will affect only mac and linux
[21:28:16] <kinuko> just adding FLAKY_
[21:28:30] <vandebo> oh, sure, use that as the whitespace change
[21:28:35] <kinuko> ok
[21:28:45] <thakis_afk> evmar: yu want to call fileSystemRepresentation on the NSString
[21:29:03] <thakis_afk> evmar: and pass that to FilePath's constructor directly
[21:29:04] *** thakis_afk is now known as thakis
[21:30:46] <thakis> evmar:
[21:30:48] <thakis> evmar: http://developer.apple.com/mac/library/documentation/Cocoa/Reference/Foundation/Classes/NSString_Class/Reference/NSString.html#//apple_ref/occ/instm/NSString/fileSystemRepresentation
[21:31:09] <vandebo> mac debug builder is no longer linking - do we carry on with only release coverage, hack in another work around, close the tree?
[21:31:27] <thakis> vandebo: are there workarounds left?
[21:31:37] <rsesek> you can try rebooting the builder, but no guarantee
[21:31:44] <thakis> keeping the tree closed for 2+ weeks doesn't seem realistic
[21:31:50] <rsesek> or see if it works itself out on this run
[21:31:54] <rsesek> sometimes it takes a couple of trys
[21:32:39] *** tonyg-cr has joined #chromium
[21:32:39] *** ChanServ sets mode: +v tonyg-cr
[21:32:51] <pathorn> just a thought, but couldn't you build without debug symbols but with _DEBUG enabled
[21:32:55] <pathorn> so you still get DCHECK's
[21:33:04] <pathorn> it then only becomes a problem if you want a stacktrace
[21:35:39] <kinuko> see one more build and try rebooting? (mac builder)
[21:37:02] <vandebo> a reboot won't hurt, though no reason to wait
[21:37:23] <rsesek> yea you're going to have to wait for windows to cycle, so I'd reboot now
[21:37:50] <evmar> thakis: thanks. currently deep in an epic "make all" to verify i didn't break anything, will change it again once that's done
[21:38:41] <kinuko> it's just failed 3 times in a row
[21:39:11] <kinuko> no 4 times
[21:39:16] <kinuko> reboot sounds good
[21:43:17] *** away01 is now known as temp01
[21:43:56] <kinuko> can anyone reboot it?
[21:43:58] <rsesek> kinuko: both tvl and mmentovai are out, so you'll need to ping a different trooper
[21:44:21] <kinuko> rsesek: I see.
[21:44:23] <rsesek> I'd try nsylvain or maruel
[21:44:23] <vandebo> thestig: ping
[21:44:36] <vandebo> maruel is out
[21:44:54] <nsylvain> hello
[21:45:00] <nsylvain> What can i do?
[21:45:08] <vandebo> mac dbg builder reboot
[21:45:13] <kinuko> great
[21:45:15] <nsylvain> ok, will do
[21:46:05] <nsylvain> rebooting now
[21:46:12] <leiz> vandebo: pong?
[21:46:25] <vandebo> nevermind, nsylvain showed up
[21:46:31] <kinuko> nsylvain: thanks
[21:47:09] <vandebo> looks like we're good to reopen now
[21:47:19] <kinuko> yeah I was about to say that
[21:47:45] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[21:47:55] *** temp01 has quit IRC
[21:47:55] <evmar> here goes try #4 on updatnig gyp
[21:48:05] <vandebo> kinuko: forgotten to cc you, the mac chromium release bot has been failing on a test, thakis is testing a fix
[21:48:19] *** temp01 has joined #chromium
[21:48:39] <kinuko> vandebo: I see. the problamtic bot...
