/irc-logs / mozilla / #developers / 2012-05-01 / end

Options:

# Session Start: Tue May 01 00:00:00 2012
# Session Ident: #developers
# [00:00] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [00:00] <jgilbert> can't the latter recognize a^nb^n?
# [00:00] <jgilbert> which puts it in CFG land, iirc
# [00:00] <jtcranmer> I don't think so
# [00:00] <ekr> I fear that PCREs may in fact be turing complete.
# [00:01] <jtcranmer> you get backrefs
# [00:01] <jgilbert> I thought one variant had counting
# [00:01] * Joins: jfkthame (jfkthame@15AA2040.A1C12133.9542EC20.IP)
# [00:02] <jhammel> ekr: i do believe they are, actually
# [00:02] * Quits: Stefan (Stefan@A64C0DEE.58530BE4.AA3FC64B.IP) (Connection reset by peer)
# [00:02] * Quits: jet (junglecode@moz-79F891EE.hsd1.ca.comcast.net) (Quit: jet)
# [00:02] <jhammel> though i don't have the link handy that proves it
# [00:02] <ekr> This chat is too small to contain it
# [00:02] <jtcranmer> excluding the ability to stuff code in PCRE
# [00:02] * AutomatedTester|AFK is now known as AutomatedTester
# [00:02] * Quits: mjschranz (mjschranz@F2D29657.F60B0462.67AC9B1.IP) (Client exited)
# [00:03] * Quits: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Connection reset by peer)
# [00:03] * Joins: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
# [00:03] * Quits: jfkthame (jfkthame@15AA2040.A1C12133.9542EC20.IP) (Quit: jfkthame)
# [00:03] * Joins: mjschranz (mjschranz@F2D29657.F60B0462.67AC9B1.IP)
# [00:04] <jprmc> sicking: can you respond in https://bugzilla.mozilla.org/show_bug.cgi?id=738528? like to unstick that one way or another
# [00:04] * Joins: Mano (mano@5A193F33.5F4A4453.39716721.IP)
# [00:04] * Joins: armenzg (armenzg@62D3EA85.3CBDB8AB.DA78B690.IP)
# [00:06] <jtcranmer> oh dear
# [00:06] <jtcranmer> perl's regular expressions give you pretty much CFG in one bat
# [00:06] <jtcranmer> (?<group num>) if you were wondering
# [00:07] <jwir3> does presShell->GetRootFrame() always return the same root frame? E.g. if I want to detect if a frame is the root frame, is it sufficient to compare it to the return value of nsPresShell::GetRootFrame() ?
# [00:08] * Quits: armenzg (armenzg@62D3EA85.3CBDB8AB.DA78B690.IP) (Ping timeout)
# [00:09] * Joins: teoli (teoli@A01051A9.6F669779.3D1CA460.IP)
# [00:09] * Quits: joey (chatzilla@moz-EFCB4CBF.princetowncable.com) (Quit: ChatZilla 0.9.88.2 [Firefox 14.0a2/20120428042005])
# [00:09] <sicking> jprmc: looking
# [00:10] <jprmc> sicking: thanks
# [00:10] * Quits: madhava (madhava@F2D29657.F60B0462.67AC9B1.IP) (Quit: madhava)
# [00:11] * Joins: armenzg (armenzg@62D3EA85.3CBDB8AB.DA78B690.IP)
# [00:13] * Quits: pnemsak (Miranda@80CFE454.10D9684B.4F33160D.IP) (Quit: pnemsak)
# [00:13] * Quits: jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP) (Ping timeout)
# [00:13] * Quits: harth (harth@88F51059.F3BBB17D.144F44FA.IP) (Input/output error)
# [00:13] * Quits: armenzg (armenzg@62D3EA85.3CBDB8AB.DA78B690.IP) (Ping timeout)
# [00:15] * Joins: mconley (mconley@moz-D640D16C.cable.teksavvy.com)
# [00:16] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [00:17] * Joins: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [00:17] * Quits: jimb (user@9E727688.17C2F9C.163DC5C6.IP) (Ping timeout)
# [00:18] * Quits: Mano (mano@5A193F33.5F4A4453.39716721.IP) (Client exited)
# [00:18] * Quits: ajuma (ajuma@F2D29657.F60B0462.67AC9B1.IP) (Quit: ajuma)
# [00:19] <krit> AryehGregor: ping
# [00:19] * Joins: joe_walker_ (joe_walker@moz-15405DDA.cable.virginmedia.com)
# [00:20] * Quits: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [00:20] * Joins: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [00:21] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [00:22] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [00:22] * Joins: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [00:23] * Joins: bholley (anonymous@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [00:24] * Quits: Standard8 (Standard8@B7F1AE36.48015583.54C3481B.IP) (Quit: ZNC - http://znc.in)
# [00:25] * Parts: teoli (teoli@A01051A9.6F669779.3D1CA460.IP)
# [00:25] * Joins: cpearce (chatzilla@538BABFE.A073F3E.97BBD552.IP)
# [00:26] * Joins: nrc (nrc@538BABFE.A073F3E.97BBD552.IP)
# [00:27] * Quits: Mardak (Mardak@moz-7B0110AD.mv.mozilla.com) (Quit: Mardak)
# [00:27] * Joins: asadotzler (asa@moz-BBE3ABD.mv.mozilla.com)
# [00:28] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [00:29] * Joins: zzzzz_ (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [00:30] * Joins: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [00:30] * bear is now known as bear-afk
# [00:31] * Quits: zzzzz_ (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [00:33] * Quits: mjschranz (mjschranz@F2D29657.F60B0462.67AC9B1.IP) (Client exited)
