[00:00:06] *** fqian has joined #chromium
[00:03:31] <thakis_afk>  /me cries
[00:03:39] *** thakis_afk is now known as thaki
[00:03:42] *** thaki is now known as thakis
[00:03:48] * thakis aslo can't type
[00:03:51] <thakis> *sigh*
[00:05:00] *** akalin has joined #chromium
[00:05:03] <akalin> greetings chromiumites
[00:05:12] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "base_unittests" on "Modules Linux" from 52136: albertb at chromium dot org)
[00:05:17] *** GeekShadow has quit IRC
[00:05:30] <akalin> does anyone have any numbers as to how many people use trunk builds?
[00:06:16] *** fqian has quit IRC
[00:06:39] <shess> Dammit.  I just decided it was going to stay open so I started a fetch ... and now I'm paying the git-gc penalty to no end.  SMASH!
[00:08:04] <jamesr> shess: cronjob that!
[00:08:27] <akalin> bleh
[00:08:49] <rsesek> akalin: because it's not Chrome I don't think there's metrics for that
[00:08:57] <akalin> i see
[00:09:13] *** hebz0rl_ is now known as hebz0rl
[00:09:25] <akalin> i'm asking because i've been able to gather metrics re. it and i wanted to double-check it
[00:09:37] <shess> I just set ac.auto to 50k, and now it happens seldom enough that I don't remember the last time.  Which is all I _really_ care about.
[00:09:38] <akalin> way more than i thought!
[00:09:49] <rsesek> akalin: do you have it broken down by OS?
[00:09:59] *** fqian has joined #chromium
[00:10:00] <akalin> rsesek: no, unfortunately
[00:10:18] <rsesek> akalin: ah. I'd bet that Linux users are the most common consumers of the waterfall builds
[00:10:25] <akalin> really?  why
[00:10:26] *** albertb has joined #chromium
[00:10:26] *** ChanServ sets mode: +v albertb
[00:10:38] <awong> akalin: because we're crazy?
[00:10:43] <rsesek> :D
[00:10:49] <akalin> gentoo mentality?
[00:10:53] <akalin> :D
[00:10:59] <albertb> Looks like I broke the tree, I'll revert my patch :(
[00:11:22] *** solsTiCe has left #chromium
[00:11:29] <awong> If I compile with -O1337, my browser will be so way faster than yours.
[00:11:51] <arv> albertb: thanks
[00:13:28] <jamesr> satish_: so the try bots use the WebKit rev listed in DEPS
[00:13:43] <satish_> yes, this change is only in chromium source
[00:13:44] <satish_> not webkit
[00:13:48] <albertb> arv: revert is in
[00:13:48] <jamesr> oh
[00:13:59] <jamesr> i think they use LKGR
[00:14:09] <jamesr> but i also think you can override that with -r
[00:14:39] <satish_> ah. that could be a reason
[00:14:44] <satish_> since the builders have been red for some time
[00:14:49] *** fqian_ has joined #chromium
[00:15:20] <trungl-bot> Tree opened by arv at chromium dot org: Tree is open
[00:15:22] *** mrossetti has quit IRC
[00:15:28] <satish_> thanks jamesr, i'll try with -r
[00:16:19] *** fqian has quit IRC
[00:16:19] *** fqian_ is now known as fqian
[00:16:21] <trungl-bot> Tree closed by buildbot at chromium dot org: Tree is closed (Automatic: "base_unittests" on "Chromium Linux x64" from 52136: albertb at chromium dot org, evan at chromium dot org (:evmar), mattm at chromium dot org (:mattm_g), pkasting at chromium dot org)
[00:16:33] <pkasting> It wasn't me!
[00:17:04] <albertb> I think this is still my patch. But I already landed the revert
[00:17:07] <awong> arv, jrg: We're way behind in webkit again.  I'd like to submit a ~250 revision roll. You guys okay with that?  It's passing most of the trys.
[00:17:13] <awong> arv, jrg: http://codereview.chromium.org/2958006/show
[00:17:53] <arv> awong: go for it
[00:18:00] <atwilson> Anyone familiar with the automation framework around to answer a question?
[00:18:26] <awong> arv: thanks. I'll be watching the layout tests.
[00:18:43] <atwilson> to whit: it seems that closing the last browser window is closing the channel from the proxy to the automation provider, even if the application isn't actually exiting. Is there special code doing that somewhere?
[00:19:17] *** dhollowa has joined #chromium
[00:19:17] *** ChanServ sets mode: +v dhollowa
[00:19:54] *** rafaelw2 has left #chromium
[00:22:49] *** jenkins has joined #chromium
[00:23:13] *** jenkins has left #chromium
[00:24:46] *** n3v3le has joined #chromium
[00:25:06] *** Venom_lnch is now known as Venom_X
[00:26:46] *** bmizeran_ has joined #chromium
[00:28:25] *** bmizerany has quit IRC
[00:28:29] <trungl-bot> Tree opened by arv at chromium dot org: Tree is open
[00:30:36] *** blueness has left #chromium
[00:31:41] *** markmentovai has left #chromium
[00:31:58] <akalin> not being able to get logging on windows chrome builds is the worst thing ever
[00:33:18] <inferno-sec> does anyone know why the Tabs test is breaking on linux. it starts breaking after my checkin but looks unrelated
[00:34:19] *** dhollowa has quit IRC
[00:36:04] <awong> I need some help with a mac build issue that's blocking the webkit roll.
[00:36:47] *** fta_ has joined #chromium
[00:37:02] <mattm_g> akalin: you can use sawbuck
[00:37:05] <awong> The mac try bots keep having issues liking tests against libwebcore.a, but I can't repro locally. any mac gurus able to help? http://build.chromium.org/buildbot/try-server/builders/mac_rel/builds/34/steps/compile/logs/stdio
[00:37:29] <akalin> mattm_g: oh, right
[00:37:30] <akalin> really?
[00:37:33] *** fta has quit IRC
[00:37:43] *** fta_ is now known as fta
[00:37:44] <eseidel> teh mac bot needs a kick: http://build.webkit.org/builders/Chromium%20Mac%20Release/builds/10547/steps/gclient/logs/stdio
[00:38:28] <xji_> anyway to get the actual pixel test result from trybot? For example, I have a failed test in http://test-results.appspot.com/dashboards/flakiness_dashboard.html#showExpectations=true&tests=fast%2Ftext%2Fatsui-spacing-features.html,  but how to get the pixel test result? And is the result of trybot the same as the one run in buildbot after changelist commit?
[00:39:07] <awong> eseidel: I think someone in 1950 has to do the mac bot kick.  I tried figuring out webkit bot access on friday, and couldn't find a way to do it remotely...
[00:39:23] <eseidel> awong: I emailed dglakzo and yaar
[00:39:34] *** Erkan_Yilmaz_ has joined #chromium
[00:39:52] *** malavv has quit IRC
[00:39:57] <jamesr> eseidel, awong: i can kick it
[00:40:01] <jamesr> what sorta kicking does it need?
[00:40:06] <awong> jamesr: svn cleanup...
[00:40:06] <jamesr> svn cleanup somewhere?
[00:40:16] <jamesr> k. dglazkov and yaar aren't around currently
[00:40:46] <eseidel> ha, now the linux bot too
[00:40:49] <awong> jamesr: is there a way to fix those remotely?
[00:40:57] <awong> couldn't find any docs
[00:41:01] <jamesr> awong: believe no
[00:41:07] <awong> suck.
[00:41:10] <jamesr> word
[00:41:17] <mattm_g> akalin: yeah, http://code.google.com/p/sawbuck/
[00:42:46] *** Erkan_Yilmaz has quit IRC
[00:43:35] *** tonikitoo has joined #chromium
[00:45:51] <xji_> eseidel: ping
[00:46:11] <pkasting> awong: Are you going to be ready to roll soon?
[00:46:20] <awong> pkasting: no
[00:46:23] <pkasting> awong: If not I'm inclined to close the tree for .grd landings
[00:46:31] <awong> pkasting: *sigh* go for it.
[00:46:42] *** jrmuizel has joined #chromium
[00:46:49] <pkasting> jrg, hwennborg, arv: Any objections to closing for .grd landings?
[00:47:04] <arv> pkasting no
[00:47:11] <eseidel> xji_: pong
[00:47:12] <arv> no objections
[00:47:47] <trungl-bot> Tree closed by pkasting at chromium dot org: Tree closed for .grd landings
[00:48:01] <pkasting> I'll wait a little for the bots to catch up some and for me to re-sync
[00:50:03] *** jrmuizel has quit IRC
[00:50:13] <xji_> eseidel: I have a question regarding to how to get the pixel test result from trybot. are you able to see the questsion posted? my run is 2125
[00:51:32] *** zaspire_ has joined #chromium
[00:51:55] *** zaspire has quit IRC
[00:52:46] <eseidel> xji_: what try bot?
[00:53:00] <eseidel> the EWS bots are not try bots. :)  and don't post results
[00:53:11] <eseidel> if you mean maruel's try bots, I know nothing about those
[00:54:02] <xji_> eseidel: chromium's layout test trybot, the one we use when run "gcl try xxx -b layout_linux". what is EWS.
[00:54:39] <thakis> rsesek: around?
[00:54:50] <eseidel> xji_: I would ask maruel or jamesr or someone who works on chromium
[00:55:07] <xji_> eseidel: ok. thanks anyway.
[00:55:37] <xji_> jamesr: ping
[00:55:45] <eseidel> xji_: EWS is the little bubbles on webkit bugs, that warn you if your patch is going to break things
[00:56:02] <pkasting> Wait a minute, why is chromium-builder (win dbg) not compiling
[00:56:08] <pkasting> I'm not checking in on a broken tree
[00:56:15] <xji_> eseidel: I see.
[00:56:27] <pkasting> nsylvain: Can you glance at http://build.chromium.org/buildbot/waterfall/waterfall?builder=Chromium%20Builder%20(dbg) ?
[00:56:35] <nsylvain> looking
[00:56:37] <pkasting> I'm not sure what's wrong, something about not finding PDB files...
[00:56:39] *** mrossetti has joined #chromium
[00:57:15] *** Venom_X has quit IRC
[00:57:26] <nsylvain> looks like it needs a clobber, i'll trigger one
[00:58:50] <pkasting> Should I check in the .grd changes I want to commit, so they get in the clobber?
[00:59:26] <pkasting> nsylvain: ^
[01:01:13] <nsylvain> sure
[01:01:32] <pkasting> OK, landing
[01:02:13] <pkasting> Done
[01:02:42] <jamesr> xji_: pong
[01:03:14] *** agayev has left #chromium
[01:03:14] *** janm has quit IRC
[01:03:31] <xji_> jamesr: do you know is there a way to read the pixel test result from layout test trybot (the one we use when run "gcl try xxx -b layout_linux")?
[01:03:48] <jamesr> xji_: they don't preserve results
[01:03:52] *** Buglouse has joined #chromium
[01:03:55] <jamesr> afaik
[01:03:59] <rsesek> thakis: about to head out. what's up?
[01:04:08] <thakis> rsesek: sent you a review
[01:04:25] <jamesr> xji_: i think you can log in to them interactively and grab the results if you really need to. what's the context?
[01:04:26] <thakis> rsesek: but sent it to trungl as well
[01:04:46] <jamesr> do you not have access to the platform to run the test locally?
[01:04:53] <rsesek> thakis: k. I'll take a look when I get in tomorrow
[01:05:03] <awong> xji_: If you're trying to checkin new layout tests, it might be easiest to check them with with an expectation set to fail. then rebaseline it off the canary.
[01:05:26] <xji_> jamesr: I have a failed test, which is expected after my change. Last time, I remember I grabbed the test result locally, submitted it, but seems that the pixel result from buildbot is different from mine.
[01:05:39] <jamesr> xji_: you want to know why that is
[01:05:42] *** rsesek has quit IRC
[01:05:48] <xji_> jamesr: ya, that is my question, whether I can rebaseline before commit.
[01:05:54] <jamesr> xji_: if it's just platform differences (for example, text is different on every platform) then you want to do what awong said
[01:06:08] <vandebo> the windows tsan failure is probably mine, I'm looking at it
[01:07:17] <vandebo> I take that back, probably mbelshe
[01:07:42] <xji_> jamesr: it is the image result difference. I will set it as fail and rebaseline off canary. BTW, what awong said?
[01:08:33] <vandebo> I let him know
[01:09:33] *** rafaelw2 has joined #chromium
[01:11:24] *** rafaelw2 is now known as rafaelw
[01:14:52] *** rafaelw has quit IRC
[01:18:54] *** rafaelw2 has joined #chromium
[01:19:25] *** rsleevi has joined #chromium
[01:21:43] <jamesr> xji_: yes, that's what awong said
[01:23:55] *** ebraminio has quit IRC
[01:26:26] *** seventh has joined #chromium
[01:26:36] *** fta_ has joined #chromium
[01:27:34] *** fta has quit IRC
[01:27:43] *** fta_ is now known as fta
[01:28:56] *** abarth has quit IRC
[01:32:13] <thakis> <weekly cmd-q rage />
[01:32:23] <akalin> eh?
[01:32:25] <akalin> what rage
[01:32:30] <eglaysher> thakis: well at least you got it out of the way on Monday.
[01:32:39] <eglaysher> it is all uphill from here.
[01:32:44] <akalin> aw fuck
[01:32:50] <akalin> bitten by memory space exhaustion again
[01:32:53] <thakis> akalin: I HIT CMD-Q AND MY BROWSER DISAPPEARED FFFFFFFUUUUUUUUUUU
[01:32:56] <akalin> time to rebuild with svg off :(
[01:33:01] <akalin> thakis: oh
[01:33:16] <thakis> eglaysher: :-)
[01:33:24] <akalin> what's the magic incantation again?
[01:33:30] <thakis> akalin: soon you will be able to build with mathml off, too!
[01:33:32] <akalin> GYPDEFINES=enable_svg=0 ?
[01:33:39] <thakis> akalin: add enable_svg=0 to ~/.gyp/includes.gypi
[01:33:43] <akalin> wait, mathml is landing in chrome? :O
[01:33:54] <akalin> oh, cool
[01:33:58] <thakis> akalin: sounds like they want to turn it on by default in webkit
[01:34:04] <akalin> nice
[01:34:09] <jamesr> that's always been the plan
[01:34:15] <akalin> i was thinking about making a TeX -> html5 converter like jsmath
[01:34:17] *** Cheery has quit IRC
[01:34:23] <thakis> akalin: http://codepad.org/k0D9woTS
[01:34:30] <jamesr> i don't think anyone is planning to flip the switch in the immediate future
[01:34:35] <akalin> thakis: thank you
[01:34:49] *** sebmarkbage has quit IRC
[01:34:58] <thakis> akalin: if you pick the two ffmpeg flags, you will get a harmless error when building
[01:35:02] <jamesr> although that's up to us (the embedder)
[01:35:08] <akalin> harmless error?
[01:35:21] <thakis> akalin: "failed to copy libffmpegsumo.dylib" or similar
[01:35:25] <akalin> oh
[01:35:25] <akalin> ok
[01:36:44] *** awolfson has quit IRC
[01:36:57] <akalin> thakis: so do you compile chrome on clang regularly now?
[01:37:11] <thakis> akalin: i did for a while, but i haven't in 1.5 weeks or so
[01:37:22] <thakis> akalin: once i have a few spare cycles again, i will
[01:37:26] <akalin> mmm
[01:37:31] <akalin> did it work?
[01:37:40] <thakis> akalin: debug builds work
[01:37:47] <akalin> neat
[01:37:51] <thakis> release builds don't link due to bugs in clang's inliner
[01:37:53] *** yutak_home has joined #chromium
[01:37:54] *** ChanServ sets mode: +v yutak_home
[01:38:04] <thakis> or symbol visibility emitter
[01:38:29] <thakis> the debug unit_tests binary produced by clang crashes halfway through
[01:38:31] <trungl-bot> Tree opened by arv at chromium dot org: Tree is open
[01:38:36] *** skrul has quit IRC
[01:39:51] *** satish_ has quit IRC
[01:41:45] *** abarth has joined #chromium
[01:41:46] *** ChanServ sets mode: +v abarth
[01:42:23] *** NthDeGeek has joined #chromium
[01:43:15] <pkasting> arv: Why did you open?
[01:43:35] <trungl-bot> Tree closed by pkasting at chromium dot org: Tree closed for .grd landings
[01:43:36] <akalin> thakis: did you have to nuke xcodebuild when you set your new defines?
[01:43:42] <akalin> or at least for enable_svg=0
[01:43:50] <thakis> akalin: i think so, yes
[01:43:51] <pkasting> arc: The tests haven't even cycled for the broken build, let alone the correct one
[01:43:54] <pkasting> er, arv
[01:43:56] <akalin> ugh
[01:44:04] <arv> pkasting: sorry
[01:44:07] <jamesr> if you rerun gclient runhooks it should dirty things up appropriately
[01:44:16] <thakis> jamesr: it doesn't
[01:44:18] <jamesr> can xcodebuild not track that?
[01:44:25] <jamesr> it works OK on linux+make
[01:45:13] <thakis> jamesr: it can probably be made to work somehow, but enabling/disabling svg happens less often than, say, a grd change
[01:45:16] <thakis> so making this work isn't as important
[01:45:34] <jamesr> you have to rebuild all of webcore anyway so it's not much of a difference in build time
[01:45:51] <akalin> jamesr: i did run gclient runhooks
[01:46:28] <akalin> my co-worker informed me that i should change my login/chromium id/irc name to 'frak'
[01:46:53] <atwilson> yeah, that seems like a win.
[01:47:01] <evmar_afk> can i land some stuff?