[21:49:05] <vandebo> yea, the problems it was having masked this failure until recent builds
[21:49:20] *** chrisccoulson has quit IRC
[21:49:44] <kinuko> good catch.  it was hard to see the real issue
[21:51:13] <thakis> pkasting saw it
[21:51:54] *** Beetny has joined #chromium
[21:56:31] *** awolfson has quit IRC
[21:56:50] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "unit_tests" on "Vista Tests (dbg)(1)" from 51548: dantasse at chromium dot org)
[21:57:23] *** leavittx_ has quit IRC
[21:57:52] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[21:59:06] *** dr_win has joined #chromium
[22:00:14] *** adnap has joined #chromium
[22:00:39] <eglaysher> I'm resubmitting a patch that bork linux heapcheck on the bot, but which passes heapcheck locally and valgrind everywhere. It has a speculative fix, but if heapcheck goes red, it is me.
[22:00:52] <vandebo> thanks
[22:01:04] <adnap> Hey, is there a key I can old when I press enter in the search bar to do an "I'm feeling lucky" search (in Google Chrome)?
[22:01:08] <adnap> *hold
[22:02:11] *** chrisccoulson has joined #chromium
[22:03:34] <rsesek> vandebo, kinuko: still fails to link unit_tests
[22:03:44] <kinuko> grr
[22:04:10] <vandebo> not surprising
[22:04:10] <rsesek> if I had to guess, it's the libvpx roll that pushed us over the edge
[22:05:17] <vandebo> what about omitting debug symbols from libwebkit ?
[22:05:44] *** evmar is now known as evmar_afk
[22:05:52] <rsesek> possible. the real answer is 10.6
[22:05:57] <evmar_afk> be back in 20, home my gyp change sticks.
[22:06:01] <vandebo> yes, but as a stop gap
[22:06:05] <rsesek> but that's probably not going to happen at least until next week
[22:06:37] <adnap> Is this the an appropriate place for the question I asked (concerning Google Chrome)?
[22:06:50] <rsesek> see topic
[22:07:00] *** awolfson has joined #chromium
[22:07:20] <adnap> rsesek: My bad!
[22:07:33] *** adnap has left #chromium
[22:09:06] *** michaeln has joined #chromium
[22:09:06] *** ChanServ sets mode: +v michaeln
[22:11:57] <dcheng> Is crashing when creating an application shortcut a known issue on dev?
[22:12:30] <rsesek> yes
[22:13:34] <leiz> I thought we fixed that?
[22:13:46] <rsesek> googlechromereleases.blogspot.com
[22:15:50] *** leavittx_ has joined #chromium
[22:16:24] <leiz> ah, it's fixed on trunk, waiting for the next dev channel release
[22:17:16] <rsesek> seems like it
[22:17:42] *** rsesek has quit IRC
[22:17:51] <thakis> vandebo: my patch seems to fix a unit test crash after a clobber build
[22:18:14] <vandebo> thakis: great
[22:18:31] <thakis> vandebo: lg?
[22:19:33] <vandebo> yes, will
[22:22:59] *** dantasse has quit IRC
[22:25:03] <vandebo> Are the 10.5 and 10.6 binarier different? Why not change the 10.5 tests to pull from the 10.6 builder?
[22:25:22] <vandebo> binaries even
[22:25:42] <vandebo> er, right, that's not a 10.6 builder, it's a releas builder
[22:25:44] <vandebo> nevermind
[22:26:04] *** Gina has joined #chromium
[22:28:29] <Gina> Has anyone tried adding their own tests to pagecycler?  I am following the test data directory structure here http://www.chromium.org/developers/page-cyclers  but it is not cycling through the pages (just loading and staying on my first page).  If I use the index.html file from acid3, it works fine, so I think there is some issue connecting my index.html files with the head.js.  Does anyone have any experience with this?
[22:28:55] *** MikeSmithW3C has quit IRC
[22:30:51] *** MikeSmithW3C has joined #chromium
[22:31:01] <sberry> What possible use could this pattern be put to const char *_pattern = "*.anim[1-9j]";??
[22:31:25] *** Dataforce has joined #chromium
[22:32:14] *** Dataforce` has quit IRC
[22:33:08] *** Precea is now known as Precea[BNC]
[22:34:28] *** Zxcvb_ has joined #chromium
[22:35:19] *** erikkay has joined #chromium
[22:35:20] *** ChanServ sets mode: +v erikkay
[22:39:26] *** Zxcvb_ has quit IRC
[22:39:31] *** jschuh has joined #chromium
[22:43:49] *** shepazu has joined #chromium
[22:44:10] *** andybons has joined #chromium
[22:46:51] *** rohitrao has quit IRC
[22:52:21] *** maikmerten has quit IRC
[23:00:12] *** duffydack has joined #chromium
[23:03:34] <jochen__> how can i upload a patch to codereview that includes files in third_party/ ?