# [00:33] * Quits: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP) (Quit: Computer has gone to sleep.)
# [00:34] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [00:34] * Joins: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [00:35] * Quits: squib (squib-@moz-3F6F2A9C.ep.wisc.edu) (Quit: Leaving)
# [00:36] <mwu> jhford: how much do you know about the tegras?
# [00:36] <jhford> i don't work on them
# [00:36] <mwu> hm ok
# [00:36] <jhford> what did you want to know?
# [00:36] * Joins: loadbang (loadbang@moz-28BBF200.range86-162.btcentralplus.com)
# [00:37] * Quits: wlach (wlach@moz-14AFFC14.vif.net) (Quit: Leaving)
# [00:37] <mwu> just wondering if I just managed to corrupt the filesystem on the tegra
# [00:38] * Quits: Waldo (waldo@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [00:38] * Joins: mjschranz (mjschranz@F2D29657.F60B0462.67AC9B1.IP)
# [00:38] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [00:39] * Joins: zzzzz_ (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [00:39] * zzzzz_ is now known as zzzzz
# [00:40] * Quits: dseif (dseif@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [00:41] * Quits: mjschranz (mjschranz@F2D29657.F60B0462.67AC9B1.IP) (Client exited)
# [00:41] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [00:42] * Joins: jfkthame (jfkthame@15AA2040.A1C12133.9542EC20.IP)
# [00:42] <taras> bz: ping
# [00:42] <mwu> bear-afk: ping
# [00:42] * mcote is now known as mcote|afk
# [00:42] <bear-afk> which tegra?
# [00:42] * Quits: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Input/output error)
# [00:42] * Joins: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [00:43] <bear-afk> mwu - what tegra?
# [00:43] * Joins: jprmc (jprmc@A21EB03E.3CBDB8AB.DA78B690.IP)
# [00:43] * Quits: Metapioca (lg@moz-548650B4.rev.sfr.net) (Quit: Lost terminal)
# [00:43] * Joins: wlach (wlach@moz-14AFFC14.vif.net)
# [00:43] * mwu checks
# [00:43] * Quits: kaze (kaze@moz-7E0F0F9E.fbx.proxad.net) (Quit: marre)
# [00:44] * Joins: dseif (dseif@F2D29657.F60B0462.67AC9B1.IP)
# [00:44] <mwu> bear-afk: 041 - https://tbpl.mozilla.org/php/getParsedLog.php?id=11337454&tree=Firefox&full=1
# [00:44] <taras> roc: around?
# [00:44] <bear-afk> mwu - let me check
# [00:45] * Joins: mjschranz (mjschranz@F2D29657.F60B0462.67AC9B1.IP)
# [00:45] <mwu> probably something in /mnt/sdcard
# [00:45] <taras> cjones: around?
# [00:45] <philor> nah, that's just something we do now, and haven't gotten around to filing yet
# [00:46] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [00:46] <philor> because it will be fixed by magical unicorn dust
# [00:46] <mwu> ah ok
# [00:46] * Quits: mcsmurf (mcsmurf@moz-E3C18463.dip.t-dialin.net) (Quit: )
# [00:46] <bear-afk> mwu - that's just us being more verbose about what the failure is
# [00:46] <bear-afk> we have code that for many small reasons we just can't manage to "stick" when we land it
# [00:46] * Joins: anky (anky@C772EAAE.744E7036.A3D1B221.IP)
# [00:46] <bear-afk> that will start cleaning those up auto-magically
# [00:46] <mwu> bear-afk: well, as long as there aren't any files named £≈tΓ░i¢.█C
# [00:46] * Joins: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [00:47] <cjones> hey taras
# [00:47] <mwu> mm
# [00:47] <pcwalton> cjones: what do you think about retaining layer trees in background tabs? taras tells me that google chrome does this to some extent