[01:47:02] *** sbyer has left #chromium
[01:47:03] <evmar_afk> i see everyone else is
[01:47:11] <akalin> damn, already taken :(
[01:47:37] *** evmar_afk is now known as evmar
[01:47:41] <evmar> oh, i see, accidentally open
[01:47:51] * evmar will commit in the evening or sumthin
[01:47:52] *** jaffacakes has joined #chromium
[01:48:16] <thakis> evmar: the tree won't ever reopen
[01:48:18] <thakis> chrome's done
[01:48:22] <thakis> everyone go home
[01:48:55] <akalin> it hasn't even been ported to go yet
[01:49:21] *** sbyer has joined #chromium
[01:49:21] *** ChanServ sets mode: +v sbyer
[01:49:31] <akalin> is svg seriously that heavyweight?
[01:49:56] <akalin> is it not a separate library from webcore or something?
[01:50:08] <akalin> i'm just wondering why enabling/disabling it requires basically a full rebuild
[01:50:39] <jamesr> it's retarded
[01:50:57] <jamesr> it requires a full rebuild because of #ifdefs in headers that are included everywhere
[01:51:05] <akalin> i see
[01:51:06] <akalin> blurggh
[01:51:08] <jamesr> and it's heavy enough to cause link failures because it has a lot of classes / template overhead
[01:51:15] <jamesr> and the implementation is kinda dumb
[01:51:20] <thakis> reminds me of mng
[01:51:37] *** RT|Chatzilla has joined #chromium
[01:51:38] <akalin> what's dumb about the implementation
[01:52:02] *** ebraminio has joined #chromium
[01:52:24] <tony^work> doesn't changing any c define require a full rebuild (command line changed)?
[01:53:54] <akalin> lol C :(
[01:55:43] <awong> arv: ping
[01:57:09] <arv> awong: ack
[01:58:53] <awong> arv: thakis found that in http://trac.webkit.org/changeset/63038, removing CSSPropertyWebkitPadding{End,Start} breaks the mac chromium build.  Is it okay if I just add those two bad into the switch?
[01:59:12] <atwilson> Anyone have any tips for debugging the animation provider framework - in particular, I want to debug the browser instance that's created when I run ui_tests.exe.
[02:00:03] <arv> awong: Yes
[02:00:22] <jamesr> akalin: it just has some legacy oddness. same as any old codebase
[02:00:23] <arv> awong: Those are never part of the switch since they get handled before the switch
[02:01:28] <thakis> awong: do you build on mac locally?
[02:01:43] <awong> arv: I'm confused now...are you saying they shouldn't be in the switch?
[02:01:55] <awong> thakis: I've got a local mac build going now, yes.
[02:02:04] <thakis> awong: they don't need to be 'cause they're handled earlier, but gcc complains if they aren't in the switch
[02:02:18] <arv> awong: Our build seems to enfoce that all cases are covered
[02:02:26] <awong> thakis: right. so they need to be in the switch for -Werror compliance.
[02:02:30] <arv> awong: the webkit builds do not seem to be that strict
[02:02:32] <thakis> yupp
[02:02:48] <awong> If this were chromium code, I'd do a default: NOT_REACHED()
[02:03:04] <thakis> arv: ours is that strict only since a few weeks. thomasvl put in a lot of work to make that happen on mac
[02:03:11] <arv> They should preferably be in another section than where they were before since the are implemented
[02:03:17] <thakis> (and i think evmar is doing/did the same for linux)
[02:03:36] <arv> makes sense
[02:03:59] <awong> is there an equiv of not reached in webkit?  Or should I just stuff it into another seciton with a break, and a comment that says "shouldn't get executed"?
[02:04:39] *** albertb has quit IRC
[02:04:39] <arv> awong: Can you put them after the shorthand section with a comment like: Diractional properties are resolved before the switch
[02:05:00] *** abarth has quit IRC
[02:05:19] *** abarth has joined #chromium
[02:05:20] *** ChanServ sets mode: +v abarth
[02:05:27] <tony^work> awong: ASSERT_NOT_REACHED(), IIRC
[02:05:53] <awong> arv: Sure. "Diractional" is the right word?
[02:06:18] <awong> tony^work: Cool. I'll give that a shot.
[02:06:20] *** MikeSmithW3C^ has joined #chromium
[02:06:25] <arv> awong: directional
[02:06:30] <awong> arv: okay
[02:06:38] <awong> sorry for the silly question...
[02:06:56] <arv> awong: These are resolved by resolveDirectionAwareProperty before the switch
[02:07:30] <awong> arv: ic
[02:07:42] <akalin> aargh
[02:07:47] <jrg> arv: need to hit the road early.  Still working on Mac memory bots; there are a ton of things backed up in there.  Will send you some TBRs tonight.
[02:07:48] <akalin> i had includes.gypi instead of include.gypi :(
[02:08:36] *** tedoc2000 has quit IRC
[02:09:25] <evmar> thakis: i turned of -Werror for webkit
[02:09:30] <arv> awong: There should be 4 direction aware css properties; margin-start, margin-end, padding-start, pading-end
[02:09:39] <evmar> thakis: we have no way to do quick fixes for webkit, and there's so much red tape around slow fixes
[02:09:39] *** MikeSmithW3C has quit IRC
[02:09:42] <thakis> evmar: oh. too noisy?
[02:09:52] <evmar> we have -Wall, just not -Werror i think
[02:10:51] <awong> arv: should WebKitMarginEnd and WebKitMarginStart also be moved into that section?  Currently, they're still in the unimplemented -webkit- properties section.
[02:10:54] *** hasimo_ has quit IRC
[02:11:32] *** hasimo has joined #chromium
[02:13:05] <pkasting> jrg, hwennborg, arv: Are you guys looking at the red and orange bots?  It looks like there are some longstanding ones as well as some new ones that came up after the commits after my .grd changes (the .grd changes themselves are looking clean so far but haven't all cycled)
[02:13:37] <jrg> pkasting: I'm working through memory bots which are angry red, but it takes time.
[02:13:54] *** hasimo has joined #chromium
[02:14:02] <pkasting> K... I just want to make sure my holding the tree closed doesn't make us overlook anything
[02:14:31] <jrg> pkasting: the purple is because we had to restart chromium.master.
[02:15:31] <jrg> pkasting: looks like much of the orange self-cleaned, but feel free to keep tree closed a bit after you are done to let things settle.  The Mac valgrind bot has a ton of problems (accum errors) which will make it harder to sort out... best to not let the cardhouse fall in other areas if possible.
[02:16:33] *** elmargol has quit IRC
[02:17:01] *** fta_ has joined #chromium
[02:17:13] *** evmar is now known as evmar_givingupfo
[02:17:17] *** evmar_givingupfo is now known as evmar_afk
[02:17:21] *** sytse has quit IRC
[02:17:26] *** sytse has joined #chromium
[02:17:51] *** fta has quit IRC
[02:17:57] *** fta_ is now known as fta
[02:18:35] *** elmargol has joined #chromium
[02:19:05] *** legion13 has joined #chromium
[02:19:40] <akalin> humph
[02:19:41] *** AaronMT has quit IRC
[02:19:46] <akalin> i'm getting compile errors with enable_svg=0
[02:20:02] <akalin> e.g. Users/akalin/src/chromium/src/third_party/WebKit/WebCore/WebCore.gyp/../../../../xcodebuild/DerivedSources/Debug/webcore/bindings/V8HTMLIFrameElement.cpp:203: error: 'class WebCore::HTMLIFrameElement' has no member named 'getSVGDocument'
[02:20:12] <akalin> did a runhooks and then nuked xcodebuild
[02:20:27] <akalin> anyone have any ideas?
[02:20:56] <jamesr> sounds like the bindings are missing something
[02:20:58] <jamesr> what rev are you at?
[02:21:08] <arv> The NaCL failures seems legit but I cannot figure out what changed caused it. Nothing in the vicinity of the breakage looks related
[02:21:16] <akalin> jamesr: 52136
[02:21:31] <jamesr> what webkit rev?
[02:21:50] <jamesr> it looks like that's behind the appropriate #ifdefs
[02:22:03] <akalin> 62844
[02:22:25] <jamesr> dunno. that should work
[02:22:26] <akalin> should i sync to head and pray?
[02:22:33] *** tonyg-cr has quit IRC
[02:22:35] <akalin> although it sounds like the tree is in a crappy state now
[02:23:21] <thakis> akalin: did you nuke all of xcodebuild or only xcodebuild/Debug? the generated stuff is in xcodebuild/DerivedSources/Debug
[02:23:25] <thakis> you need to delete that too
[02:23:32] <akalin> i nuked xcodebuild
[02:23:57] <akalin> what's the command to fidn the lkgr?
[02:24:12] *** MikeSmithW3C^ has quit IRC
[02:24:29] <thakis> trungl-bot: lkgr
[02:24:30] <trungl-bot> thakis: 52133
[02:24:32] <thakis> akalin: ^
[02:24:35] <akalin> aha!
[02:24:53] <akalin> i'm only a few after that
[02:24:53] *** MikeSmithW3C has joined #chromium
[02:24:54] <thakis> or do a bing search for "lkgr"
[02:24:55] <akalin> blurgh
[02:25:41] <thakis> hm, looks like this only works if you do your bing search with google
[02:25:49] <thakis> bing proper doesn't find the right hit
[02:26:47] <arv> Any NaCL people here?
[02:27:33] *** slightlyoff has joined #chromium
[02:27:34] *** ChanServ sets mode: +v slightlyoff
[02:29:27] <pkasting> jrg, hwennborg, arv: It looks to me like at this point the .grd changes seem basically clean; most have cycled through, and red/orange either was already present or showed up after.  I'm leaving the tree closed at jrg's suggestion above, but at this point feel free to reopen or manage however you like
[02:29:50] *** elmargol has quit IRC
[02:30:05] *** rafaelw2 has quit IRC
[02:30:11] <arv> pkasting: thanks. yeah the tree is too red to be open
[02:30:18] *** fqian has quit IRC
[02:31:06] <nirnimesh> "can't map file, errno=12" during 'ld' on leopard is the 32-bit problem right, with the current solution to use snow leopard?
[02:31:16] <dcheng> jam2: For crbug.com/48031, I'm not entirely convinced it's caused by printing.
[02:31:30] <dcheng> jam2: Or PDFs for that matter.
[02:33:17] *** General1337 has quit IRC
[02:36:44] <slightlyoff> my, that's one unhappy mac bot we have there
[02:36:52] *** tonyg-cr has joined #chromium
[02:36:53] *** ChanServ sets mode: +v tonyg-cr
[02:36:56] <thakis> nirnimesh: yes
[02:36:56] *** elmargol has joined #chromium
[02:37:09] <thakis> nirnimesh: you can kinda work around the problem if you don't want to go to 10.6 yet
[02:38:43] *** dmaclach has quit IRC
[02:38:48] *** General1337 has joined #chromium
[02:45:15] <awong> arv: do you think I need a review for the CSSComputedStyleDeclaration change?
[02:45:39] <arv> No, build fixes are excempt
[02:45:43] <awong> arv: k
[02:45:51] <arv> awong:  bug#?
[02:46:07] <awong> arv: I didn't create a bug...
[02:46:22] *** chrisccoulson_ has quit IRC
[02:49:30] <trungl-bot> Tree closed by arv at chromium dot org: Tree closed
[02:50:30] <trungl-bot> Tree closed by arv at chromium dot org: Tree is closed
[02:50:36] <slightlyoff> arv: any idea whats got things closed now?
[02:52:50] *** ebraminio has quit IRC
[02:52:51] <arv> slightlyoff: things are just red all over. I cannot pin point the cause
[02:53:01] <slightlyoff> ok
[02:54:01] *** abarth has quit IRC
[02:54:46] <awong> arv: https://bugs.webkit.org/attachment.cgi?id=61304&action=review
[02:58:27] <arv> awong:  You need to call out that that the patch is unreviewed in the ChangeLog
[02:58:57] <awong> arv: yep...I'm getting it reviewed actually.
[02:59:07] <arv> awong: even better
[02:59:17] <awong> does it look about right to you though?
[02:59:47] *** abarth has joined #chromium
[02:59:48] *** ChanServ sets mode: +v abarth
[02:59:54] *** abarth has quit IRC
[03:01:20] *** fta_ has joined #chromium
[03:03:09] *** fta has quit IRC
[03:03:12] *** fta_ is now known as fta
[03:06:06] *** jaffacakes has left #chromium
[03:06:59] *** hebz0rl_ has joined #chromium
[03:07:25] *** leeight has joined #chromium
[03:07:48] *** yutak_home has quit IRC
[03:07:59] *** leeight has left #chromium
[03:08:56] *** mazda has joined #chromium
[03:08:57] *** ChanServ sets mode: +v mazda
[03:11:06] *** hebz0rl has quit IRC
[03:14:46] *** jshin has quit IRC
[03:15:38] *** agayev has joined #chromium
[03:19:20] <arv> awong: looks great. Are you doing another roll?
[03:19:48] <awong> arv: running it through try servers now. unlikely to actually make it tonight.
[03:19:58] <awong> I think I'm going ot pass it off to pavel in moscow when he wakes up.
[03:20:28] <awong> http://codereview.chromium.org/2958006/show if you're curious
[03:20:31] *** Dykam has joined #chromium
[03:21:03] *** trungl is now known as trungl_afk
[03:22:22] <awong> Alternately, if someone in asia is feeling adventurous, they're welcome to pickup the patch and commit once it passes the try servers.
[03:23:48] <jamesr> those two numbers in the roll are really different from one another :P
[03:24:02] <awong> hence the leap comment :-/
[03:26:17] *** tkent has joined #chromium
[03:26:19] *** ChanServ sets mode: +v tkent
[03:27:56] *** leeight has joined #chromium
[03:27:58] *** leeight has left #chromium
[03:30:47] *** awolfson has joined #chromium
[03:33:34] *** urbanape has quit IRC
[03:33:44] *** hbono has joined #chromium
[03:35:58] *** ebraminio has joined #chromium
[03:37:14] *** bevc_work has quit IRC
[03:38:47] *** mrossetti has quit IRC
[03:42:17] <arv> pkasting: the ordering of the cros resources broke a test
[03:51:17] *** fta_ has joined #chromium
[03:51:50] *** fta has quit IRC
[03:51:59] *** fta_ is now known as fta
[03:55:27] *** oshima has joined #chromium
[03:55:28] *** ChanServ sets mode: +v oshima
[03:57:57] *** csilv has left #chromium
[03:58:55] *** ebraminio has quit IRC
[03:59:47] *** hebz0rl_ has quit IRC
[04:02:22] <oshima> hello is there a sheriff?
[04:04:25] <kermit> does browsing history ever expire?
[04:05:42] <oshima> is anyone looking at tree closure?
[04:07:35] *** bweinstein has quit IRC
[04:08:37] *** trungl has joined #chromium
[04:08:37] *** ChanServ sets mode: +v trungl
[04:08:59] <hbono> oshima: arv?
[04:09:21] *** stuartmorgan has quit IRC
[04:09:38] <arv> hbono: I'm marking things as FAIL_. We had a lot of fallout duo to massive grd changes
[04:10:35] <hbono> arv: thank you for your description.
[04:10:59] *** dave_levin_ has joined #chromium
[04:11:22] <oshima> arv: i have two changes I'd like to check in to disable tests for chromeos try bots.
[04:11:44] <oshima> http://codereview.chromium.org/2909011/show
[04:11:47] <arv> I just marked the cros tests as failing
[04:11:54] <oshima> http://codereview.chromium.org/2895012/show
[04:12:04] <arv> this is different
[04:12:08] <arv> oshima: lgtm
[04:12:54] <oshima> arv: thanks! they've been failing for quite some time on trybots because they fails only on dbg build :(
[04:13:45] *** dave_levin has quit IRC
[04:13:46] *** dave_levin_ is now known as dave_levin
[04:14:11] <dave_levin> awong: hi
[04:14:33] *** tyoshino has joined #chromium
[04:16:53] *** arv is now known as arv_away
[04:20:38] *** tonyg-cr has quit IRC
[04:21:23] *** pathorn has quit IRC
[04:23:14] *** tfarina has joined #chromium
[04:24:01] <tfarina> hum, is someone looking the tree? is there any sheriff around?
[04:24:39] <jamesr> arv is
[04:24:58] <jamesr> we're still recovering from .grd, i believe
[04:25:14] *** zaspire_ has quit IRC
[04:25:15] <tfarina> wow, is tooking much time :(
[04:25:49] *** ebraminio has joined #chromium
[04:26:03] *** rafaelw2 has joined #chromium
[04:26:14] *** dr_win has quit IRC
[04:26:46] <tfarina> it's looking pretty red though.
[04:27:42] <jamesr> yup. i'm just a messenger, have to ask arv
[04:29:54] <tfarina> np, looks like the bots are still compiling. the should cycle green soon.
[04:30:05] <tfarina> I just wanted to commit my change.
[04:30:38] *** legion13 has quit IRC
[04:32:14] * trungl votes to shoot jamesr.