[23:04:26] <vandebo> mpcomplete: do you think 51540 is related to this mac valgrind failure? http://build.chromium.org/buildbot/memory/builders/Chromium%20Mac%20UI%20(valgrind)(2)/builds/3479/steps/memory%20test:%20ui/logs/stdio
[23:04:28] *** nexex has quit IRC
[23:04:40] *** trungl has joined #chromium
[23:04:41] *** ChanServ sets mode: +v trungl
[23:04:51] <mpcomplete> looking
[23:05:45] <andybons> trungl: ping
[23:05:54] *** nexex has joined #chromium
[23:05:55] <trungl> andybons: gnip
[23:05:57] <thakis> i have a somewhat reproducible chrome crash
[23:06:01] <pathorn> Gina: each of the test pages include a bit of javascript at the top to do the cycling
[23:06:08] <thakis> only on dev channel tho
[23:06:09] <andybons> trungl: see bug i cc'd you on?
[23:06:14] <thakis> how do i find it on crash/?
[23:07:20] <mpcomplete> vandebo: doesn't look like it to me. i didn't touch any of the files in those callstacks
[23:07:51] <trungl> thakis: were you planning on doing something about measuring memory consumption/sharing (of our own processes) on Mac? (or did you already do something?)
[23:07:56] <dpranke> anybody familiar with the mac gyp implementation around? I'm wondering if it's possible to compile on the mac using make instead of xcode, and if not, if there is a reason it wouldn't be possible to make it work?
[23:08:37] <thakis> trungl: i'm not planning on doing it. i did create the infrastructure necessary for it (task_ts for all child processes available in browser process)
[23:08:51] <motownavi> dpranke: it's theoretically possible.
[23:08:54] <jochen__> there are a few special rules in gyp files for xcode stuff that only the xcode generator pays attention to, dpranke
[23:08:58] <trungl> hrm
[23:09:16] <motownavi> the big question is bundling and anything else that xcode does special
[23:09:16] *** dantasse has joined #chromium
[23:09:18] * trungl adds thakis to the cc list just for fun.
[23:09:27] *** johnny_g has quit IRC
[23:09:28] <thakis> trungl: i'm on it already i thought
[23:09:28] <dpranke> jochen__ : presumably those rules can be skipped?
[23:09:34] *** johnny_g has joined #chromium
[23:09:46] <jochen__> dunno, i just realized that those rules exists while reading some gyp files
[23:09:58] <trungl> thakis: of another "bug"
[23:09:59] <thakis> dpranke: akalin can tell you why this is hard
[23:10:12] <thakis> dpranke: he tried
[23:10:12] *** tittiatcoke has quit IRC
[23:10:21] <dpranke> thakis: ah. interesting.
[23:11:29] <jochen__> i tried to port the xcode fix for grd changes to msvs btw
[23:11:43] <jochen__> test-compiling right now
[23:11:51] <thakis> jochen__: !!
[23:14:15] <jochen__> i'd upload it to codereview, if I knew how...
[23:14:19] <thakis> my dump question seems to be answered at http://www.chromium.org/for-testers/bug-reporting-guidelines/hanging-tabs
[23:14:23] <thakis> jochen__: upload.py
[23:14:32] <jochen__> but what args?
[23:14:37] <thakis> none
[23:14:49] <thakis> assuming you're in an svn checkout
[23:16:36] <jochen__> http://codereview.chromium.org/2876041
[23:19:30] *** tittiatcoke has joined #chromium
[23:20:56] *** tittiatcoke has quit IRC
[23:22:30] <vandebo> codf63 needs to be rebooted
[23:23:11] *** andybons has quit IRC
[23:23:19] <vandebo> nsylvain: ^^
[23:23:40] *** duffydack has left #chromium
[23:25:12] <vandebo> thestig: ? ^^
[23:26:40] *** ROBOd has quit IRC
[23:27:04] *** Venom_X has joined #chromium
[23:28:47] <thakis> it it possible to get the number of pending tasks on a message loop in a test?