# [00:47] <pcwalton> and obviously helps tab switching snappiness
# [00:48] <cjones> what problem are you trying to solve?
# [00:48] * Quits: beaufour (beaufour@18D5CC88.C7EE4FB2.ECED8BE3.IP) (Quit: beaufour)
# [00:48] <philor> mwu: you might still be busting things, though
# [00:48] <pcwalton> not all background tabs, but if we can predict which ones are most likely to be switched to...
# [00:48] <joe> we cannot
# [00:48] <mwu> philor: seems likely with the talos redness
# [00:48] <pcwalton> cjones: most tab switching jank is rerendering pages
# [00:48] <@roc> taras: yes
# [00:48] <pcwalton> seems like retained layers could help here
# [00:48] <philor> mwu: yeah, the set on smaug, the "browser failed to close after being initialized" is unknown
# [00:48] <@roc> how about we just speed up page rendering
# [00:48] <taras> roc: see pcwalton comment above
# [00:48] <joe> roc: +1
# [00:49] <@roc> see my comment above
# [00:49] <mwu> bear-afk: any idea what's going on with the talos red?
# [00:49] <taras> roc: we seem to render pages faster if i just switch between 2 tabs
# [00:49] <cjones> pcwalton, it would be easier to snapshot on switch away
# [00:49] <taras> and slower if more
# [00:49] <pcwalton> taras: is it image decoding in your benchmarks or what?
# [00:49] <@roc> we can retain layer trees pretty easily --- if you can come up with a heuristic that doesn't destroy memory usage
# [00:49] <taras> roc: is there some knob to tune?
# [00:50] <pcwalton> yeah, memory usage is obviously the big worry here
# [00:50] <@roc> taras: is that image decoding maybe?
# [00:50] <pcwalton> I'd like to know precisely what google chrome does
# [00:50] <@roc> "use a lot of memory" perhaps
# [00:50] <taras> roc: my image cache is set to 500mb
# [00:50] <taras> so i doubt it
# [00:50] <philor> oh, hello rnewman
# [00:50] <cjones> tab switch involves a lot of stuff
# [00:50] <cjones> lots of things can be slow
# [00:51] <@roc> I think there's a lot we could do to speed up page rendering
# [00:51] <cjones> pcwalton, what's the advantage of keeping a layer tree over a snapshot?
# [00:51] <@roc> that wouldn't require space/time tradeoffs
# [00:51] <joe> please do not jump to solutions before exploring what the problem space is
# [00:51] <pcwalton> cjones: nothing, I just mean the general idea of snapshots
# [00:51] <cjones> generally snapshotting is the only way for us to hit a re-render target on tab switch
# [00:51] <cjones> and not even that is guaranteed in desktop
# [00:52] <taras> roc: i'm trying to get a complete picture of why our tab strip performs so poorly
# [00:52] <taras> in particular for tab switching
# [00:52] <@roc> good
# [00:52] <philor> m-c is closed because I don't trust us
# [00:52] * Quits: eflores (eflores@538BABFE.A073F3E.97BBD552.IP) (Connection reset by peer)
# [00:52] <taras> so if you could make a list of things we could to help
# [00:52] * Joins: eflores (eflores@538BABFE.A073F3E.97BBD552.IP)
# [00:52] <joe> taras: measure everything :)
# [00:52] <rnewman> philor: what did I do?
# [00:52] <taras> so far i know the main diff is that webkit keeps displaylists for every tab
# [00:53] <mwu> if most of xul opt survives, it probably isn't rnewman's fault
# [00:53] <joe> and differentiate between page drawing time and tab strip drawing time
# [00:53] * Quits: TheLink (TheLink@moz-2C115FCB.pools.arcor-ip.net) (Client exited)
# [00:53] <taras> joe: don't worry, i always measure
# [00:53] <rnewman> landed equivalent change on Aurora with no real issue: https://tbpl.mozilla.org/?tree=Mozilla-Aurora&rev=a3a4c8efa239