[04:35:27] *** fta_ has joined #chromium
[04:36:35] *** fta has quit IRC
[04:36:36] *** fta_ is now known as fta
[04:38:06] *** johnny_g_ has joined #chromium
[04:39:38] *** johnny_g_ has quit IRC
[04:39:43] *** jamesr has quit IRC
[04:40:30] *** johnny_g has quit IRC
[04:40:56] *** tfarina has quit IRC
[04:41:37] *** akalin has quit IRC
[04:41:57] *** thakis_ has joined #chromium
[04:41:57] *** ChanServ sets mode: +v thakis_
[04:42:19] *** tonyg-cr has joined #chromium
[04:42:19] *** ChanServ sets mode: +v tonyg-cr
[04:44:36] *** trungl has quit IRC
[04:45:12] *** tonyg-cr has quit IRC
[04:48:27] *** nick_ has joined #chromium
[04:48:55] *** nick_ is now known as Guest25386
[04:51:13] *** Guest25386 has quit IRC
[04:51:27] *** ebraminio has quit IRC
[04:52:30] *** tonyg-cr has joined #chromium
[04:52:30] *** ChanServ sets mode: +v tonyg-cr
[04:54:02] *** Aria has joined #chromium
[05:00:45] *** shoe has quit IRC
[05:01:03] *** shoe has joined #chromium
[05:10:45] *** slightlyoff has quit IRC
[05:19:08] *** rafaelw2 has quit IRC
[05:22:50] *** trungl has joined #chromium
[05:22:50] *** ChanServ sets mode: +v trungl
[05:22:59] *** fta_ has joined #chromium
[05:24:23] *** fta has quit IRC
[05:24:35] *** fta_ is now known as fta
[05:32:12] *** dave_levin has quit IRC
[05:39:40] *** dmaclach has joined #chromium
[05:53:04] *** dr_win has joined #chromium
[05:55:15] <oshima> arv: ping
[05:56:45] *** thakis_ has quit IRC
[05:58:27] *** loislo has quit IRC
[06:15:52] *** fta_ has joined #chromium
[06:16:45] *** fta has quit IRC
[06:16:50] *** fta_ is now known as fta
[06:18:15] *** loislo has joined #chromium
[06:19:02] *** suzhe has joined #chromium
[06:19:02] *** ChanServ sets mode: +v suzhe
[06:19:51] *** bandu has quit IRC
[06:20:46] *** dave_levin has joined #chromium
[06:25:13] *** dmaclach has quit IRC
[06:25:27] *** elmargol has quit IRC
[06:26:16] *** dmaclach has joined #chromium
[06:27:22] *** pathorn has joined #chromium
[06:28:41] *** loislo has quit IRC
[06:30:21] *** agayev has left #chromium
[06:31:38] *** elmargol has joined #chromium
[06:32:32] *** nayankk has quit IRC
[06:32:55] *** nayankk has joined #chromium
[06:33:27] *** pankajm has joined #chromium
[06:38:06] *** tedoc2000 has joined #chromium
[06:46:00] *** amruthraj has quit IRC
[06:47:02] *** thakis_ has joined #chromium
[06:47:02] *** ChanServ sets mode: +v thakis_
[06:50:14] *** dave_levin has quit IRC
[06:50:48] *** bers has joined #chromium
[06:57:35] *** michaeln has quit IRC
[07:02:24] *** fqian has joined #chromium
[07:03:11] *** johnny_g has joined #chromium
[07:11:18] *** pfeldman has joined #chromium
[07:12:50] *** pfeldman_ has joined #chromium
[07:13:45] *** tedoc2000 has quit IRC
[07:16:07] *** bevc has joined #chromium
[07:16:28] *** pfeldman has quit IRC
[07:16:28] *** pfeldman_ is now known as pfeldman
[07:16:32] *** kphanee has quit IRC
[07:16:53] *** kphanee has joined #chromium
[07:17:17] *** kphanee has left #chromium
[07:20:23] *** fqian__ has joined #chromium
[07:21:46] *** fqian has quit IRC
[07:21:46] *** fqian__ is now known as fqian
[07:24:53] *** loislo has joined #chromium
[07:28:02] *** tonyg-cr has quit IRC
[07:29:29] *** avinashhebbur has joined #chromium
[07:29:37] *** ebraminio has joined #chromium
[07:29:37] *** Aria has left #chromium
[07:30:02] *** Aria has joined #chromium
[07:30:56] <avinashhebbur> hi... can anyone tell me how the history page is designed in chrome browser....
[07:31:00] *** viro101 has quit IRC
[07:31:20] *** viro101 has joined #chromium
[07:32:02] *** bmizeran_ has quit IRC
[07:36:51] *** thakis_ has quit IRC
[07:40:34] *** bweinstein has joined #chromium
[07:40:38] *** mranima has joined #chromium
[07:41:05] *** ebraminio1 has joined #chromium
[07:41:05] <mranima> hello?
[07:41:14] *** bweinstein has quit IRC
[07:41:22] *** bweinstein has joined #chromium
[07:41:33] <mranima> I have a concert with Chrome?
[07:42:07] *** Buglouse has quit IRC
[07:42:08] *** ebraminio has quit IRC
[07:45:37] *** fqian has quit IRC
[07:45:43] *** mranima has left #chromium
[07:47:27] *** alt-dot-net-geek has joined #chromium
[07:48:53] *** bevc has quit IRC
[07:52:04] *** estes has joined #chromium
[07:52:19] *** thakis_ has joined #chromium
[07:52:19] *** ChanServ sets mode: +v thakis_
[07:53:05] *** dmaclach has quit IRC
[07:53:31] *** dmaclach has joined #chromium
[07:55:01] *** fta_ has joined #chromium
[07:55:53] *** fta has quit IRC
[07:56:04] *** fta_ is now known as fta
[07:57:36] *** bers has quit IRC
[07:58:44] *** Precea is now known as Precea[BNC]
[07:59:12] *** rsleevi has quit IRC
[08:00:44] *** js2 has quit IRC
[08:02:48] *** FunkyPenguin has quit IRC
[08:05:51] *** johnny_g has quit IRC
[08:06:48] *** MikeSmithW3C^ has joined #chromium
[08:09:27] *** MikeSmithW3C^ has quit IRC
[08:10:09] *** MikeSmithW3C has quit IRC
[08:11:34] *** MikeSmithW3C has joined #chromium
[08:13:24] *** n3v3le1 has joined #chromium
[08:13:31] *** n3v3le has quit IRC
[08:14:19] <thakis_> avinashhebbur: hit cmd-opt-u to "view source"
[08:14:24] <thakis_> (or whatever the shortcut is on windows)
[08:18:35] <avinashhebbur> @thakis_:in btw thank u... actully i m in search of the backend of the page... are they running a server or just creating a html page and showing the history items in the pages...
[08:19:27] <thakis_> avinashhebbur: that's in chrome/browser/dom_ui/history*
[08:19:34] <thakis_> (not running a server)
[08:21:11] <avinashhebbur> okay... yup i got it man.... thanks a lot:)
[08:21:56] <trungl> thakis_!
[08:22:01] *** barcon332 has quit IRC
[08:22:11] 
[08:22:16] <trungl> Indeed!
[08:22:25] <thakis_> ZZZZZZZZZZZZZZZZZZZZZ
[08:22:35] *** Weems has joined #chromium
[08:22:53] <trungl> I think Chrome needs a dedicated (native) vuvuzela button.
[08:22:57] <suzhe> thakis_:  are you awake?
[08:23:10] <thakis_> suzhe: barely
[08:23:32] <trungl> now that we've cleared some room on the toolbar, we surely have room for it
[08:23:38] <suzhe> thakis_: I'm still busying porting browser_keyevents_browsertest.cc to mac.
[08:23:58] <trungl> maybe we should have a dedicated vuvuzela-bar
[08:23:59] <thakis_> suzhe: cool. where's that test? what does it test?
[08:24:07] <suzhe> thakis_: But I encountered an issue, may be you can help me on it.
[08:24:11] <trungl> browser key events?
[08:24:22] * trungl ducks.
[08:24:26] <thakis_> suzhe: the interactive_ui_tests not running on the bots?
[08:24:30] *** satorux has joined #chromium
[08:24:31] *** ChanServ sets mode: +v satorux
[08:24:51] <suzhe> thakis_: I moved the test into browser_tests :)
[08:25:20] <thakis_> :-)
[08:25:25] <suzhe> thakis_: but the problem is, when I start the test in commandline, the browser window can never get the focus.
[08:26:28] <suzhe> thakis_: Do you know the reason?
[08:26:42] <satorux> the tree is closed?
[08:27:04] <trungl> trungl-bot: treestatus
[08:27:05] <trungl-bot> trungl: Tree status set by arv at chromium dot org: Tree is closed
[08:27:49] <thakis_> suzhe: not off-hand
[08:28:00] <thakis_> suzhe: the test uses the normal browser window classes, right?
[08:28:36] <suzhe> thakis_: yes. Should be no difference than normal one, except for some testing code.
[08:29:23] *** [Nef] has quit IRC
[08:29:25] <suzhe> thakis_: If I add some delay in the test, then I can click the window to focus it, then the test works.
[08:29:48] *** viro101 has quit IRC
[08:30:11] <thakis_> suzhe: let me check something; i have an idea
[08:30:39] <suzhe> thakis_: Do you want me to upload the CL? I may need to clean up the code.
[08:30:54] <thakis_> no need i think
[08:31:13] <suzhe> thakis_: good. Thanks.
[08:31:50] <suzhe> thakis_: besides, things like [window orderFront] do not work.
[08:33:56] <thakis_> suzhe: i confused this with CocoaTestHelper which calls orderBack:. looks like that's not the culprit here
[08:34:45] <suzhe> thakis_: where is it called?
[08:34:55] *** pamg_ has joined #chromium
[08:34:55] *** ChanServ sets mode: +v pamg_
[08:35:05] <thakis_> that's called only in unit_tests i think, not in browser_tests
[08:36:35] <suzhe> thakis_: yes.
[08:38:07] <pamg_> Ping hwennborg ?
[08:38:11] *** pamg_ is now known as pamg
[08:38:44] *** ebraminio has joined #chromium
[08:39:42] *** Aria has quit IRC
[08:40:57] <suzhe> thakis_: seems that any program started from commandline can't get focus. Just tested Safari, the same behavior.
[08:41:56] *** ebraminio1 has quit IRC
[08:42:38] <thakis_> suzhe: heh. `osascript -e 'tell application "Safari" to activate'` somehow does it tho
[08:42:45] <pamg> Does anyone know why the tree has been closed for 8 hours?
[08:43:17] *** pfeldman has quit IRC
[08:43:40] <suzhe> thakis_: how to start browser_tests with this method?
[08:44:30] *** Erkan_Yilmaz_ has quit IRC
[08:45:53] *** fta_ has joined #chromium
[08:48:00] *** fta has quit IRC
[08:48:05] *** fta_ is now known as fta
[08:49:20] *** star-affinity has joined #chromium
[08:49:28] *** estes has quit IRC
[08:50:50] <trungl> hmmm, lots of badness on the tree
[08:51:29] <thakis_> suzhe: could try `NSApplication activateIgnoringOtherApps:YES`
[08:51:45] <suzhe> thakis_: let me try it.
[08:51:59] <pamg> Hi trungl. Any ideas why the tree is closed? I mean, it's clearly red, but is there any info about what's important there?
[08:52:36] <pamg> Unless someone objects, I'm going to land a build fix for ARM.
[08:52:38] <suzhe> thakis_: I need to switch branch and build again :(
[08:52:59] <trungl> pamg: I don't know, I'm looking at the tree right now
[08:53:11] <trungl> pamg: landing a build fix for arm sounds good to me
[08:53:25] <thakis_> pamg: http://echelog.matzon.dk/logs/browse/chromium/1278972000
[08:53:32] <thakis_> looks lke nobody knows what's wrong
[08:53:38] <thakis_> mass clobbering might or might not help
[08:53:42] *** ebraminio has quit IRC
[08:54:05] <pamg> I'll land this fix, then, and we can see if another run fixes anything.
[08:54:25] *** estes has joined #chromium
[08:54:42] <pamg> The browser_tests at least look maybe flaky.
[08:55:11] <thakis_> i'm off to bed
[08:55:14] <thakis_> good night everyone
[08:55:16] *** thakis_ has quit IRC
[08:55:38] *** eseidel has quit IRC
[08:55:44] <trungl> pamg: browser_tests Toolstrip seems to be failing consistently right now
[09:00:16] *** FullFlannelJacke has quit IRC
[09:00:22] <trungl> (also looks like a consistent perf regression on startup tests)
[09:02:41] <trungl> ExtensionBrowserTests.Toolstrip was already disabled for Linux Views
[09:07:21] *** eseidel has joined #chromium
[09:09:02] <trungl> i wonder if one of the .grd changes didn't delete a bitmap resource that's used by that test
[09:10:45] <trungl> (or moved, or something)
[09:10:48] <pamg> Hmmm.
[09:10:52] <trungl> it wouldn't shock me if it were r52144
[09:11:50] *** pdelgallego has joined #chromium
[09:14:37] <pamg> trungl: You're in MTV now?
[09:14:44] <trungl> pamg: yes
[09:14:51] <trungl> well, at home
[09:14:53] *** apavlov has joined #chromium
[09:14:59] <trungl> (in SF)
[09:15:10] <pamg> In it's-midnight-there, anyway.
[09:15:58] <pamg> Do you particularly want to track down this Toolstrip failure? 'Cause it's 9 AM here and I just bumped my feature to M7, so I can do it. :)
[09:16:25] <trungl> pamg: I probably should get some sleep, so you're welcome to it. :)
[09:17:24] <trungl-bot> Tree closed by pam at chromium dot org (:pamg): Tree is closed (ARM should cycle green; Toolstrip -> pamg)
[09:18:08] <trungl> pamg: Thanks and good luck with that. Sorry about leaving you guys with a broken tree. :(
[09:18:26] <pamg> Still need someone to look at the layout test problems, failing to complete.
[09:18:56] <trungl> That seems to have been broken since even longer.
[09:19:04] <pamg> Yup.
[09:19:21] <pamg> Looks like the non-Pacific sheriff is in London this time, so he should be coming along before too much longer too.
[09:21:44] <trungl> he'll have his work cut out for him
[09:21:49] <trungl> anyway, I'm off
[09:21:52] <trungl> 'night
[09:21:56] *** trungl has quit IRC
[09:23:37] *** Omen_20 has quit IRC
[09:30:42] *** estes has quit IRC
[09:32:03] *** Omen_20 has joined #chromium
[09:35:51] *** pathorn has quit IRC
[09:36:38] <trungl-bot> Tree closed by pam at chromium dot org (:pamg): Tree is closed (ARM should cycle green; Toolstrip -> pamg; webkit_tests -> ??? (needs http://trac.webkit.org/changeset/63087))
[09:36:40] *** awong__ has joined #chromium
[09:37:14] *** awong__ has quit IRC
[09:37:32] *** awong__ has joined #chromium
[09:37:56] *** awong__ has left #chromium
[09:38:04] *** awong__ has joined #chromium
[09:38:15] *** awong__ is now known as awong_home
[09:38:38] <awong_home> pamg: ping
[09:38:54] <pamg> awong_home: pong
[09:39:21] <awong_home> I was thinking of attempting a big webkit roll...
[09:39:44] <awong_home> any chance that the tree will open?
[09:40:03] <pamg> You'd have to ask a sheriff. ;)
[09:40:08] <awong_home> hah
[09:40:43] <pamg> OK, well. ARM should be good. I'm working on the browser_tests. The webkit_tests will hopefully be fixed as long as you roll in 63087 too.
[09:41:09] <pamg> There's a perf regression in the startup tests, it looks like. Would be nice to have a handle on that.
[09:42:01] *** rniwa has joined #chromium
[09:42:30] <awong_home> hmm...this webkit roll is 309 revisions.
[09:42:49] <awong_home> if startup tests is acting odd, I'm less inclined to attempt it
[09:43:14] <pamg> Yow. Can it be broken up?
[09:43:22] *** bauerb has joined #chromium
[09:43:22] *** ChanServ sets mode: +v bauerb
[09:43:23] <awong_home> nope.
[09:43:30] <awong_home> been trying to roll this sucker since thursday. :(
[09:44:17] <pamg> You could improve your chances by investigating the startup_tests issue. :)
[09:44:41] <pamg> I'm trying to see if we can pull in that WebKit revision singly, since it's blocking every WK bot.
[09:45:19] <awong_home> so the big roll has passed all thre try servers
[09:45:21] *** [Nef] has joined #chromium
[09:45:26] <awong_home> http://codereview.chromium.org/2958006/show
[09:45:38] <awong_home> almost
[09:46:03] <awong_home> I'm not sure how you'd pull in a single webkit revision.
[09:46:26] <awong_home> I don't think I can debug startup_tests. All I've got is a little personal mac laptop here :-/
[09:46:26] <hbono> pamg: by the way, is it possible to manually install irclib.py to fix the bots?
[09:46:52] <pamg> hbono: I don't know. I'm guessing not, since they appear to reinstall it every run.
[09:47:02] <pamg> This external dependency makes me sad.
[09:47:33] <awong_home> Monday PST was an utter mess for the tree :(
[09:47:44] *** alt-dot-net-geek has quit IRC
[09:49:04] *** iamdharma has quit IRC
[09:49:45] <hbono> pamg: thank you.
[09:50:58] <pamg> Ah, startup_test is not a perf problem, it's actually failing its tests. :/
[09:51:05] <awong_home> yeah
[09:51:47] * pamg pushes it onto the stack.
[09:53:23] *** Beetny` has joined #chromium
[09:55:31] *** Beetny has quit IRC
[09:56:35] <pamg> Unfortunately I have to go for 90 min or so. Hopefully Sheriff  hwennborg will be arriving soon.
[09:57:03] <awong_home> yeah
[09:57:15] <awong_home> later!
[09:57:40] <awong_home> I'm not able to stay up for 3+ hours, so I don't think I'm going to stab at the webkit roll either :(
[09:58:02] <pamg> Sorry. Good luck tomorrow!
[09:58:19] <awong_home> BTW, startup_tests error I think is GRD related.
[09:58:33] <awong_home> everything dying on a resource_bundle.cc check failure.
[09:59:00] <pamg> OK, it may be fixed when i fix Toolstrips then.