[23:29:07] <vandebo> not that I saw.. was looking for something like that recently
[23:29:36] <thakis> sadness
[23:29:51] *** Alystair has quit IRC
[23:30:04] *** joecool has joined #chromium
[23:31:08] <vandebo> anyone know what's up with pagecycler more js on vista perf ?  It keeps timing out
[23:31:25] <joecool> ah good, running chromium 5.0.375.86, on linux, google requests getting hijacked by results5.google.com.  Is this is a known bug?  I'm used to this sort of thing with windows
[23:32:02] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "net_unittests" on "Modules XP (dbg)" from 51567: rdsmith at google dot com)
[23:34:03] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open
[23:35:16] <leiz> vandebo: I won't see that highlighted in my irc client unless you actually type in my nick:
[23:35:32] <vandebo> indeed
[23:36:24] <leiz> reboot in progress
[23:36:27] <vandebo> thanks
[23:37:53] *** GeekShad0w has joined #chromium
[23:39:05] <kinuko> will module xp test crash fix itself?  I'm not quite sure why it crashed.
[23:39:26] <vandebo> I suspect flake.  If it fixes itself, I'll disable the test
[23:39:46] <kinuko> vandebo: I see. thx
[23:40:07] *** GeekShadow has quit IRC
[23:41:53] <vandebo> somewhere between 51505 and 51538 morejs perf test on vista started crashing
[23:42:18] *** tonyg-cr has quit IRC
[23:45:14] *** Zaba has quit IRC
[23:48:30] *** tonyg-cr has joined #chromium
[23:48:30] *** ChanServ sets mode: +v tonyg-cr
[23:48:30] <kinuko> sigh it's been suffering gclient exceptions before the crash exposes
[23:48:55] <vandebo> right, 51505 is the last good run, and then it comes out of bot problems with the crash
[23:50:09] <vandebo> http://build.chromium.org/buildbot/perf/dashboard/ui/changelog.html?url=/trunk/src&range=51505:51538&mode=html
[23:50:13] <thakis> bleh, unit_tests now fails to link locally for me
[23:50:15] <thakis> (svg disabled)
[23:50:37] <vandebo> thakis: time to upgrade...
[23:51:22] *** hrna has quit IRC
[23:53:09] <leiz> vandebo: btw, you can locally edit ~/.gyp_includes (or whatever that file is) to add gyp options to disable svg and what not
[23:53:25] <leiz> that way you don't need a master restart
[23:54:00] <vandebo> you mean you could, since you can log into the bots?
[23:54:00] *** Zaba has joined #chromium
[23:54:30] *** GeekShadow has joined #chromium
[23:55:03] <leiz> sure, but what gyp options do we want to add?
[23:55:27] <vandebo> thakis: does going to -O2 make things link for you?
[23:55:30] *** erikkay has quit IRC
[23:55:43] *** [Nef] has quit IRC
[23:55:49] <thakis> vandebo: i will try
[23:56:04] <thakis> but trying stuff takes time, since it requires a full rebuid :-/
[23:56:10] <vandebo> indeed
[23:56:25] <vandebo> removing -g will almost certainly fix things
[23:56:32] *** cbentzel has quit IRC
[23:56:40] *** GeekShad0w has quit IRC
[23:56:47] <vandebo> but -O2 seems like a more graceful bandaid
[23:57:11] *** mrossetti has quit IRC
[23:58:38] <trungl-bot> Tree opened by vandebo at chromium dot org: Tree is open (help wanted: morejs perf test is crashing)
[23:58:47] <leiz> what about turning on the fastbuild option, that would just turn on -g1
[23:59:14] <vandebo> that would be smaller, and easy to do from .gyp_includes ?





top