# [00:53] <taras> but having an idea of wtf i'm measuring helps :)
# [00:53] <rnewman> if you're referring to https://tbpl.mozilla.org/?rev=23b86b775f6f
# [00:53] <mwu> yeah
# [00:54] * Quits: wlach (wlach@moz-14AFFC14.vif.net) (Input/output error)
# [00:54] <mwu> our tegras probably sploded then
# [00:54] <taras> joe: i know tab strip perf is a small part of page switching
# [00:54] <taras> and it depends on complexity of the page being switched to
# [00:54] <taras> s/small/relatively small/
# [00:54] <joe> in that case, i'd focus on what we spend our page draw time on
# [00:55] <joe> and utterly ignore the tab strip
# [00:55] * Quits: Cwiiis (cwiiis@moz-F15E698.croy.cable.virginmedia.com) (Quit: Leaving)
# [00:55] <rnewman> philor: also, later push https://tbpl.mozilla.org/?rev=992588c2eab6 is largely green
# [00:56] <philor> rnewman: the things that are not green on that push are exactly what I'm worried about
# [00:56] <pcwalton> I think we should measure more, but we shouldn't rule snapshotting out
# [00:56] <philor> dunno, maybe we build a11y on Android?
# [00:56] * Joins: wlach (wlach@moz-14AFFC14.vif.net)
# [00:56] <pcwalton> because if chrome is snapshotting and we aren't, then we'll always lose on complex pages
# [00:56] <joe> i don't think anybody is ruling anything out
# [00:56] * Quits: bholley (anonymous@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Quit: bholley)
# [00:56] <pcwalton> that's fair
# [00:57] <joe> making painting faster is unlikely to make our user experience worse though
# [00:57] <taras> joe: i would like to see more input on possible solutions other than "measure"
# [00:57] <joe> taras: i have no idea what our problems are, so I can offer no input on solutions
# [00:57] <rnewman> philor: looks like a network/sutagent thing
# [00:57] * Quits: @smaug (chatzilla@moz-AD13B184.elisa-mobile.fi) (Ping timeout)
# [00:58] <joe> in my experience, our problems are generally correlated with the solutions to those problems
# [00:58] <philor> rnewman: "utils.talosError: 'browser failed to close after being initialized'"?
# [00:58] * Quits: drice (derice@1606D15F.E628B196.8E155D4E.IP) (Quit: Leaving.)
# [00:58] * Quits: mjschranz (mjschranz@F2D29657.F60B0462.67AC9B1.IP) (Client exited)
# [00:58] <rnewman> ah, earlier in the log
# [00:58] * joduinn-mtg is now known as joduinn-brb
# [00:59] <philor> I'd expect that to be a more shutdown hang or totally busted kind of thing
# [00:59] * Quits: robhawkes (robhawkes@moz-33A339B7.dsl.cnl.uk.net) (Quit: Leaving...)
# [00:59] <rnewman> if this is being built as org.mozilla.fennec, the change I commit should have a net zero effect on the output source
# [00:59] * joe afk
# [01:00] <rnewman> 04-30 14:42:50.215 W/System.err( 1438): java.io.FileNotFoundException: /mnt/sdcard/tests/fennec-15.0a1.en-US.android-arm.apk (No such file or directory)
# [01:00] <rnewman> ...!
# [01:00] <rnewman> these logs are not reassuring at all
# [01:00] <rnewman> how can you spot a real problem?!
# [01:00] <philor> step 1: read literally thousands of them
# [01:00] <rnewman> heh
# [01:00] * Quits: jfkthame (jfkthame@15AA2040.A1C12133.9542EC20.IP) (Quit: jfkthame)
# [01:01] <philor> but don't you mean net zero effect if it's being built as org.mozilla.gecko?
# [01:01] <rnewman> yes, sorry
# [01:01] <philor> which, um, it wouldn't be
# [01:01] <rnewman> which it is not
# [01:02] <rnewman> give me a moment to investigate this
# [01:02] * Joins: Waldo (waldo@moz-BBE3ABD.mv.mozilla.com)
# [01:03] * Quits: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP) (Ping timeout)
# [01:03] * Quits: zuzelvp (zuzelvp@2112147D.C3507A2D.9A8C35B4.IP) (Ping timeout)
# [01:05] * Quits: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Ping timeout)
# [01:07] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [01:08] * Quits: Boriss (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: Boriss)
# [01:08] <Bas> taras: Yes, a bunch, today's a public holiday here, but I'll be sending an e-mail. Profiling shows (for chrome, i.e. tab switching with -nothing- in the tabs), D2D is a -lot- more expensive than GDI, and there's several reasons, I already discussed it a little with Jeff.
# [01:08] <rnewman> I can't find anywhere that the string name of that process matters
# [01:08] <Bas> taras: We have several options to address although I'd like to get Azure-Thebes on-by-default first so we only need to address them once. A bunch of the stuff that needs to happen is also in layout.