[09:59:12] <pamg> (when, if, whatever...)
[09:59:21] <awong_home> heh
[09:59:23] <awong_home> good luck
[09:59:27] <awong_home> I'm going to go sleep
[09:59:37] <awong_home> will send a note to hwennborg
[09:59:42] <pamg> Thanks.
[10:13:53] *** fta has quit IRC
[10:16:18] <awong_home> anyone still awake?
[10:16:21] <awong_home> or around?
[10:16:35] <awong_home> I'm tempted to believe that the startup_test breakage is due to http://src.chromium.org/viewvc/chrome?view=rev&revision=52144.
[10:16:52] <awong_home> wanted to know if anyone's got a windows box that could test reverting that.
[10:17:52] <pamg> awong_home: That's a huge grd change, looks like it also broke Toolstrips. I'd like to see if we can fix ratehr than revert, since it's so big.
[10:18:26] *** pfeldman has joined #chromium
[10:18:35] <awong_home> pamg: right...I was hoping to do a local test to verify if it indeed blew-up those tests.
[10:18:50] <awong_home> As is, I'm only about 80% confident about it breaking startup_tests.
[10:19:39] <pamg> I'll do that when I get back, if nobody else has yet.
[10:19:43] <awong_home> ok
[10:20:01] <awong_home> really going to go to sleep this time. (not useful w/o a windows box...grrr...)
[10:20:20] <trungl-bot> Tree closed by pam at chromium dot org (:pamg): Tree is closed (Toolstrip and startup_tests ->  r52144? ; webkit_tests -> ??? (needs http://trac.webkit.org/changeset/63087))
[10:20:44] <pfeldman> awong: yt?
[10:21:00] <pfeldman> awong_home: yt?
[10:21:07] <awong_home> yeah
[10:21:19] <pfeldman> awong_home: win failure?
[10:21:38] <awong_home> startup_tests at least
[10:21:42] <pfeldman> awong_home: rest of the bots seem green, only strange win bug
[10:21:56] <pfeldman> awong_home: did you try re-sending it to try bots? (does it repro?)
[10:22:06] <awong_home> I have no clue.
[10:22:14] *** fta has joined #chromium
[10:22:15] <awong_home> just started looking at it about 10 minutes ago.
[10:22:19] <pfeldman> awong_home: k
[10:22:23] <pfeldman> awong_home: go to sleep
[10:22:30] <awong_home> :)
[10:22:36] <pfeldman> awong_home: i was on my way to office
[10:22:39] <pfeldman> i think it is flaky
[10:22:54] <awong_home> if you look at the log for startup_tests, it's resource_bundle.cc throwing errors
[10:23:18] <awong_home> looks like a bitmap that was still being reference was removed from the GRD. don't know which one though.
[10:23:31] <awong_home> anyways, good luck
[10:23:42] <pfeldman> heh
[10:24:38] <pfeldman> awong_home:  btw, why didn't you roll earlier?
[10:24:57] *** rniwa has left #chromium
[10:25:00] <awong_home> you mean today?
[10:25:43] <awong_home> there's been a cascade of closures due to grd changes, etc. for the last 2 work days.  and webkit hasn't been rollable for significant portions of the time too due to various bugs, etc.
[10:26:39] <awong_home> finally got things beaten into a rollable state around ~5pm PST.  But the tree was burning, and seems to have kept burning.
[10:27:01] *** eseidel has quit IRC
[10:28:01] *** shenki has quit IRC
[10:32:34] *** janm has joined #chromium
[10:36:11] *** shenki has joined #chromium
[10:37:37] *** ROBOd has joined #chromium
[10:39:04] *** awong_home has left #chromium
[10:42:45] *** hrna has joined #chromium
[10:45:53] <pfeldman> awong_: i'll just roll
[10:46:09] <pfeldman> we are 300 revisions behind
[10:48:20] <Peter`> oh, that figures, I was wondering why some things didn't work in Chromium yet
[10:54:51] <hwennborg> pamg: i'm not sheriff today (that was Fri-Mon)
[11:06:52] *** kashey has quit IRC
[11:07:28] *** kashey has joined #chromium
[11:08:56] *** fta has quit IRC
[11:10:13] *** kashey has quit IRC
[11:10:16] *** tonikitoo has quit IRC
[11:10:28] *** kashey has joined #chromium
[11:10:58] *** chrisccoulson has joined #chromium
[11:15:21] *** mitchell687 has quit IRC
[11:16:28] *** FullFlannelJacke has joined #chromium
[11:20:05] *** allanwoj has joined #chromium
[11:20:16] *** General13372 has joined #chromium
[11:20:20] *** fta has joined #chromium
[11:22:56] *** GeekShadow has joined #chromium
[11:23:53] *** General1337 has quit IRC
[11:24:24] *** tonikitoo has joined #chromium
[11:25:08] *** podivilov has joined #chromium
[11:26:09] *** podivilov has left #chromium
[11:32:17] *** dr_win has quit IRC
[11:37:33] *** anddam has joined #chromium
[11:39:18] <anddam> hello
[11:41:35] <anddam> I joined youtube html5 beta but still Chrome shows videos using Flash, is that due to builtin flash plugin?
[11:42:50] *** yuzo has quit IRC
[11:42:50] <jochen__> see topic please
[11:43:17] <anddam> my bad
[11:46:37] <anddam> bye
[11:46:38] *** anddam has left #chromium
[11:47:26] *** avinashhebbur has quit IRC
[11:47:28] *** dr_win has joined #chromium
[11:47:49] *** avinashhebbur has joined #chromium
[11:48:18] *** avinashhebbur has left #chromium
[11:50:53] *** fta has quit IRC
[11:52:05] *** fta has joined #chromium
[11:58:48] *** obscurant1st has joined #chromium
[11:58:49] <obscurant1st> somebody know how to pack in crx for chrome themes in linux via terminal?
[12:04:28] *** vithos has quit IRC
[12:07:03] *** MikeSmithW3C has quit IRC
[12:09:35] *** vithos has joined #chromium
[12:14:43] *** mnaganov has quit IRC
[12:16:57] *** phrearch has joined #chromium
[12:16:59] <phrearch> hi
[12:17:28] <phrearch> how would i setup a cookie with chromium's websocket handler?
[12:17:42] <phrearch> i would like to setup subprotocol and a cookie for authentication
[12:17:49] <phrearch> or rather use a session if possible
[12:22:33] *** mnaganov has joined #chromium
[12:24:43] *** c_zahmad has quit IRC
[12:27:50] *** MikeSmithW3C has joined #chromium
[12:31:08] *** solsTiCe has joined #chromium
[12:31:42] *** Al2O3 has quit IRC
[12:32:17] <solsTiCe> hi. html5 video is still broken. here http://camendesign.com/code/video_for_everybody I think I am using ogg version and sound seems to stop and restart every second but playing continuasly. that make is very strange. video seems ok
[12:36:45] *** ROBOd has quit IRC
[12:37:55] *** fta_ has joined #chromium
[12:38:12] *** ROBOd has joined #chromium
[12:39:18] *** fta has quit IRC
[12:39:21] *** fta_ is now known as fta
[12:42:40] *** hbono has quit IRC
[12:47:03] *** solsTiCe has left #chromium
[12:49:12] *** loislo has quit IRC
[12:49:30] *** awolfson has quit IRC
[12:49:46] *** loislo has joined #chromium
[12:49:59] <star-affinity> any idea why this site won't render properly in Chrome, but works in Safari? http://www.ownit.se/
[12:50:06] <star-affinity> is it because of all the Flash elements?
[12:56:12] *** duffydack has joined #chromium
[12:56:18] *** duffydack has left #chromium
[13:05:51] <jorlow> star-affinity: i loaded it.  it was weird.  i refreshed it...now it seems fine
[13:05:57] <jorlow> what version of chrome are you on?
[13:09:10] <star-affinity> jorlow: ok, doesn't get better when I refresh.
[13:09:17] <star-affinity> I'm on Chrome 5.0.375.99 on Mac OS X
[13:09:26] <star-affinity> Flash 10.1
[13:09:53] <jorlow> star-affinity: prob a real bug
[13:09:56] <jorlow> i'm on dev channel
[13:10:03] <jorlow> so my guess it was fixed at some point
[13:10:10] <jorlow> maybe file?  new.crbug.com
[13:10:50] <star-affinity> jorlow: alright ? maybe I should try with the dev. release too and see.
[13:11:01] <jorlow> if you can, that'd be helpful
[13:12:42] *** Sonderblade has joined #chromium
[13:13:29] *** alt-dot-net-geek has joined #chromium
[13:15:08] <star-affinity> jorlow: works fine in the dev. version
[13:15:32] <jorlow> star-affinity: please do file a bug then
[13:19:07] <star-affinity> jorlow: is it sufficient if I use the ?Report Bug or Broken Web Site? from the Help menu?
[13:19:31] *** chrisccoulson has quit IRC
[13:19:45] <jorlow> star-affinity: new.crbug.com will get more attention
[13:19:55] <star-affinity> alright. :)
[13:23:58] *** NthDeGeek has quit IRC
[13:24:12] <star-affinity> jorlow: hmm.. strange - now it loads fine in the stable release too. Maybe restarting Chrome cleared something up?
[13:26:33] *** chrisccoulson has joined #chromium
[13:27:51] *** qinjia has joined #chromium
[13:28:58] <trungl-bot> Tree closed by pam at chromium dot org (:pamg): Tree is closed (Toolstrip and startup_tests ->  r52144? (pamg looking) ; webkit_tests -> ??? (needs http://trac.webkit.org/changeset/63087))
[13:31:05] *** andrix has joined #chromium
[13:35:57] *** tonikitoo has quit IRC
[13:36:12] *** rubenbb has quit IRC
[13:37:26] *** alt-dot-net-geek has quit IRC
[13:43:13] *** rubenbb has joined #chromium
[13:49:50] *** apavlov has quit IRC
[13:51:07] *** fta_ has joined #chromium
[13:51:57] *** fta has quit IRC
[13:52:05] *** fta_ is now known as fta
[13:58:25] *** apavlov has joined #chromium
[14:12:11] *** andrix1 has joined #chromium
[14:12:39] *** lgombos has joined #chromium
[14:13:11] *** romainhuet has joined #chromium
[14:13:35] *** andrix has quit IRC
[14:20:03] *** kashey has quit IRC
[14:21:00] *** kashey has joined #chromium
[14:26:48] *** mnaganov has quit IRC
[14:26:53] *** podivilov1 has joined #chromium
[14:28:05] *** apavlov has quit IRC
[14:29:34] *** apavlov has joined #chromium
[14:30:52] *** n3v3le1 is now known as n3v3le
[14:34:46] *** c_zahmad has joined #chromium
[14:39:45] *** mazda has quit IRC
[14:41:08] *** MX80 has quit IRC
[14:43:54] *** fta_ has joined #chromium
[14:44:54] *** fta has quit IRC
[14:45:04] *** fta_ is now known as fta
[14:47:22] *** urbanape has joined #chromium
[14:49:50] *** AaronMT has joined #chromium
[14:50:25] *** barcon332 has joined #chromium
[14:50:54] *** n3v3le1 has joined #chromium
[14:51:48] <trungl-bot> Tree closed by pfeldman at chromium dot org: Tree is closed (Toolstrip and startup_tests ->  r52144? (pamg looking))
[14:53:40] *** n3v3le has quit IRC
[14:54:11] *** MX80 has joined #chromium
[14:54:57] <pamg> Any gyp experts around? The problem with browser_tests (and many others) is that they're not getting the theme_resources, and some resource moved from app_resources to theme_resources. I haven't yet figured out the gyp magic to fix it.
[14:58:39] *** qinjia has quit IRC
[15:12:03] *** agl has joined #chromium
[15:12:03] *** ChanServ sets mode: +v agl
[15:13:11] *** xenor has joined #chromium
[15:13:30] *** rvargas has quit IRC
[15:13:33] *** happygrue has quit IRC
[15:13:59] *** happygrue has joined #chromium
[15:20:36] *** xenor has left #chromium
[15:33:14] *** happygrue_ has joined #chromium
[15:36:43] *** happygrue has quit IRC
[15:37:17] *** mazda has joined #chromium
[15:42:52] *** bent-mozilla has joined #chromium
[15:45:27] *** seventh has quit IRC
[15:45:58] *** AryehGregor has quit IRC
[15:50:36] *** aroben has joined #chromium
[15:50:49] *** aroben has quit IRC
[15:50:49] *** aroben has joined #chromium
[15:59:04] *** dmaclach has quit IRC
[15:59:42] *** zer0her0 has joined #chromium
[15:59:52] *** AryehGregor has joined #chromium
[16:03:22] <agl> pamg: I have a GRD change pending which I would like to get in before MTV are too active. Could you leave the tree closed when you're done?
[16:06:47] <pamg> agl: Sure.
[16:06:52] <nkostylev> agl: I have pending GRD change too - browser_resources.grd
[16:07:26] <agl> nkostylev: want to land them both? Should save some time.
[16:07:40] <agl> nkostylev: I think we can land <GRD 1> <GRD 2> <ws change>
[16:07:55] *** awidegreen has joined #chromium
[16:08:34] <nkostylev> agl: makes sense
[16:09:12] *** rsesek has joined #chromium
[16:09:12] *** ChanServ sets mode: +v rsesek
[16:09:25] *** awolfson has joined #chromium
[16:10:46] <nkostylev> agl: I'll submit my change once your GRD1 is fine
[16:12:39] <pamg> I'm just waiting for browser_tests to clear, to make sure the revert fixed the error.
[16:12:58] <agl> I'll land once trungl-bot says that the tree status has changed.
[16:13:08] <pamg> Oof, Vista not looking good. Sigh.
[16:13:44] <pamg> Oh, right. The rollback was another grd change. agl, you might as well put in all your grd changes and we'll proof them all at once.
[16:14:28] <pamg> If you ever work in an EMEA office, don't admit you can help fix a broken tree. :(
[16:14:28] <agl> pamg: ok, landing now.
[16:14:40] *** skerner has quit IRC
[16:14:47] <agl> landed
[16:15:18] <nkostylev> waiting for your change to go green, will land mine after
[16:16:13] *** alt-dot-net-geek has joined #chromium
[16:18:56] <gavinp> can anyone talk buildbots with me for a second?
[16:19:12] <gavinp> i'm getting tryfailures that look a lot like a buildbot is using an old webkit.  is this possible?
[16:19:14] *** skerner has joined #chromium
[16:19:15] *** ChanServ sets mode: +v skerner
[16:19:35] <agl> gavinp: what's the URL of the failure?
[16:19:50] *** Kunalagon has joined #chromium
[16:21:12] *** fta_ has joined #chromium
[16:21:36] *** js2 has joined #chromium
[16:21:42] *** fta has quit IRC
[16:21:52] *** fta_ is now known as fta
[16:22:57] <trungl-bot> Tree closed by pam at chromium dot org (:pamg): Tree is closed (grd revert and changes landing -> agl)
[16:23:58] <trungl-bot> Tree closed by nkostylev at chromium dot org: Tree is closed (grd revert and changes landing -> agl, nkostylev)
[16:26:04] <nkostylev> agl: revert needed
[16:26:24] <agl> nkostylev: no, I just need to fix ChromeOS
[16:26:38] <nkostylev> agl: ok then
[16:27:15] *** hebz0rl has joined #chromium
[16:27:41] <nkostylev> agl: Chromium Builder is green. Should I wait for Chromium XP to go green too before landing my grd change?
[16:28:21] <agl> nkostylev: I would land it now
[16:28:38] <nkostylev> agl: submitting
[16:31:01] *** duffydack has joined #chromium
[16:32:26] *** janm has quit IRC
[16:34:53] <nkostylev> landed
[16:34:56] *** trungl has joined #chromium
[16:34:57] *** ChanServ sets mode: +v trungl
[16:35:12] <trungl> 'morning, Chromium.
[16:35:13] <trungl-bot> trungl: Good morning!
[16:35:30] *** karpar has joined #chromium
[16:37:05] *** alt-dot-net-geek has quit IRC
[16:40:07] *** thakis_ has joined #chromium
[16:40:07] *** ChanServ sets mode: +v thakis_
[16:40:31] 
[16:40:46] <thakis_> trungl: !
[16:41:13] <trungl-bot> Tree closed by pfeldman at chromium dot org: Tree is closed (grd revert and changes landing -> agl, nkostylev, webkit_tests -> ??? (needs http://trac.webkit.org/changeset/63204))
[16:42:14] *** duffydack has left #chromium
[16:43:14] *** stuartmorgan has joined #chromium
[16:43:14] *** ChanServ sets mode: +v stuartmorgan
[16:43:34] *** thakis_ has quit IRC
[16:47:18] <mabbo> can anyone familiar with Skia explain to me what is so awesome about RESAMPLE_AWESOME?
[16:49:09] <agl> mabbo: it's slower and higher quality
[16:49:20] <gavinp> agl: I am sorry for wandering off
[16:49:40] <mabbo> agl: better than linear resampling, then
[16:49:41] <gavinp> i figured out my problem: trybots were using a last known good chromium, and the deps file there specified a too-old webkit
[16:50:04] <gavinp> so i'm now trying to specify a webkit rev manually to git try, and mostly failing and just coming to dislike git try
[16:50:15] <agl> mabbo: yes, Brett would know the details, but it's a little early PST right now.