# [01:08] <taras> Bas: ok, thanks
# [01:08] * Joins: harth (harth@moz-C2C235AC.bb.sky.com)
# [01:09] * Quits: Pike (Pike@moz-E31CD2CB.mozilla.org) (Quit: ChatZilla 0.9.88.2 [Firefox 12.0/20120420145725])
# [01:09] <taras> Bas: i'm also looking into content swithcing perf too
# [01:09] * Quits: karl (karl@moz-9B5DFBA8.jetstream.xtra.co.nz) (Ping timeout)
# [01:09] <taras> though seems that with gfx accel
# [01:09] <philor> well, we were unbusted on 0a796d07499a and we aren't now, and I know how to fix that
# [01:09] <taras> chrome is actually worse than content
# [01:09] <Bas> taras: If we want to be better purely by changing the UI, the basic story is: Remove gradients, make all borders -single- color.
# [01:10] <taras> Bas: i got good news..ui folks are going to add more svg and more rounded borders
# [01:10] * Joins: Boriss (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [01:10] * gregglind is now known as gregglind_away
# [01:10] * jlund|afk is now known as jlund|away
# [01:11] * Quits: NeilAway (neil@moz-32AA0D01.in-addr.btopenworld.com) (Ping timeout)
# [01:11] <taras> Bas: are bordrs inherently slow?
# [01:11] <taras> or is it our implementation
# [01:11] <Bas> taras: Mostly our imlementation.
# [01:11] * Quits: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [01:11] <Bas> SVG is more or less inherently slow.
# [01:12] <Bas> But also partially our implementation, not on the GFX side though.
# [01:12] * Joins: NeilAway (neil@moz-32AA0D01.in-addr.btopenworld.com)
# [01:12] <Waldo> web 1.0 borders: fast; web 2.0 bordrs: slow; progress!
# [01:12] * coop|mtg is now known as coop
# [01:12] <taras> Bas: so...if you take out svg..borders are bad because ____?
# [01:13] * Joins: RyanVM (chatzilla@moz-D04D3C77.phlapa.fios.verizon.net)
# [01:13] <Bas> taras: Because we jump through 7 hoops to draw them -exactly- according to the spec, unlike anyone else.
# [01:13] * Quits: asadotzler (asa@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [01:14] * Quits: smooney (smooney@moz-BBE3ABD.mv.mozilla.com) (Quit: smooney)
# [01:14] <taras> Bas: can we not draw according to spec?
# [01:14] <taras> :)
# [01:15] <Waldo> getting tests for this stuff upstreamed to w3c would probably help with that
# [01:15] <taras> Bas: anyway, i dont want to waste you time off
# [01:16] <taras> thanks for the bits of info, will wait on email
# [01:16] * joduinn-brb is now known as joduinn-mtg
# [01:17] * AutomatedTester is now known as AutomatedTester|AFK
# [01:18] * Joins: smooney (smooney@moz-BBE3ABD.mv.mozilla.com)
# [01:20] * Quits: joe_walker_ (joe_walker@moz-15405DDA.cable.virginmedia.com) (Quit: Leaving)
# [01:20] * Joins: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [01:20] * Quits: smagnin (pike@moz-1816673.w90-57.abo.wanadoo.fr) (Client exited)
# [01:21] * Joins: KLB (Kenneth_Ba@moz-C233EF2B.maine.res.rr.com)
# [01:21] * Quits: tchevalier (Instantbir@moz-BA7C7754.w90-0.abo.wanadoo.fr) (Quit: Instantbird -- http://www.instantbird.com)
# [01:21] * Joins: darktrojan (geoff@moz-30B3CCFD.telstraclear.net)
# [01:23] * AutomatedTester|AFK is now known as AutomatedTester
# [01:24] <taras> billm: is there a way to force incremental gc?
# [01:25] <Bas> taras: We can, and in my opinion should, not draw according to spec, and just seam.
# [01:25] * Joins: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com)
# [01:25] <Bas> Roc's also exploring some other options involving totally changing our drawing methods.
# [01:25] <Bas> But that's very early in the process.
# [01:25] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [01:26] * Quits: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [01:27] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [01:27] * Quits: TheCrap (TheCrap@moz-3BC7BEEC.dip0.t-ipconnect.de) (Ping timeout)
# [01:27] * Joins: ilya1 (Adium@moz-CD91E596.google.com)
# [01:28] * ilya1 is now known as isherman
# [01:28] * Quits: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com) (Ping timeout)
# [01:29] <rnewman> anyone know where the talos run_tests.py lives? doesn't seem to be in m-c
# [01:29] <isherman> bz: Wanted to get your thoughts on [ https://bugs.webkit.org/show_bug.cgi?id=85150 ] — the related Gecko bug is [ https://bugzilla.mozilla.org/show_bug.cgi?id=740979 ].