[16:50:45] <gavinp> so either my git try -s third_party/WebKit@xxxx will work for me, or I'll just wait for lkg to include a recent webkit
[16:50:47] <gavinp> and i'll be good
[16:50:59] <trungl> gavinp: if there's a more recent revision of chromium which includes that webkit change, then it's easy to specify that
[16:51:16] *** janm has joined #chromium
[16:52:20] <gavin>  /forcenick gavinp bill
[16:53:48] <trungl> gavinp: otoh, it looks to me like -s third_party/WebKit@... specifies a branch, not a revision
[16:54:40] *** malavv has joined #chromium
[16:54:51] *** andybons has joined #chromium
[16:54:51] *** ChanServ sets mode: +v andybons
[16:54:54] <gavinp> trungl, hrm.  so perhaps i am doing it wrong.
[16:55:01] *** elmargol has quit IRC
[16:55:20] <bauerb> gavinp: i think it should work with both
[16:55:34] *** rohitkc has joined #chromium
[16:55:55] <bauerb> trychange.py says "Use @rev or @branch or @branch1..branch2 to specify the revision/branch to diff against."
[16:56:11] <gavinp> bauerb, it seems not, at least the gclient output on tryserver is not showing a happy rev of webkit.
[16:56:26] <gavinp> i'll try trungl's suggestion
[16:57:14] <andybons> estade, oshima: grd change coming when the tree opens up, though I don't think it should break anything since it's simply adding an <if> to choose between two different entries. therefore it shouldn't offset any of the generated numbers
[16:57:17] <trungl> gavinp: if a more recent revision of chromium has a suitably new DEPS, then -r <cr_rev> should do the trick
[16:57:19] *** iamdharma has joined #chromium
[16:57:29] *** elmargol has joined #chromium
[16:58:00] <gavinp> trungl, yeah.  i am doing exactly that, with the rev that bumped the webkit deps to what makes me happy
[16:58:31] *** dmaclach has joined #chromium
[16:58:34] <andybons> estade, oshima: review is at http://codereview.chromium.org/2973003/show. Mac-only change. Trybot failures on other platforms don't seem related at all.
[16:58:38] <trungl> (the risk, evidently, is that some things will probably fail because you're not testing on a green revision)
[16:58:51] <gavin> pnivag? inpgav? gavip?
[16:58:56] <trungl> though it's not like lkgr guarantees greenness anyway
[16:59:09] <trungl> gavin: what's your problem?
[16:59:14] * gavinp confused
[16:59:17] <gavinp> what is it, gavin?
[16:59:18] <gavin> sorry, I'm just kidding around
[16:59:26] <gavin> everytime someone says "gavinp" it pings me
[16:59:36] <gavinp> oh.
[16:59:58] <trungl> you should consider changing your regex (or whatever)
[17:00:26] <gavinp> i reccomend gavin\($\|[^p]\)
[17:00:38] <trungl> (or just getting rid of the alert)
[17:00:39] *** shoe has quit IRC
[17:00:52] <gavin> thanks for the tip! :)
[17:00:56] <gavinp> gavin, unfortuantely i can't really change my userid here
[17:01:06] *** shoe has joined #chromium
[17:01:07] <gavinp> i want it to be the same as my chromium.org userid
[17:01:30] <gavin> no worries
[17:06:18] *** rohitrao has joined #chromium
[17:06:18] *** ChanServ sets mode: +v rohitrao
[17:06:27] <nkostylev> agl: both bots are green, I'm sending ws change
[17:07:09] *** tonyg-cr has joined #chromium
[17:07:10] *** ChanServ sets mode: +v tonyg-cr
[17:07:18] <gavinp> trungl, git try -r seems happier than the -W and -s attempts from earlier.
[17:08:06] *** n3v3le has joined #chromium
[17:08:34] <agl> nkostylev: ok
[17:09:00] <trungl> gavinp: I've actually used -r before (successfully). ;)
[17:09:42] *** n3v3le1 has quit IRC
[17:09:45] *** mrossetti has joined #chromium
[17:10:42] <trungl> one could presumably also include a suitable change to DEPS in the change one is trying, but the considerations are a bit more complicated (since you'd need the patch to apply)
[17:11:29] <gavinp> oh, sure
[17:11:31] <gavinp> yeah
[17:12:01] *** mrossetti has quit IRC
[17:12:46] *** Erkan_Yilmaz has joined #chromium
[17:12:49] <rsesek> does anyone know when pink comes back?
[17:13:25] <trungl> pink won't be in fashion for another decade, I'm afraid
[17:13:40] <trungl> but you might check pinkerton's calendar
[17:13:44] <rsesek> :( and I just bought a bright pink tux
[17:13:48] <rsesek> it's not on his calendar
[17:14:12] <trungl> d'oh
[17:14:34] <trungl> he's scheduled to be sheriff next Monday and Tuesday
[17:15:25] <trungl> rsesek: actually, he should be back tomorrow
[17:15:27] *** [Nef] has quit IRC
[17:15:32] <rsesek> cool
[17:16:32] *** Precea[BNC] is now known as Precea
[17:16:33] *** pfeldman has quit IRC
[17:17:31] <mabbo> agl: thanks. To be clear, is that brettw you refer to as Brett?
[17:17:38] <agl> mabbo: yes
[17:18:29] * trungl thinks the awesomeness is just in the name. Clever marketing.
[17:18:45] <mabbo> agl: great. I don't really need to know so much about the awesome resampling itself however, just a bug relating to it. By chance do you know if, when the resampling is chosen, is it possible to find out if a scale transform will be applied later?
[17:19:39] <mabbo> agl: in short, right now we're resampling down images, then scaling them back up later. It makes things less pretty than basic linear does.
[17:20:05] <agl> mabbo: That knowledge is probably in WebKit when it does the sampling but, to be honest, I don't really know that code.
[17:20:16] *** n3v3le has left #chromium
[17:20:29] *** aroben has quit IRC
[17:20:57] <mabbo> agl: heh, thanks anyways. I'll talk to brettw, and dig deeper into this. Not a complex bug, just don't want to shake anything loose by accident.
[17:22:37] *** aroben has joined #chromium
[17:22:40] *** aroben has quit IRC
[17:22:40] *** aroben has joined #chromium
[17:22:57] *** agayev has joined #chromium
[17:26:41] *** kliegs has quit IRC
[17:27:15] *** trungl has quit IRC
[17:29:10] *** jrmuizel has joined #chromium
[17:31:38] *** RobWork2 has quit IRC
[17:31:48] *** star-affinity has quit IRC
[17:31:57] *** tonikitoo has joined #chromium
[17:33:40] *** mrossetti has joined #chromium
[17:35:08] *** joth_ has joined #chromium
[17:35:08] *** ChanServ sets mode: +v joth_
[17:36:25] *** Venom_X has joined #chromium
[17:37:04] *** RobWork has joined #chromium
[17:37:15] *** tedoc2000 has joined #chromium
[17:39:57] *** dave_levin has joined #chromium
[17:39:58] *** ChanServ sets mode: +v dave_levin
[17:39:59] *** bent-mozilla has quit IRC
[17:41:52] *** tfarina has joined #chromium
[17:43:54] *** duffydack has joined #chromium
[17:48:38] *** karpar has quit IRC
[17:50:13] *** pseudomonas has joined #chromium
[17:50:52] *** duffy has joined #chromium
[17:51:16] *** duffydack has quit IRC
[17:54:59] *** pseudomonas is now known as pseudomonas|away
[17:55:12] <fta> who's doing the new menu edit/zoom controls on linux?
[17:56:08] <fta> http://people.ubuntu.com/~fta/chromium-menu-artifacts.ogv
[17:56:10] <rsesek> erg
[17:56:35] <rsesek> fta: ^
[17:57:20] <fta> just wanted to know if it's a known bug or if i should file one
[17:57:25] <rsesek> do not know
[17:57:37] *** awong_home has joined #chromium
[17:58:17] *** masterov has quit IRC
[17:58:17] <agl> nkostylev: good to remove the GRD bit from the tree status?
[17:58:28] *** mazda has quit IRC
[17:59:30] *** mazda has joined #chromium
[17:59:30] *** ChanServ sets mode: +v mazda
[18:00:00] *** rohan has joined #chromium
[18:00:09] <rohan> could someone please check if this bug repros? http://code.google.com/p/chromium/issues/detail?id=45968
[18:03:01] <nkostylev> agl: Chromium Builder and Chromium Builder (dbg) are green - ok?
[18:03:27] <pcgod> rohan: there is no black background on windows but the button is sometimes invisible
[18:04:09] <rohan> pcgod: yes, one way or the other, there is a visual problem with the button
[18:04:19] *** johnny_g has joined #chromium
[18:04:42] *** duffy has quit IRC
[18:04:52] <pcgod> rohan: that button is a flash object...
[18:05:49] <agl> nkostylev: yea, I think we're good
[18:07:34] *** awong_home has quit IRC
[18:07:37] *** magn3ts has joined #chromium
[18:07:49] <magn3ts> Anyone using Chromium nightlies with Kerebos auth?
[18:08:04] <agl> (we support Kerberos?)
[18:08:20] <magn3ts> Uh, >_> I think so?
[18:08:54] <agl> oh, GSS-API, sure
[18:09:02] <magn3ts> also Kerberos :o
[18:09:22] <rohan> pcgod: oh ok.. is it to be expected then?
[18:09:24] <trungl-bot> Tree closed by nkostylev at chromium dot org: Tree is closed (webkit_tests -> ??? (needs http://trac.webkit.org/changeset/63204))
[18:09:32] <rohan> somehow, even firefox 4 beta is having the same problem
[18:09:33] *** Barkhorn has joined #chromium
[18:09:49] <agl> magn3ts: I suspect that none of the developers use it. I wouldn't be surprised if it's broken.
[18:09:56] <magn3ts> hehe
[18:10:04] <magn3ts> I wondered as much :)
[18:10:48] <pcgod> rohan: probably not but that explains why it happens only for that button and not for the others
[18:12:16] *** agayev has quit IRC
[18:12:24] *** agayev has joined #chromium
[18:13:26] <agayev> magn3ts: i got disconnected and didn't hear the rest of the conversation, but if you have a problem with kerberos auth, let me know
[18:13:30] *** nkostylev_ has joined #chromium
[18:13:39] *** [Nef] has joined #chromium
[18:14:00] *** Zaba has quit IRC
[18:14:02] *** markmentovai has joined #chromium
[18:14:02] *** ChanServ sets mode: +v markmentovai
[18:14:16] <magn3ts> agayev, I can do a bit of checking but I do believe that is whats causing the crashing I'm seeing. It's when I hit several of our corp sites.
[18:14:33] <magn3ts> and it takes down the whole browser quite sadly :(
[18:14:50] *** nkostylev_ has quit IRC
[18:15:49] <agayev> does it happen with the latest dev build?
[18:16:03] *** dmaclach has quit IRC
[18:16:14] *** dmaclach has joined #chromium
[18:16:17] *** ChanServ sets mode: +v dmaclach
[18:16:41] *** SkyLined2 has joined #chromium
[18:16:42] *** duffy has joined #chromium
[18:17:44] *** duffy has quit IRC
[18:21:16] *** SkyLined3 has quit IRC
[18:21:34] *** cbentzel has joined #chromium
[18:21:49] <agayev> cbentzel: magn3ts has problems with kerberos-auth
[18:22:51] <cbentzel> mang3ts: Sorry about the kerberos-auth issue. For now, you can workaround the crash with --disable-auth-negotiate-cname-lookup
[18:23:06] <cbentzel> magn3ts: The next dev channel release should include the bug fix
[18:23:26] *** jshin has joined #chromium
[18:23:26] *** ChanServ sets mode: +v jshin
[18:23:41] <magn3ts> cbentzel, fantastic! I just wanted to make sure it was a known bug and you've got workaround details and a fix. :)
[18:24:57] *** Zaba has joined #chromium
[18:30:05] *** erikkay has joined #chromium
[18:30:05] *** ChanServ sets mode: +v erikkay
[18:30:13] *** MikeSmithW3C has quit IRC
[18:30:33] *** bauerb has quit IRC
[18:30:39] *** loislo has quit IRC
[18:31:19] *** obscurant1st has quit IRC
[18:31:50] *** masterov has joined #chromium
[18:32:12] *** duffydack has joined #chromium
[18:34:51] *** apavlov has quit IRC
[18:35:13] *** MikeSmithW3C has joined #chromium
[18:35:48] *** malavv has quit IRC
[18:37:11] *** MichealH has joined #chromium
[18:38:10] *** paul_uk has joined #chromium
[18:39:00] <markmentovai> estade, jam2, oshima: what?s the story here with the tree?
[18:39:24] <paul_uk> hi all, is there a known problem with chrome for linux 6.0.457.0?  It doesnt seem able to create sessions for multiple subdomains in php.
[18:39:30] *** barcon332 has quit IRC
[18:39:37] <markmentovai> nkostylev: you closed it, you got any ideas?
[18:39:50] *** cbentzel has quit IRC
[18:39:54] <oshima> markmentovai: i just came in, and trying to understand.
[18:40:03] *** jamesr has joined #chromium
[18:40:03] *** ChanServ sets mode: +v jamesr
[18:40:05] <markmentovai> me too.
[18:41:02] <pamg> I think the layout tests are OK now, and the tree can be opened.
[18:41:13] <pamg> It's been closed for about 20 hours now.
[18:41:18] <markmentovai> i suspect that?s the case too, but we need to go metered.
[18:42:02] *** Sarten-X has quit IRC
[18:42:27] <pamg> The problem with the layout tests was a dead sourceforge server, which was preventing the autocheckout of... some Python module the WebKit scripts use.
[18:42:42] <oshima> what about webkit (dbg)(2)?
[18:42:45] <markmentovai> yup, i remember that from yesterday. webkit needs to become more resilient about that.
[18:42:49] <nkostylev> markmentovai: I've just landed grd change
[18:42:58] <nkostylev> nothing changed more from my side
[18:43:01] <pamg> The referenced WebKit changeset changed the download server. It looks like maybe the old one is back up now though.
[18:43:06] <markmentovai> nkostylev: does that mean that you no longer have a reason to hold the tree closed?
[18:43:09] <oshima> looks like it's started after r52170
[18:43:50] <nkostylev> it could be opened back webkit_tests are the only thing that is red
[18:44:00] <oshima> pfeldman: i'm going to revert r52170
[18:44:04] <jamesr> oh i see
[18:44:12] <jamesr> err, wrong chan
[18:44:32] <markmentovai> oshima: hang on
[18:44:38] <markmentovai> awong: are you the webkit gardener today?
[18:44:45] <oshima> markmentovai: ok
[18:46:05] *** fta_ has joined #chromium
[18:46:21] <markmentovai> oshima: ok, apparently we're not hearing from pfeldman or awong
[18:46:53] *** fta has quit IRC
[18:46:56] <oshima> markmentovai: ok, i'll revert.
[18:47:02] *** fta_ is now known as fta
[18:47:12] <podivilov1> please don't do it
[18:47:16] *** janm has quit IRC
[18:48:39] <oshima> sorry, i've just done it.
[18:48:51] <oshima> can you explain?
[18:48:58] <markmentovai> it?s versioned. we?ll see what happens.
[18:49:11] *** Aria has joined #chromium
[18:49:51] <markmentovai> oshima: while you?re waiting for this, you might want to look at the memory waterfall, there seems to be a good deal of redness over there too.
[18:50:20] <markmentovai> i?m going to lunch, i?ll be back later if you?d like help greening things up if your co-sheriffs don?t show.
[18:50:31] <oshima> markmentovai: will do
[18:50:56] <trungl-bot> Tree closed by oshima at chromium dot org: Tree is closed (r52170 reverted. waiting for next cycle)
[18:50:59] *** RT|Chatzilla has quit IRC
[18:51:42] *** agl has quit IRC
[18:52:32] *** Barkhorn has left #chromium
[18:53:01] *** bweinstein has quit IRC
[18:53:34] *** malavv has joined #chromium
[18:57:19] *** rryk has joined #chromium
[18:57:22] *** jschuh has joined #chromium
[18:57:52] *** rryk has quit IRC
[18:58:26] *** jeremyselier has joined #chromium
[18:59:29] *** jeremyselier has quit IRC
[19:00:00] *** jeremyselier has joined #chromium
[19:00:02] *** janm has joined #chromium
[19:00:40] *** jeremyselier has quit IRC
[19:01:11] *** jeremyselier has joined #chromium
[19:01:23] *** allanwoj has quit IRC
[19:01:51] *** jeremyselier has quit IRC
[19:02:22] *** jeremyselier has joined #chromium
[19:02:37] <awong> markmentovai: I'm back now
[19:02:54] <awong> what's going on?
[19:03:02] *** jeremyselier has quit IRC
[19:03:33] *** jeremyselier has joined #chromium
[19:04:13] *** jeremyselier has quit IRC
[19:04:44] *** jeremyselier has joined #chromium
[19:05:24] *** jeremyselier has quit IRC
[19:06:04] *** ebraminio has joined #chromium
[19:06:13] <awong> I see we reverted teh webkit roll.  Did we think it was responsible for the browser_Tests failure?  or was it the webkit (dbg) failure that signaled the revert?
[19:06:14] *** jeremyselier has joined #chromium
[19:06:38] *** jeremyselier has quit IRC
[19:06:41] <oshima> awong: i reverted r52170 and we're waiting now.
[19:07:06] *** jeremyselier has joined #chromium
[19:07:29] *** rvargas has joined #chromium
[19:07:29] *** ChanServ sets mode: +v rvargas
[19:07:45] *** jeremyselier has quit IRC
[19:08:06] <awong> oshima: okay. which failure is the revert targeting? both?
[19:08:17] *** jeremyselier has joined #chromium
[19:08:49] <oshima> webkit_tests
[19:08:56] *** jeremyselier has quit IRC
[19:09:04] <andybons> awong: browser tests have been flaky for a while
[19:09:11] <oshima> awong: podivilov1 explained to me that it's been fixed in upstream, but it was after i reverted
[19:09:17] <awong> ic
[19:09:28] *** jeremyselier has joined #chromium
[19:09:34] <oshima> please go ahead with new roll once it turns green.
[19:10:07] *** jeremyselier has quit IRC
[19:10:15] <awong> oshima: sounds good.  Given how broken things have been, should we throttle commits after the roll?
[19:10:30] <awong> I don't want to rollback again if possible.
[19:10:41] *** jeremyselier has joined #chromium
[19:11:06] <oshima> awong: webkit linux are failing now :(. are these now depends on newer version?
[19:11:38] <awong> yes
[19:11:38] <awong> when you unroll that change, most of the webkit bots will fail.
[19:11:38] <awong> :-/
[19:11:38] <andybons> awong, oshima: throttling is a good idea. the tree has been closed for a while now.
[19:11:42] <oshima> or just flaky
[19:11:42] <oshima> ok
[19:12:01] <awong> let me prep a new roll with podivilov1's fixes upstream
[19:12:14] <oshima> ok, let's roll new one then.
[19:12:21] *** kashey has quit IRC
[19:13:04] *** duffydack has quit IRC
[19:13:23] <oshima> andybons: i haven't done it before. how does it work?
[19:13:41] <oshima> awong: is new webkit change ready?
[19:14:01] <awong> oshima: not yet. trying to figure out what change I should go to.
[19:14:03] <andybons> oshima: just look at the tree. if the bots fall behind then simply close to allow for a catchup.
[19:14:51] <awong> oshima: nevermind, I'll just redo the exact roll.  We'll still fail 2 tests in WebKit dbg (2), but I'll fix that on the next roll.
[19:15:33] *** jamesr has quit IRC
[19:15:42] *** csilv has joined #chromium
[19:15:42] *** ChanServ sets mode: +v csilv
[19:15:53] <oshima> awong: ok, so i'll open the tree.
[19:16:04] <awong> oshima: I think we should keep it closed actually...
[19:16:31] <awong> and put a notice to ask on irc for permission to commit, so we can control the rate of commits...
[19:16:31] <oshima> awong: oh, ok.
[19:17:07] <andybons> oshima: permission to commit http://codereview.chromium.org/2973003/show
[19:17:21] <trungl-bot> Unknown tree status set by oshima at chromium dot org: Tree is close (webkit -> awong, ask on irc for permission to commit)
[19:18:11] *** mazda has quit IRC
[19:18:15] <andybons> oshima: it's grd change, but shouldn't break anything due to the condiitonal.
[19:18:31] <awong> andybons: won't that trigger a clobber of all bots anyways?
[19:18:38] <thakis> how a bout grd change ban during business hours? :-)
[19:18:50] <thakis> at least after a 20h tree closure
[19:19:10] *** ph0nk has joined #chromium
[19:19:11] *** mazda has joined #chromium
[19:19:11] *** ChanServ sets mode: +v mazda
[19:19:16] <andybons> awong: i don't know.
[19:20:03] <oshima> awong, andybons: it won't trigger. my understanding is that grd change requires clobber on windows, but not on others.
[19:20:06] <oshima> am i wrong?
[19:20:10] <sky_> you're right.
[19:20:36] <oshima> sky_: so this change should be fine right?
[19:20:39] <andybons> plus this won't offset the ID mappings at all
[19:20:48] *** fta_ has joined #chromium
[19:21:18] <awong> reroll is prepped for whenever we're ready
[19:21:30] <oshima> andybons, awong: ok, go ahead.
[19:21:31] *** magn3ts_ has joined #chromium
[19:21:32] <thakis> what's wrong with webkit_tests?
[19:21:35] *** happygrue_ has quit IRC
[19:21:35] *** magn3ts has quit IRC
[19:21:42] <andybons> oshima: thanks.
[19:21:42] <oshima> i'll keep close eye on it.
[19:21:42] *** happygrue has joined #chromium
[19:21:42] <awong> thakis: I need to reroll webkit
[19:21:44] <sky_> oshima: If the resources are not on windows, yes, it should be fine.
[19:21:52] <thakis> awong: i see
[19:22:04] *** fta has quit IRC
[19:22:15] <awong> webkit roll committed.
[19:22:16] *** fta_ is now known as fta
[19:23:06] *** Namegduf has quit IRC
[19:23:25] <trungl-bot> Tree closed by ajwong at chromium dot org: Tree is closed (webkit -> awong, ask on irc for permission to commit)
[19:23:29] <pamg> thakis: Whose business hours? :)
[19:23:42] <thakis> pamg: everyone's :-)
[19:23:43] <nsylvain> waterfall seems pretty slow again.  I'm going to kill the top 3 rows at the top to see if it helps
[19:23:49] <awong> pamg: antarctica's
[19:23:50] *** Namegduf has joined #chromium
[19:24:09] *** thakis_ has joined #chromium
[19:24:09] *** ChanServ sets mode: +v thakis_
[19:24:15] <awong> we need to support our emperor penguin chick committer community.
[19:24:17] <pamg> "No grd changes from May through September. kthxbai"
[19:24:30] <paul_uk> hi all, is there a known problem with chrome for linux 6.0.457.0?  It doesnt seem able to create sessions for multiple subdomains in php.
[19:24:50] <andybons> paul_uk: see topic
[19:24:53] <pamg> Hm, I guess I mean September through May. Tree sheriffing melts my brain.
[19:26:47] *** rohan has quit IRC
[19:26:51] *** bweinstein has joined #chromium
[19:27:58] <rvargas> oshima: what about http://codereview.chromium.org/2945002/show
[19:28:17] *** bweinstein has quit IRC
[19:28:36] *** bweinstein has joined #chromium
[19:29:18] *** cbentzel has joined #chromium
[19:30:51] *** paul_uk has left #chromium
[19:31:30] *** feldstein has joined #chromium
[19:31:30] *** ChanServ sets mode: +v feldstein
[19:31:53] *** ebraminio has quit IRC
[19:33:41] <oshima> rvargas: let me see...
[19:33:50] <awong> oshima: linux compile looks to have failed.
[19:34:05] *** mirandac has joined #chromium
[19:34:06] *** ChanServ sets mode: +v mirandac
[19:34:30] <oshima> awong: ok, let me check.
[19:34:39] <awong> I'm looking at it too
[19:34:40] <oshima> rvargas: please hold.
[19:34:49] <andybons> me too
[19:36:25] *** fta_ has joined #chromium
[19:37:15] *** fishd has joined #chromium
[19:37:15] *** ChanServ sets mode: +v fishd
[19:37:16] <oshima> awong: looks like webkit rool.
[19:37:24] <shess> Is there an URL which provides ONLY the tree status?  Times like now, I'm periodically refreshing in order to see if it's worth checkpointing my current work so I can go check something in.  I can hack something up based on the presubmit checks, but it might be more reasonable to have a direct link from the waterfall.
[19:37:39] *** glotov has joined #chromium
[19:37:45] <awong> oshima: odd...this is just an undo of the roll you had earlier.
[19:38:00] <oshima> shess: git cl tree, if you're using git
[19:38:13] *** fta has quit IRC
[19:38:16] *** fta_ is now known as fta
[19:38:16] <glotov> Hi! May I commit. My trybots are green: http://codereview.chromium.org/2920001/show
[19:38:49] <awong> oshima: crap, I'm off by 1 number.
[19:38:53] <shess> oshima: cool.
[19:38:59] <fishd> hmm... my mac build of chrome won't link
[19:39:29] <andybons> fishd: are you at head?
[19:39:30] <oshima> awong: you mean the revision number?
[19:39:37] <awong> oshima: yeah. I'm going to push one more revision in.
[19:39:40] <oshima> ok,
[19:39:42] <shess> fishd: you on 10.5?
[19:39:49] <pcgod> shess: http://chromium-status.appspot.com/current
[19:39:50] <oshima> please go ahead.
[19:40:59] *** duffydack has joined #chromium
[19:41:24] <shess> fishd: I'll assume so.  I put this in ~/.gyp/include.gypi:
[19:41:29] <awong> comitted.
[19:41:37] <oshima> rvargas, gltov: can you hold a bit. I want them to pass at least compilation phrase.
[19:41:53] <shess> fishd: { variables': { 'enable_svg': '0', }, }
[19:42:16] <rvargas> oshima: shure, just let me know when it is ok.
[19:42:27] <shess> fishd: except mine's more formatted.  This lets chromium build on 10.5, but unit_tests won't.
[19:42:41] <oshima> rvargas: will do.
[19:42:48] <shess> pcgod: even more cool!
[19:43:35] *** jeremyselier has quit IRC
[19:43:47] <shess> pcgod: Except AFIACT ?reload=X doesn't do what I want :-(.
[19:43:47] <fishd> andybons: at head
[19:43:54] <fishd> shess: on 10.5
[19:44:08] <fishd> shess: thanks
[19:44:25] <bryner> quick question: is there a good way to create a Callback where you want to bind some of the function parameters at the time you create the callback?
[19:45:07] <glotov> oshima: no problem, will do tomorrow. It's late in my timezone
[19:45:32] <awong> bryner: Task lets you do that.  AFAIK, callback does not support currying.
[19:45:40] <shess> fishd: Needs a clobber build after gclient sync, BTW.
[19:46:19] <fishd> yeah
[19:46:21] <fishd> thx
[19:46:22] <awong> bryner: NewRunnableMethod lets you prebind all parameters, and NewCallback lets you prebind none. Mid-states are missing...
[19:46:34] <bryner> hm, ok
[19:47:26] <bryner> i guess i can just use a member for the parameter i want to prebind
[19:48:11] <bryner> (i definitely need to support 1 unbound parameter, so Task won't work)
[19:48:27] <awong> bryner: that's what I've been doing, writing little adapters.  stoyan did some work to make currying work, but it is only included for unittests.
[19:48:38] <bryner> oh, well this is actually in a test
[19:48:40] <bryner> where's that at?
[19:48:50] <awong> bryner: errr..forgot exactly. look for "mutant"
[19:48:55] <bryner> lol
[19:48:56] <bryner> ok
[19:49:13] *** pseudomonas|away has quit IRC
[19:49:24] <awong> should be in the testing directory top level I think, but not sure.
[19:49:37] <bryner> i think i found it
[19:50:12] *** slightlyoff has joined #chromium
[19:50:12] *** ChanServ sets mode: +v slightlyoff
[19:50:40] *** duffydack has left #chromium
[19:51:00] *** MichealH has quit IRC
[19:52:23] *** hebz0rl has quit IRC
[19:52:58] <eroman> dear powers that be, may i commit http://codereview.chromium.org/2905007/show ?
[19:53:03] <awong> oshima: FYI, after the current webkit roll, we should have 2 breaks in windows again. I prepping the next webkit roll which should fix those too.
[19:54:04] <feldstein> awong or whoever is in charge: can i commit http://codereview.chromium.org/2804035/show ?  It's a mac change for translate bars, very localized.
[19:54:30] *** janm has quit IRC
[19:55:21] *** fqian has joined #chromium
[19:55:47] <davidben> Is the git mirror down? (Also, may I commit http://codereview.chromium.org/2899005/show when I'm able to talk to the repository again?)
[19:56:45] *** podivilov1 has quit IRC
[19:56:55] * awong is looking at the changes now.
[19:57:00] <oshima> ok, it passed compilation.
[19:57:14] <oshima> rvargas: lgtm ->
[19:57:37] <oshima> eroman: lgtm
[19:58:07] <eroman> oshima: thanks. so i can commit now?
[19:58:16] <oshima> eroman: yes, please go head.
[19:58:24] <rvargas> oshima:thanks
[19:58:24] <oshima> feldstein: lgtm
[19:58:44] <feldstein> thanks
[19:59:36] <oshima> davidben: lgtm
[20:01:19] *** tfarina has quit IRC
[20:01:19] *** jlebar has quit IRC
[20:01:34] *** jamesr has joined #chromium
[20:01:34] *** ChanServ sets mode: +v jamesr
[20:03:18] *** duffydack has joined #chromium
[20:03:32] <feldstein> how do i commit when the tree is closed?
[20:03:44] <awong> feldstein: pass a -f to whatever command you're suing.
[20:03:54] <awong> eg., git cl dcommit -f
[20:04:03] <feldstein> great, thanks
[20:07:27] <stuartmorgan> oshima: can I commit http://codereview.chromium.org/2979002/show ? It only affects a feature that's behind a flag
[20:07:48] *** duffydack has left #chromium
[20:08:16] <oshima> stuartmorgan: lgtm
[20:08:20] <stuartmorgan> thanks
[20:09:25] <gavinp> awong: 52184 builds green?
[20:09:54] <awong> gavinp: yep, it seems to compile.
[20:10:09] <gavinp> excellent.  i'm using it as a basis for my tries until we're green.
[20:10:25] <awong> gavinp: ok.  FYI, you may see 2 win layout test failures.
[20:11:40] <awong> oshima: a bunch of bots are hung on ui_tests on the revert since most of the m are timing out. Since that build is pretty useless anyways, I'm going to interrupt some of them.
[20:12:46] <Tommi> awong: permission to commit CF change? http://codereview.chromium.org/2895010/show
[20:13:44] *** janm has joined #chromium
[20:14:19] *** Tommi is now known as tommi
[20:14:23] *** ebraminio has joined #chromium
[20:15:07] <awong> Tommi: go for it.
[20:15:34] <tommi> thanks
[20:15:42] <oshima> awong: ok,
[20:16:11] *** jlebar has joined #chromium
[20:18:54] *** phrearch has quit IRC
[20:19:01] *** MichealH has joined #chromium
[20:20:28] *** bweinstein has quit IRC
[20:21:51] <stuartmorgan> oshima: why are you disabling test because of leaks?
[20:21:59] <stuartmorgan> oshima: that's what the suppressions file is for
[20:23:17] *** tfarina has joined #chromium
[20:23:36] *** fta_ has joined #chromium
[20:24:04] *** eseidel has joined #chromium
[20:25:17] *** fta has quit IRC
[20:25:30] *** fta_ is now known as fta
[20:26:28] <pkasting> Can I commit a change that fixes some line endings?
[20:26:35] <stuartmorgan> oshima: the only time test should be disabled is if they hang or crash
[20:26:47] <oshima> stuarmorgan: sorry my mistake.
[20:26:53] *** bent-mozilla has joined #chromium
[20:27:12] <oshima> pkasting: go ahead.
[20:27:29] *** maikmerten has joined #chromium
[20:29:07] *** janm has quit IRC
[20:29:08] <oshima> stuartmorgan: it has 50 of them and it takes some time to find suppression that works for all of them.
[20:30:10] <oshima> let me look into chromeos valgrind failures first, and will fix it.
[20:30:19] <stuartmorgan> oshima: Losing test coverage because of a few leaks (repeated however many times) is really bad
[20:31:41] <awong> stuartmorgan: do you think you might be able to help out oshima in triaging the suppressions?
[20:31:51] *** rniwa has joined #chromium
[20:31:51] *** ChanServ sets mode: +v rniwa
[20:31:59] <stuartmorgan> sure
[20:32:07] <oshima> stuartmorgan: thanks
[20:32:47] *** arv_away has quit IRC
[20:33:08] <oshima> stuartmorgan: i'll revert my changes at least.
[20:34:14] <awong> oshima: btw, do you know where estade and jam are? You seem to be taking the brunt of the work on this morning's sheriffing. :)
[20:34:27] *** bweinstein has joined #chromium
[20:35:33] <oshima> no i don't know.
[20:36:09] <oshima> i'l
[20:36:14] <oshima> l ping them
[20:36:15] <stuartmorgan> estade isn't in yet
[20:36:59] *** estes has joined #chromium
[20:37:47] *** Buglouse has joined #chromium
[20:38:32] <gavinp> awong: is a failure of ExtensionApiTest.FavIconPermission popular today?
[20:38:46] <awong> gavinp: don't know :(
[20:42:44] <johnny_g> oshima, stuartmorgan: i think i have a fix for the leaks in question, that might be even easier than adding suppressions
[20:43:15] <stuartmorgan> johnny_g: which ones?
[20:43:25] <johnny_g> DesktopNotificationTest*
[20:43:31] <oshima> ok, cl?
[20:43:51] <johnny_g> still making it...
[20:49:48] <thakis> awong: ok for http://codereview.chromium.org/2932009/show ?
[20:50:18] <thakis> oshima: ^
[20:51:00] *** urbanape has quit IRC
[20:51:50] <trungl-bot> Tree closed by ajwong at chromium dot org: Tree is closed (oshima thorottling, ask on irc to commit. webkit -> awong)
[20:52:07] <jhawkins> oshima: I have a smallish CL ready to go
[20:52:13] *** roc has joined #chromium
[20:52:27] <oshima> thakis: lgtm
[20:52:50] <oshima> jhawkins: please put cl here.
[20:53:04] <jhawkins> oshima: http://codereview.chromium.org/2941009/show
[20:53:24] <jhawkins> oshima: tests aren't done, so I'll ask again after lunch
[20:53:35] <oshima> jhawkins: ok
[20:54:34] <johnny_g> oshima: http://codereview.chromium.org/2903011/show
[20:54:56] *** peloverde has joined #chromium
[20:55:17] <oshima> johnny_g: lgtm. please update BUG=
[20:55:56] <johnny_g> oshima:  whoops... done
[20:56:09] <oshima> johnny_g: lgtm
[20:56:41] <peloverde> fta: What's going on with your chromium-codecs-ffmpeg package? it seems to have failed to build a month ago and stuck
[20:56:50] *** thakis_ has quit IRC
[20:56:55] <awong> oshima: fyi, we're almost done cycling through the webkit roll. just back to 2 failures on windows webkit dbg which should be fixed in the next roll
[20:58:19] *** chronarion has joined #chromium
[20:59:34] <oshima> awong: cool. do you think we can open the tree?