# [01:30] <isherman> bz: fantasai suggested that you might be a good person to talk to
# [01:30] * Quits: Mnyromyr (Mnyromyr@B2521176.7B0892CB.771966F7.IP) (Quit: ChatZilla 0.9.86 [SeaMonkey 1.1.19/2010030105])
# [01:30] <philor> rnewman: http://mxr.mozilla.org/build/source/talos/talos/run_tests.py
# [01:31] <taras> Bas: so can i get that email today?
# [01:31] * catlee is now known as catlee-away
# [01:31] <isherman> bz: basically, we'd like for the UA's styling of the :-autofill pseudo-class to override author styling of lower specificity, as this is the commonly useful case
# [01:31] * Quits: Xaquseg (xaquseg@moz-D4633F43.pixelhunger.com) (Quit: disconnecting)
# [01:31] <rnewman> philor: ta
# [01:32] * Quits: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se) (Ping timeout)
# [01:32] <jhammel> rnewman: looking for anything in particular?
# [01:32] <isherman> bz: but we'd like to give authors the flexibility to intentionally override the :-autofill styling if they really insist, since they can do so using JS hackery anyway
# [01:32] <rnewman> jhammel: I'm wondering how talos determines that the browser has closed
# [01:32] <isherman> bz: Any suggestions on how to approach this?
# [01:32] <rnewman> because Android is in control of some of Fennec's processes
# [01:32] * Joins: jamesr (jamesr@moz-3E81052.red-213-96-0.staticip.rima-tde.net)
# [01:32] <philor> because he broke it
# [01:32] <rnewman> and I just renamed one of them to org.mozilla.fennec.PasswordsProvider from org.mozilla.gecko.PasswordsProvider, so now the package name is a prefix
# [01:32] <jhammel> rnewman: oh, in bad ways :(
# [01:33] <rnewman> so the behavior is unchanged, but now talos thinks Fennec isn't closing
# [01:33] <jhammel> and fennec even less reliably :(
# [01:33] <rnewman> if talos is checking for string prefix names or something, we need to change that
# [01:33] <rnewman> or parent PIDs
# [01:33] <isherman> bz: At first blush, it looks like the CSS cascading order [ http://www.w3.org/TR/CSS21/cascade.html#cascading-order ] explicitly disallows this… but it seems sucky to force site authors to resort to JS hackery for what is really a styling issue
# [01:34] <jhammel> rnewman: so ffprocess.py and ffprocess_remote would be the places to look
# [01:34] * Quits: smooney (smooney@moz-BBE3ABD.mv.mozilla.com) (Quit: smooney)
# [01:34] <rnewman> 94 def ProcessesWithNames(self, *process_names):
# [01:34] <rnewman> 95 """Returns a list of processes running with the given name(s).
# [01:34] <rnewman> urgh
# [01:34] <rnewman> OK, so this *should* have broken months ago, if it wasn't for a naming accident
# [01:34] <jhammel> likely
# [01:35] <jaws> dholbert: ping?
# [01:35] <dholbert> jaws, pong
# [01:36] * Quits: anky (anky@C772EAAE.744E7036.A3D1B221.IP) (Client exited)
# [01:36] <jaws> dholbert: if a container is display:-moz-box and -moz-box-orient:vertical, and the body has dir=rtl, the elements are mirrored vertically. i'm about to file this as a bug, but wanted to run it by you first. thoughts?
# [01:36] * Quits: mconley (mconley@moz-D640D16C.cable.teksavvy.com) (Input/output error)
# [01:37] * jaws is working on a reftest for this situation
# [01:37] <dholbert> jaws, I'd say that's a bug, yeah
# [01:37] <dholbert> jaws, though -moz-box behavior is sort of defined by the implementation, rather than by any spec :)
# [01:37] <billm> taras: in what setting?
# [01:37] <dholbert> jaws, but I doubt that's the intended behavior
# [01:37] <rnewman> jhammel: we can't even rename ours to be foo.$apppackage.whatever, because the RE is so general :/
# [01:37] <jaws> dholbert: ah ok, well then i will file the bug :)
# [01:38] <rnewman> gah
# [01:38] <dholbert> jaws, thanks!
# [01:38] * rnewman contemplates reversing the string
# [01:38] <romaxa> bsmedberg: ping
# [01:38] <jhammel> rnewman: could you file a bug? im mostly only guessing what you're talking about
# [01:38] <mwu> whew, so I didn't destroy the test machines
# [01:38] <rnewman> got a preferred component, jhammel?
# [01:38] <rnewman> this it outside my realm
# [01:38] <jhammel> Testing::Talos
# [01:38] <rnewman> ta
# [01:38] <firebot> Check-in: http://hg.mozilla.org/mozilla-central/rev/6e34995a746e - Phil Ringnalda - Back out 23b86b775f6f (bug 749727) for Android Talos bustage on a CLOSED TREE
# [01:38] <jhammel> thank you :)
# [01:39] * Joins: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP)
# [01:39] * Joins: smooney (smooney@moz-BBE3ABD.mv.mozilla.com)
# [01:39] * AutomatedTester is now known as AutomatedTester|AFK
# [01:40] * Quits: jdm (jdm@moz-D662800A.cm-7-1b.dynamic.ziggo.nl) (Client exited)
# [01:41] * Joins: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [01:41] <taras> billm: is there a way to force igc all the time?
# [01:42] <taras> even if it leaks
# [01:43] * AutomatedTester|AFK is now known as AutomatedTester
# [01:44] <rnewman> jhammel: thanks: https://bugzilla.mozilla.org/show_bug.cgi?id=750548
# [01:45] <jhammel> rnewman: thanks; and bonus points for starting the bug with 'alas' ;)
# [01:45] <rnewman> :D
# [01:46] <rnewman> you should see how many bugs I've filed titled "More considered handling of ..."
# [01:46] * Quits: jprmc (jprmc@A21EB03E.3CBDB8AB.DA78B690.IP) (Ping timeout)
# [01:46] <billm> taras: well, there's an about:config option. search for gc_incremental
# [01:46] <rnewman> philor: thanks for backout, will reland with some awful hack
# [01:47] * Joins: davidb (davidb@moz-68BF56A6.dsl.bell.ca)
# [01:47] * Joins: lmandel (lmandel@FE1F74.86ED00A7.971E19F6.IP)
# [01:47] <billm> taras: also, the leak is fixed now
# [01:48] * coop is now known as coop|afk
# [01:48] <philor> org.m0zilla.f3nn3c would be nice
# [01:48] <rnewman> philor: :D
# [01:48] <philor> does the preprocessor have a l33t option?
# [01:48] <rnewman> need a preprocessor directive for l33t
# [01:48] <rnewman> lol
# [01:48] <rnewman> ^5
# [01:48] * Quits: davidb (davidb@moz-68BF56A6.dsl.bell.ca) (Input/output error)
# [01:49] * Joins: squib (squib@moz-B01B5D55.dhcp.mdsn.wi.charter.com)
# [01:50] * Quits: nhirata (nhirata.bu@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: nhirata)
# [01:50] * Quits: dseif (dseif@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [01:52] * Joins: jprmc (jprmc@346FA63F.3CBDB8AB.DA78B690.IP)
# [01:52] * Joins: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com)
# [01:53] * Joins: birtles (chatzilla@moz-4B5534A0.mozilla.or.jp)
# [01:53] <taras> billm: cool, is there an eta for IGC being turned back on?
# [01:54] <billm> taras: a few days. I want to look at some memory usage issues that showed up last time it was enabled.
# [01:55] <taras> billm: cool, happy to hear that the leak is out of the way
# [01:55] * wlach is now known as wlach|afk
# [01:55] <billm> taras: yeah, it's a big relief :-)
# [01:55] * Quits: jprmc (jprmc@346FA63F.3CBDB8AB.DA78B690.IP) (Ping timeout)
# [01:56] <mwu> kinetik: ping
# [01:57] * Quits: WG9s (bill@moz-7A06A043.hsd1.ma.comcast.net) (Quit: ChatZilla 0.9.87-4.1450hg.fc16 [XULRunner 12.0/20120424094117])
# [01:58] <kinetik> mwu: hi
# [01:58] <mwu> hey
# [01:58] * Quits: IanN (chatzilla@moz-3F5A461C.cable.virginmedia.com) (Quit: ChatZilla 0.9.88.2 [SeaMonkey 2.12a1/20120428142341])
# [01:58] <mwu> so I was wondering if you're still working on the pulseaudio backend
# [01:59] * heycam|away is now known as heycam
# [01:59] <kinetik> mwu: i haven't had a chance to do anything since we talked about it last, but i plan to get back to it eventually