[20:59:45] *** janm has joined #chromium
[21:00:09] <awong> I'm tempted to say yes...let me file a supression for those 2 layout tests now.
[21:00:21] <oshima> k
[21:00:50] *** romainhuet has quit IRC
[21:01:56] <stuartmorgan> oshima: ExtensionPrefStore suppressions landed; if there are any I missed I'll get them in the next pass. I'll keep wading through the memory tree after lunch
[21:03:07] *** romainhuet has joined #chromium
[21:03:14] <oshima> stuartmorgan: thanks. i'm working on valgrind failures. jrg said he's working on mac.
[21:03:17] <awong> oshima: before reopening, we should go through all the errors on the builds and make sure we know what's causing them..
[21:03:24] *** romainhuet has quit IRC
[21:04:59] <fta> peloverde, i was following trunk (hence it was 0.6+xx), but fbarchard told me it's not ready, so i'm now following the rev tracked by chromium (so it's 0.5+xx)
[21:05:08] <fta> and i can't go back rev-wise
[21:05:15] <peloverde> fta: ok
[21:05:32] <awong> oshima: I'm going to new test expectations for the failing win layout tests from the webkit roll.
[21:05:42] *** jlebar has quit IRC
[21:05:57] <oshima> awong: checking the tree again and seems like there are a few bots that are stuck :(
[21:06:07] <oshima> awong: go ahaed.
[21:06:16] *** peloverde has quit IRC
[21:06:28] <fta> peloverde, so it will auto-restart once chromium catches up on ffmpeg 0.6
[21:07:07] *** eseidel has quit IRC
[21:07:32] <awong> oshima: committed.
[21:07:36] <awong> looking at bots now. *sigh*
[21:08:07] *** Vanderhuge has quit IRC
[21:08:11] <slightlyoff> awong: any thoughts on reducing the # of A/V DLLs?
[21:08:28] <awong> slightlyoff: make it build for us :)
[21:08:32] <awong> we did it in mac/linux.
[21:08:49] <awong> is it causing an issue?
[21:08:51] <slightlyoff> any idea why it's not working on windows?
[21:08:57] <slightlyoff> awong: slower startups than I'd like
[21:09:05] <slightlyoff> awong: scratch that, *need*
[21:12:03] <awong> slightlyoff: the reason it doesn't work in windows is that the build system is completely different.
[21:12:16] <awong> in mac/liux, we managed to integrate into gyp to build the DSOs natively.
[21:12:19] *** duffydack has joined #chromium
[21:12:56] <awong> However, ffmpeg is not MSVC compatible at all.  So we basically build by hand using mSYS, and MinGw with the ./configure provided to us.
[21:12:57] <slightlyoff> where would I look for those changes? chrome.gyp?
[21:13:02] <awong> oh man...
[21:13:03] <slightlyoff> ugg
[21:13:08] <awong> third_party/ffmpeg/ffmpeg.gyp
[21:13:12] <awong> FYI, this is a can of worms.
[21:13:33] <awong> The path forward is likely this: (1) put MSYS + MinGw into the platform deps for a gclient checkout on window.s
[21:13:45] <slightlyoff> (not a great way to make friends)
[21:13:48] <awong> (2) teach gyp to move into "gcc" mode for one target.
[21:14:07] <awong> (3) adapt the current ffmpeg.gyp to work with MSYS quirks.
[21:14:39] <awong> I talked about this briefly with bradnelson + markmentovai a few months ago.  Step #1 seems the least controversial.
[21:14:48] <awong> The next two aren't charted. I'd put this at about 2-3 weeks of solid work. :(
[21:15:20] <awong> The other option is to hack it and modify ffmpeg's build manually...but this is a pretty annoying maintenance burden on what is already a touchy manual build process.
[21:15:47] *** fta_ has joined #chromium
[21:16:29] *** fta has quit IRC
[21:16:30] *** fta_ is now known as fta
[21:16:43] <slightlyoff> well, I'll look at what's happening in ffmepg.gyp and go from there
[21:16:55] <slightlyoff> I might just do dirty tricks to keep page faults from hurting so bad
[21:17:02] <slightlyoff> will keep you posted
[21:17:16] <awong> slightlyoff: okay. that will let you know what's going on in linux/mac.  For win, you will want to look in the READMEs.
[21:17:18] *** akalin has joined #chromium
[21:17:24] <akalin> greetings chromiumites
[21:17:52] <awong> slightlyoff: good luck! definitely ping me, scherkus, or fbarchard if you have questions.
[21:18:22] <slightlyoff> yes, question: how did you ever get this working in the first place? ;-)
[21:18:40] <awong> slightlyoff: lots of help from markmentovai...and about 1 month of my life.
[21:18:49] <markmentovai> :)
[21:18:54] <markmentovai> it was a fun time
[21:18:54] <oshima> killing jobs on linux tests (dbg)(2) and mac10.6 tests. they're stuck for 2 hours
[21:19:14] *** ojan has quit IRC
[21:19:33] <jam2> oshama: can i check in http://codereview.chromium.org/2948009/show?  it's a minor fix to pepper that shouldnt be triggered for anyone else
[21:19:41] *** duffydack has quit IRC
[21:20:31] <oshima> jam2: lgtm
[21:21:02] <jam2> thanks
[21:21:11] <jam2> btw let me know if you need any help, i'm sheriff too
[21:21:40] <jam2> im keeping my eye on the tree
[21:21:52] <akalin> oshima: can I check in http://codereview.chromium.org/2979003/show ?  all trybots have passed
[21:22:58] <oshima> jam2: i think we need to restart linux test (dbg)2 and mac10.6 test
[21:23:13] <oshima> akalin: lgtm
[21:23:14] <awong> jam2: I think there's a number of issues that aren't well understood still.  For instance, startup_Test's twentytabs-* tests have been flaky since last night.
[21:23:54] <oshima> nsylvain: ping
[21:24:46] <oshima> any troopers here?
[21:25:24] *** pathorn has joined #chromium
[21:25:24] *** ChanServ sets mode: +v pathorn
[21:25:29] <jam2> hey
[21:25:42] <jam2> ok i can try restarting them, although i dont htink i have hte pw
[21:25:45] <jam2> let me ask around
[21:25:49] *** maikmerten has quit IRC
[21:25:52] <awong> jam2: I've got the password
[21:26:00] <oshima> i have, unfortunately :p
[21:26:12] *** duffydack has joined #chromium
[21:26:14] <jam2> i've never restarted a mac/linux.  do you just ssh and restart?
[21:26:20] <awong> there's instructions
[21:26:22] <awong> let me find them
[21:26:24] <jam2> thanks
[21:26:34] <jhawkins> oshima: ready for http://codereview.chromium.org/2941009/show ?
[21:26:37] *** sebmarkbage has joined #chromium
[21:26:38] <oshima> awong: can you restart mac bot? i'll restart linux one
[21:26:50] <awong> oshima: okay
[21:27:56] <oshima> jhawkins: hold a sec.
[21:30:19] *** eseidel has joined #chromium
[21:30:42] *** awidegreen has quit IRC
[21:30:58] *** duffydack has left #chromium
[21:31:16] *** thakis_ has joined #chromium
[21:31:17] *** ChanServ sets mode: +v thakis_
[21:32:42] *** mnissler_ has joined #chromium
[21:33:23] <awong> oshima: rebooting macbot now.
[21:33:35] <jrg> oshima: yes I'm working on it
[21:34:07] <oshima> awong: i rebooted linux test bot
[21:34:52] <oshima> jrg: thanks
[21:36:29] *** dr_win has quit IRC
[21:40:21] <oshima> grabbing lunch, but will be back soon.
[21:41:09] *** seanparent has joined #chromium
[21:41:15] *** ChanServ sets mode: +v seanparent
[21:41:28] *** zer0her0 has quit IRC
[21:41:33] <trungl-bot> Tree closed by ajwong at chromium dot org: Tree is closed (awong thorottling, ask on irc to commit.)
[21:41:44] *** dr_win has joined #chromium
[21:42:07] *** eglaysher has quit IRC
[21:43:08] *** MichealH_ has joined #chromium
[21:46:29] *** MichealH has quit IRC
[21:46:31] *** MichealH__ has joined #chromium
[21:46:48] <awong> seanparent: re http://codereview.chromium.org/2859043/show, there are grd changes in there.
[21:47:02] <awong> will that necessitate a clobber of the windows bots?
[21:47:03] *** MichealH__ is now known as MichealH
[21:47:14] *** janm has quit IRC
[21:47:21] <seanparent> Yes - pinged the sheriff's and got pre-approval for that.
[21:47:27] *** MichealH has quit IRC
[21:47:27] *** MichealH has joined #chromium
[21:47:43] <seanparent> Likely won't require a Win clobber - they are Chromium-OS specific.
[21:48:07] <markmentovai> awong: (you may be aware) mac debug compile?
[21:48:24] <jam2> ajwong: mac compile falure
[21:48:33] <jam2> ah, mmentovai just beat me to it :)
[21:48:39] <markmentovai> :)
[21:48:45] <awong> I'm aware now
[21:48:46] <awong> :P
[21:48:54] <jam2> took forever for the stdio to load, not sure why
[21:49:05] <markmentovai> it's because we're in bad shape
[21:49:10] *** MichealH_ has quit IRC
[21:51:28] *** Vanderhuge has joined #chromium
[21:52:18] <awong> jam2: I don't know enough about grd changes to know the effects.  Could you take a peek at seanparent's CL and double-check the clobber safety?
[21:52:32] <awong> just want to be extra careful given the last 36-hours.
[21:52:41] *** chrisccoulson has quit IRC
[21:53:07] <jam2> awong: i can take al ook, but i dont know what to look for :)
[21:53:27] <jam2> i think we might need two runs, as per usual grd changes
[21:53:30] <markmentovai> it'll likely screw up the windows tests for a cycle
[21:53:45] <markmentovai> i'd hold off 'til the compile failure is under control
[21:53:51] <jam2> although, given how red the tree is, i'd say we shoould wait
[21:54:00] <jam2> seanparent: let's wait until the tree is greenf irst
[21:54:23] <awong> danggit.
[21:54:24] <seanparent> Okay - any eta?
[21:54:34] <awong> I accidnetally rolled the DEPs with my commit.
[21:54:38] <jam2> not sure
[21:55:59] *** taf2 has joined #chromium
[21:56:44] <jam2> awong: it's ok i had an accidental checkin yesterday too :)
[21:56:52] <awong> unrolling
[21:56:52] <jhawkins> awong: I was supposed to commit a while back, still good to go?
[21:57:39] <awong> jhawkins: what was your CL again?
[21:57:53] <awong> deps unrolled. mac should cycle green.
[21:58:13] <jhawkins> http://codereview.chromium.org/2941009/show
[21:58:45] <awong> jhawkins: lgtm
[21:58:53] <jhawkins> thanks
[21:59:30] <markmentovai> awong: lgtm.
[21:59:42] <awong> markmentovai: thanks.  I'll go fix upstream now
[21:59:46] <markmentovai> thanks.
[22:00:19] *** janm has joined #chromium
[22:01:00] *** kliegs has joined #chromium
[22:01:05] <awong> seanparent: I would expect at least another couple of hours before the tree opens up.
[22:01:08] *** kliegs_ has joined #chromium
[22:01:19] *** kliegs_ has quit IRC
[22:01:29] <seanparent> Okay - I'll check back...
[22:01:33] <awong> sorry
[22:01:43] <seanparent> It happens.
[22:02:01] <awong> random thought, would it be sensible to batch up multiple grd changes?
[22:02:29] <awong> pkasting had a huge one reverted this morning.  If that's going to be reapplied today, should we do that at hte same time as seanparent's?
[22:02:58] *** jlebar has joined #chromium
[22:03:27] *** tfarina has quit IRC
[22:03:29] <vandebo> awong: it's hard to get them into the same compile
[22:03:47] <awong> vandebo: :(
[22:04:30] <vandebo> you could do it easily by integrating the two changes into a single commit, but then you can't revert them individually and you conflate change history, etc
[22:04:39] <markmentovai> pamg: looks like r52088 is the source of the leaks that linux heapcheck is detecting. http://build.chromium.org/buildbot/memory/builders/Linux%20Heapcheck/builds/4857
[22:04:55] <stuartmorgan> markmentovai: suppressions are in for that
[22:05:02] <markmentovai> stuartmorgan: thanks
[22:05:08] *** fta has quit IRC
[22:05:22] <markmentovai> vandebo: you can get close if you get both of them to get their commits ready and have them pull the trigger at the same time
[22:05:41] <vandebo> markmentovai: unless the changes touch the same file
[22:05:56] *** chrisccoulson has joined #chromium
[22:06:29] <awong> vandebo: one git client. :)
[22:06:43] *** urbanape has joined #chromium
[22:06:45] <awong> and ascript
[22:06:52] <markmentovai> well, hope for no conflicts, have guy #2 do a quick update before committing...
[22:07:13] <markmentovai> the window's not THAT tiny with compile times as they are now
[22:07:24] * awong hates that stl uses size_t for size().
[22:08:17] <oshima> back now
[22:08:35] <awong> webkit style question: ASSERT(index < m_labels.size()); is failing because index is signed (and should be). Do I just do something like ASSERT(static_cast<size_t>(index) < m_labee.size())?
[22:08:58] <jam2> sgtm
[22:09:05] <awong> done.
[22:09:24] <stuartmorgan> markmentovai: actually, I lied; I was thinking of the same leaks in memcheck. I'll add heapcheck suppressions and update the bug
[22:09:34] <markmentovai> stuartmorgan: thank you
[22:09:41] <jamesr> awong: failing or compile warning/erroring? static_cast<size_t> is the right way to handle the compile warning/error
[22:09:54] <jamesr> with an && index >=0 if you are super paranoid
[22:09:59] <awong> jamesr: compile warning.
[22:10:11] *** janm has quit IRC
[22:10:18] <awong> There's an if(index >=0) check above already...but might as well assert it too.
[22:10:40] <jamesr> i actually prefer making the other side of the comparison signed, but i think either way is fine
[22:11:38] <awong> jamesr: yeah, I disliked unsigned. But given the specific usage, signed -> unsigned makes more semantic sense.
[22:11:50] <thakis> jamesr: this way he doesn't need the index >= 0 part tho
[22:11:54] <fishd> when a renderer crashes during a ui_test, is there a way to get the stack dump in the renderer process?
[22:11:59] *** thakis_ has quit IRC
[22:12:33] <awong> thakis: good point.
[22:13:10] <markmentovai> careful going to signed, too, if you do that. these things have a tendency to get even weirder when size_t is 64 bits but int is 32. more stupid things to have to check.
[22:13:44] <awong> tears.
[22:14:07] <jamesr> whine. yeah if index is non-negative go to size_t
[22:14:12] <markmentovai> 'zactly
[22:15:29] *** cristim has left #chromium
[22:16:09] <awong> bah, I'm just going to do jamesr's original paranoid suggestion. More explicit, and my head's too scattered to be confident.
[22:16:11] <markmentovai> i've had some really annoying cases where you had an off_t and a size_t. off_t would always be signed64, size_t can be unsigned32 or unsigned64 depending on programming mode. so depending on the model, the max size_t would be either less than or greater than the max off_t. now write an assertion, try to keep it readable, and try to keep the compiler from complaining.
[22:16:45] <awong> clearly the solution is to rewrite chrome in ruby.
[22:16:48] <thakis> DCHECK(true)
[22:16:57] <markmentovai> thanks, nico
[22:17:10] <oshima> awong: like that:)
[22:17:49] <jhawkins> anyone know the appropriate steps need to check in a new imaged file into chrome?
[22:18:05] *** nickcarter has joined #chromium
[22:18:05] <thakis> jhawkins: add the png (or whatever) in its own cl and commit it
[22:18:20] <thakis> jhawkins: else the try job for the cl that wants to use the image won't work
[22:18:30] <jhawkins> thakis: does the new file need to be in some list, so it gets compiled into a resource file?
[22:18:49] *** trungl has joined #chromium
[22:18:50] *** ChanServ sets mode: +v trungl
[22:18:57] <thakis> jhawkins: it needs to references from a grd file, sure
[22:19:05] <thakis> jhawkins: but that belongs into the "real" cl i believe
[22:19:12] <nickcarter> awong: permission to commit http://codereview.chromium.org/2844037 ?
[22:20:11] <jhawkins> thakis: ok, thanks
[22:20:28] <thakis> s/references/be referenced/
[22:20:31] *** jshin has quit IRC
[22:21:12] <awong> nickcarter: go for it
[22:23:11] <nickcarter> awong: vielen Dank
[22:23:25] <pkasting> awong, seanparent, vandebo: My .grd change might be able to re-land after a GRIT fix of tony^work's lands (I'll have to rebuild with it and see if the problems are gone).  I can probably batch it with seanparent's change and with another .grd change of mine.
[22:23:27] *** janm has joined #chromium
[22:23:30] *** leavittx has joined #chromium
[22:23:34] <tony^work> pkasting: I'll try landing now
[22:24:00] <awong> nickcarter: ????
[22:24:31] <awong> pkasting: sgtm.
[22:25:28] <tony^work> awong: can I land my grit fix now?
[22:25:54] <awong> tony^work: go for it.
[22:27:20] <tony^work> poor buildbot
[22:27:32] <pkasting> buildbot and codereview have both been getting slower and slower
[22:27:44] <pkasting> Is there something we can do to throw more hardware at them?
[22:27:45] <awong> are we hitting a quota somewhere?
[22:27:47] *** janm has quit IRC
[22:27:51] <rohitrao> stuartmorgan: did you change something recently about flash or core animation or layers?
[22:28:08] <stuartmorgan> rohitrao: nope, not me
[22:28:16] <stuartmorgan> Why?
[22:28:22] <jam2> jpkasting: not much we can do about codereview, it's up to app engine's datastore's speed
[22:28:28] <rohitrao> I just failed to make the findbar disappear
[22:28:31] <seanparent> pkasting: no objection if you want to batch with my cl.
[22:28:32] <awong> jam2: no way. really?
[22:28:43] <awong> jam2: are we doing something really dumb with the data model?
[22:29:35] <rohitrao> stuartmorgan: nevermind, I'm back to flash 10.0 :)
[22:29:40] <stuartmorgan> haha
[22:30:11] <awong> oshima: can I pass the tree throttle back to you? I need to go fix up webkit canary bots
[22:30:12] *** bent-mozilla has quit IRC
[22:30:21] <oshima> sure
[22:30:29] <awong> thx
[22:31:03] *** bent-mozilla has joined #chromium
[22:31:04] <trungl-bot> Tree closed by ajwong at chromium dot org: Tree is closed (oshima thorottling, ask on irc to commit.)
[22:31:44] <tony^work> ok, landed my change and forced a clobber (my change is a grd change)
[22:32:24] *** hrna has quit IRC
[22:33:53] *** mnissler_ has quit IRC
[22:33:58] <tony^work> the waterfall's not actually loading for me
[22:34:33] <mabbo> waterfall page is slow, but it does load. eventually.
[22:34:34] <oshima> tony^work: it's very slow now for some reason.
[22:34:35] *** cbentzel has quit IRC
[22:34:47] <tony^work> has anyone investigated?
[22:35:31] <oshima> not that i know of. we've been busy fixing tree itself.
[22:35:42] *** evmar_afk is now known as evmar
[22:35:52] <evmar> i have a small linux-only change
[22:36:08] <oshima> evmar: cl?
[22:36:09] <tony^work> oshima: I will ssh to the bot to see if anything obvious is awry
[22:36:27] <jhawkins> oshima: have a CL that only uploads new images http://codereview.chromium.org/2917010/show
[22:36:33] *** andrix1 has quit IRC
[22:36:33] *** Kunalagon has quit IRC
[22:36:36] <evmar> http://codereview.chromium.org/2931011
[22:36:46] <oshima> jhawkins: lgtm
[22:37:11] <oshima> evmar: lgtm
[22:37:36] *** smead has joined #chromium
[22:38:23] <vandebo> oshima: how about http://codereview.chromium.org/2806052
[22:38:27] <oshima> tony^work: ok. i'm not familiar with how water fall works. Is it build bots that causing slowness?
[22:38:52] <evmar> oh actually it depends on another change
[22:39:23] <tony^work> actually, I don't have the password for the master
[22:39:43] <awong> tony^work: it's not the same as the slave password/
[22:39:44] <awong> >
[22:39:45] <awong> ?
[22:39:54] <markmentovai> no
[22:40:14] <oshima> vandebo: do you know why WebSocket test is failing on win?
[22:40:37] <tony^work> I have like half a dozen buildbot related passwords in valentine but none of them worked
[22:40:39] <vandebo> it's just flake
[22:41:03] <markmentovai> tony^work: which machine is the master/
[22:41:14] <tony^work> codg32, I think
[22:41:19] <tony^work> I just pinged buildbot.jail
[22:41:26] *** kliegs_ has joined #chromium
[22:41:32] <tony^work> maybe I'm connecting to the wrong machine?
[22:41:58] <oshima> vandebo: lgtm
[22:42:30] <markmentovai> tony^work: do you have "Chrome - Linux/Windows/mac Servers"
[22:42:40] <markmentovai> that's the one for that machine
[22:42:51] <tony^work> markmentovai: no. can you share it or should I ask nsylvain
[22:43:07] *** eseidel has quit IRC
[22:43:14] <markmentovai> tony^work: only bevc, maruel, nsylvain can share it
[22:43:27] *** kliegs_ has quit IRC
[22:43:56] *** jeremyselier has joined #chromium
[22:44:14] <markmentovai> tony^work, oshima: it's a bunch of mac distcc compilers running
[22:44:22] <markmentovai> that's probably what's causing the master slowness
[22:44:29] <tony^work> markmentovai: I don't think distcc is supposed to run on the master
[22:44:38] <markmentovai> i don't either, i'm stopping it and will ask bev what the story is
[22:44:43] <tony^work> ok, sounds good
[22:44:47] *** abarth has joined #chromium
[22:44:47] *** ChanServ sets mode: +v abarth
[22:45:20] <oshima> thestig may know too
[22:45:52] <oshima> markmentovai: thanks a lot.
[22:46:18] * markmentovai observes load average falling
[22:46:32] <oshima> markmentovai: so master is running on mac?
[22:46:36] <markmentovai> no
[22:46:39] <markmentovai> we cross-compile
[22:46:42] <tony^work> markmentovai: maybe run sudo apt-get remove distcc?
[22:46:45] <oshima> ok,
[22:46:45] *** MichealH is now known as i
[22:46:58] *** i is now known as Iamnotmhall199
[22:47:07] <oshima> tony^work: +10
[22:47:24] *** Iamnotmhall199 is now known as MichealH
[22:47:30] <markmentovai> that's not quite right for this. i'll talk to bev when she's available and figure out the right thing.
[22:47:58] *** eseidel has joined #chromium
[22:48:30] *** bers has joined #chromium
[22:51:01] *** jshin has joined #chromium
[22:51:01] *** ChanServ sets mode: +v jshin
[22:51:51] *** bers has quit IRC
[22:54:06] *** janm has joined #chromium
[22:54:26] *** sbyer has quit IRC
[22:58:27] *** bmizerany has joined #chromium
[23:00:00] *** wjmaclean has quit IRC
[23:00:25] *** wjmaclean has joined #chromium
[23:00:33] <markmentovai> oshima, tony^work: i suspect that codg32 isn't the master after all
[23:00:40] <tony^work> heh
[23:00:45] <tony^work> maybe it's just a proxy
[23:00:47] <markmentovai> yeah
[23:00:49] <markmentovai> that's what i think
[23:00:53] <tony^work> hmm
[23:00:59] *** rniwa has quit IRC
[23:01:00] * tony^work looks for wiki page
[23:01:01] <markmentovai> let me try to find the wiki page that lists our stuff
[23:01:02] <markmentovai> heh
[23:01:38] <oshima> is this the one? https://wiki.corp.google.com/twiki/bin/view/Main/ChromeBuildbotResources
[23:01:39] <tony^work> markmentovai: try codf2
[23:01:47] <jhawkins> oshima: got a lightweight change that adds those images to browser_resources.grd http://codereview.chromium.org/2893014/show
[23:01:52] <tony^work> it's in my .ssh/config file
[23:01:54] <mpcomplete> shess: are you familiar with the mac omnibox code?
[23:02:02] <markmentovai> oh, it's on the "atlanta" wiki page
[23:02:04] <shess> mpcomplete: yes.
[23:02:28] <oshima> sorry it's internal
[23:02:37] * markmentovai feeds the puppy again
[23:02:39] <mpcomplete> shess: can you review the _mac file changes for http://codereview.chromium.org/2973006/show
[23:02:58] <oshima> does this require clobber?
[23:03:03] <mpcomplete> i think i'm screwing up retain and release for NSImages
[23:03:15] <markmentovai> this looks like the master, but it doesn't have any suspicious processes
[23:03:18] <oshima> jhawkins: does this require clobber?
[23:03:31] <jhawkins> oshima: I don't think so
[23:04:12] <jhawkins> it doesn't change the strings, which is what breaks the tests
[23:05:12] <oshima> hmm, i'm not sure. does anyone know for sure?
[23:05:27] <suzhe> 7 of 13 tests in browser_focus_uitest.cc passed on my Mac :D
[23:05:44] <tony^work> jhawkins: adds images to the end of the file?
[23:06:10] *** MichealH has quit IRC
[23:06:10] <tony^work> webkit dbg layout tests may fail.  it didn't seem to pick up my clobber
[23:06:13] <jhawkins> tony^work: almost the end. there's a chromeos/linux_views block at the end of the file
[23:06:27] <tony^work> jhawkins: that should be safe
[23:06:32] <jhawkins> oshima: ^^
[23:06:51] <oshima> ok
[23:06:54] <jhawkins> thanks
[23:08:29] <shess> mpcomplete:  OK.  Note that I'm going to conflict with you on http://codereview.chromium.org/2805070 .  Are you ready-to-go once I've reviewed?
[23:08:33] *** loislo has joined #chromium
[23:08:48] *** hebz0rl has joined #chromium
[23:09:24] <mpcomplete> shess: basically, but i can hold off until after you commit
[23:09:25] <oshima> jhawkins: lgtm
[23:10:39] <oshima> jam2: ping
[23:11:17] *** cpu has joined #chromium
[23:11:17] *** ChanServ sets mode: +v cpu
[23:13:36] <gavinp> oshima: what's a good green building rev to run tests with?
[23:13:56] <cpu> howdy, can I haz a checkin ticket please?
[23:14:44] <oshima> gavinp: r52190 is the last cl that was all green.
[23:15:00] <gavinp> oshima, you da man.  thanks.
[23:15:24] <jhawkins> oshima: I have a CL that just marks a test as fails instead of disabling it http://codereview.chromium.org/2950006/show
[23:16:03] <oshima> jhawikins: lgtm. nice, thanks.
[23:17:14] *** Cheery has joined #chromium
[23:17:23] <Cheery> I find it weird that chrome javascript is faster than python
[23:17:48] *** GeekShad0w has joined #chromium
[23:18:16] *** stevenjb has joined #chromium
[23:18:16] *** ChanServ sets mode: +v stevenjb
[23:19:03] <oshima> hi, does anyone know why mac test 10.6 dbg 3 is failing so badly?
[23:19:12] <stevenjb> oshima: I have a small change, http://codereview.chromium.org/2952006/show ready to push whenever it is a good time. Not urgent.
[23:19:23] *** GeekShadow has quit IRC
[23:19:25] *** bent-mozilla has quit IRC
[23:19:56] <oshima> stevenjb: lgtm
[23:19:58] <tony^work> oshima: the windows dbg builder didn't get my clobber in time so I think the win dbg tests may fail
[23:20:19] *** ada2358 has joined #chromium
[23:20:31] <stevenjb> oshima: Thanks, pushing.
[23:20:50] *** General1337 has joined #chromium
[23:21:23] <oshima> tony^work: ok
[23:21:44] <oshima> I have to leave for interview now. awong is kindly taking throttling role. will be back in an hour.
[23:22:40] *** estes has quit IRC
[23:22:44] *** dcheng has quit IRC
[23:23:11] <shess> oshima, awong: I have a Mac change, http://codereview.chromium.org/2805070, which is kind of substantial, but shouldn't impact non-Mac at all.  Since it's the first in a train of changes I need for M6, am wondering if I can push?
[23:23:24] <oshima> but if someone can look into mac issue, it'd be greatly appreciated. it's very slow and i couldn't retrieve log...
[23:23:32] <awong> oshima: I've got it
[23:23:44] <awong> shess: give me a sec to look
[23:23:48] <trungl-bot> Tree closed by oshima at chromium dot org: Tree is closed (awong thorottling, ask on irc to commit.)
[23:23:48] <oshima> thinks. off now
[23:24:10] *** General13372 has quit IRC
[23:25:04] *** taf2 has quit IRC
[23:25:25] <tony^work> e.g., the webkit (dbg)(3) failures are because of my grit change
[23:25:29] <tony^work> it should go green in the next cycle
[23:25:38] <tony^work> but there's going to be lots of red coming because of the missed cycle
[23:25:49] <awong> tony^work: k. thanks for the heads up.
[23:26:24] *** ROBOd has quit IRC
[23:26:43] *** dcheng has joined #chromium
[23:28:03] <mpcomplete> shess: are you saying that calling gfx::SkBitmapToNSImage for every keystroke is not a big deal?
[23:28:50] <jhawkins> oshima: another test re-enabling http://codereview.chromium.org/2979004/show
[23:29:00] <jhawkins> awong: ^^
[23:29:03] <awong> shess: is tehre any chance this might break layout tests on mac?
[23:29:49] <awong> jhawkins: does this test timeout, or does it just fail?
[23:30:09] <jhawkins> awong: fails
[23:30:14] <awong> jhawkins: lgtm
[23:30:17] <jhawkins> thanks
[23:30:40] *** johnny_g has quit IRC
[23:31:57] *** xji_ has quit IRC
[23:32:29] <awong> shess: go ahead and submit
[23:32:33] <evmar> http://codereview.chromium.org/2982002/show plz
[23:32:41] <shess> mpcomplete:  For icon-sized images, I don't think it's huge.  I guess I'm more nervous about the lifetime issues around this map than I am about the costs of SkBitmapToNSImage.  Maybe if you cleared the map when the popup is closed?
[23:32:46] *** abarth is now known as abarth|afk
[23:33:14] <shess> awong: sorry, multi-tasking.  This is all in ui chrome, not in web content.
[23:34:15] <mpcomplete> shess: if i don't put the NSImages in a map, who is responsible for destroying them?
[23:35:31] <mpcomplete> shess: it boils down to [cell setImage:gfx::SkBitmapToNSImage(bitmap)] - will 'cell' release the image?
[23:35:47] <shess> mpcomplete:  I'd assume they are autoreleased.  So the button cell will snag a reference, and they'll hang around until you next update everything.
[23:35:59] *** abarth|afk has quit IRC
[23:36:12] <motownavi> (they are autoreleased from that function)
[23:36:29] *** eseidel has quit IRC
[23:36:31] <shess> mpcomplete: Cocoa has a thing called autorelease pools.  When you say [object autorelease], it gets entered into the pool, and will be sent -release next time things reach the event loop.  If someone meanwhile does a -retain, they'll keep the reference.
[23:37:22] <shess> mpcomplete: but note that the cell keeps a reference, so for purposes of drawing and stuff, everything is all efficient.
[23:37:23] <mpcomplete> oh, so SkBitmapToNSImage is in that pool?
[23:37:28] <shess> mpcomplete: yep.
[23:37:35] <shess> [It better be!]
[23:37:39] <pkasting> My viewpoint is that we should write the simplest code that's reasonable
[23:37:45] <pkasting> And caching is more complex
[23:37:51] *** michaeln has joined #chromium
[23:37:51] *** ChanServ sets mode: +v michaeln
[23:37:53] <pkasting> Even if it's less efficient, does that matter?
[23:38:09] <pkasting> I mean, spending a microsecond more on each keypress == who cares
[23:38:11] <shess> I think this kind of cost on a per-keystroke basis without a force-multiplier is reasonable.
[23:39:02] <mpcomplete> pkasting: yeah, i can buy that
[23:39:32] <shess> awong: OK, I'm going in.  Will watch things, and revert if anything can be traced to me :-).
[23:39:35] <jhawkins> oshima: another re-enable http://codereview.chromium.org/2980007
[23:40:03] <mpcomplete> actually the reason i added the caching was so that i could control the lifetime of the GdkPixbuf/NSImage. but it sounds like i don't need to for NSImage
[23:40:20] *** eseidel has joined #chromium
[23:42:21] <thakis> skia folks: is there an easy way to clear the alpha channel of an SkBitmap to 0xff
[23:42:30] <thakis> (easier than the 3 lines to loop over all pixels that is)
[23:43:16] <evmar> please sir, i'd like to commit.  it's been so long and i am so very hungry
[23:43:30] <Cheery> http://boxbase.org/fun/knights/ <- probably won't stay in the url, but I put it anyway.
[23:43:42] <awong> evmar: hah. CL?
[23:43:45] <tony^work> markmentovai: did you figure out why the master is so slow?
[23:44:00] <markmentovai> no, but bev was going to look into it
[23:44:09] <estade> awong: me too: http://codereview.chromium.org/2996002/show
[23:44:22] <shess> mpcomplete: OK - in that case I'm worried that the lifetime isn't well-controlled.  AutocompletePopupViewMac lives as long as the window the field is in.
[23:44:33] <awong> estade: go for it
[23:44:47] <evmar> http://codereview.chromium.org/2982002/show
[23:44:47] *** bent-mozilla has joined #chromium
[23:44:56] *** eseidel has quit IRC
[23:45:02] <jhawkins> awong: another re-enable http://codereview.chromium.org/2980007
[23:45:08] <awong> evmar: go for it
[23:45:28] <mpcomplete> shess: i am going to try removing the caching and see how it looks. (as for lifetime, the SkBitmaps will live as long as the Profile)
[23:45:53] <awong> jhawkins: go for it
[23:45:58] <jhawkins> thanks
[23:47:37] <Cheery> I feel a little amused that I have to use javascript if I want to make something many as possible can see.
[23:49:10] <pkasting> awong: Can I commit a chromeos test fix?
[23:49:19] <awong> pkasting: sure
[23:49:38] <akalin> so sawbuck
[23:49:44] <akalin> is it supposed to work for release versions of chrome?
[23:49:53] <akalin> do release versions of chrome (on windows) do logging?
[23:52:30] *** alt-dot-net-geek has joined #chromium
[23:53:13] <cpu> awong: when can I commit:  http://codereview.chromium.org/2974001/show   ?
[23:53:33] <awong> cpu: how about now?
[23:53:38] <cpu> cool
[23:55:55] 
[23:57:18] <mattm_g> akalin: does for me.  Did you do select Log->Capture in sawbuck?
[23:57:52] <awong> ananta: would it be possible to get past at least hte compile stage on the try bots?
[23:57:56] *** agayev has quit IRC





top