# [01:59] <mwu> anything I can do to speed things up?
# [02:00] * Joins: njn (chatzilla@moz-71689D29.dyn.iinet.net.au)
# [02:00] * Joins: karl (karl@538BABFE.A073F3E.97BBD552.IP)
# [02:00] <kinetik> mwu: was it just the deadlock fix you need, or was there something I'm forgetting?
# [02:01] <mwu> the deadlock thing is the only fix I really need above what's currently available
# [02:01] <taras> Bas: it would be handy to have a flag to punt on drawing borders
# [02:01] * jhford-work is now known as jhford-work-away
# [02:02] * jaws is now known as jaws|away
# [02:02] <mwu> at least right now. we'll probably also want a pa backend for audiomanager for controlling volume and such
# [02:02] <mwu> but I can probably handle that
# [02:03] * Quits: billm (billm@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving)
# [02:03] * Joins: billm (billm@moz-BBE3ABD.mv.mozilla.com)
# [02:03] * Quits: jamesr (jamesr@moz-3E81052.red-213-96-0.staticip.rima-tde.net) (Quit: jamesr)
# [02:04] <mwu> kinetik: just landing the backend might be helpful too - I might be able to look into fixing the deadlock
# [02:04] <kinetik> mwu: i haven't looked too deeply, but i'm reasonably sure there's no safe way to make PA drop it's mainthread loop lock for us, so i think it'll be necessary to add a queue of refill requests and run through it outside of PA's callback
# [02:04] <kinetik> mwu: landing it is easy... want to review it? :-)
# [02:05] <mwu> hah
# [02:05] <mwu> sure, as long as that doesn't upset anyone
# [02:06] <mwu> I need to learn the pa api for real anyway
# [02:06] * Quits: Asa (asa@moz-BBE3ABD.mv.mozilla.com) (Input/output error)
# [02:06] <kinetik> i don't think anyone is clamoring to review it, and you seem like a good choice since you've been working with PA
# [02:06] <mwu> ok, go for it
# [02:06] <kinetik> i'll put up a patch today/tomorrow, i won't wire it into the build system at all yet
# [02:06] <mwu> that's fine
# [02:06] * Joins: jimb (user@moz-F4EC06CC.hsd1.or.comcast.net)
# [02:07] <mattwoodrow> bz: You worked on making sure adding table rows/columns didn't invalidate the whole table right?
# [02:07] <kinetik> excellent
# [02:08] * Quits: jhammel (jhammel@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: leaving)
# [02:09] * Quits: Waldo (waldo@moz-BBE3ABD.mv.mozilla.com) (Quit: ChatZilla 0.9.87-4.1450hg.fc15 [XULRunner 11.0/20120313114635])
# [02:10] <darktrojan> "A new tool called Collusion for Chrome shows users all the websites that secretly track you as you browse the Web" ... um, isn't collusion ours?
# [02:11] * Quits: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net) (Ping timeout)
# [02:12] <Mook_as> Collusion for Chrome isn't. https://chrome.google.com/webstore/detail/ganlifbpkcplnldliibcbegplfmcfigp
# [02:12] <dholbert> darktrojan, assuming you're reading http://gizmodo.com/5902855/collusion-for-chrome-reveals-whos-spying-on-your-surfing -- note that the video there mentions that the original collusion was for firefox, and they want to make something similar for chrome
# [02:12] <dholbert> *wanted
# [02:13] <philor> Opera probably had it first anyway
# [02:13] <dholbert> darktrojan, (kind of annoying that the article never mentions mozilla/firefox/the original collusion, though
# [02:13] <dholbert> )
# [02:13] <darktrojan> no, I'm reading something very similar on RWW
# [02:13] <dholbert> philor, probably true :D
# [02:13] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [02:13] <darktrojan> someone's clearly pushing it this week
# [02:13] * Quits: sfink (chatzilla@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [02:13] * Quits: bdahl (bdahl@moz-BBE3ABD.mv.mozilla.com) (Client exited)
# [02:13] * AutomatedTester is now known as AutomatedTester|AFK
# [02:14] <darktrojan> this article does actually mention it
# [02:14] * AutomatedTester|AFK is now known as AutomatedTester
# [02:14] * Quits: jimb (user@moz-F4EC06CC.hsd1.or.comcast.net) (Ping timeout)
# [02:14] * Quits: victorporo (victorporo@5F6B43D3.1C233438.79933D60.IP) (Quit: Linkinus - http://linkinus.com)
# [02:15] <bsmith> What is the currently-recommended way to add thread-safe AddRef()/Release() to a non-nsISupports class?
# [02:16] * Quits: @ehsan (ehsan@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [02:16] <@roc> NS_INLINE_DECL_THREADSAFE_REFCOUNTING
# [02:16] <bsmith> thanks roc.
# [02:16] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [02:17] * Joins: zuzelvp (zuzelvp@2112147D.C3507A2D.9A8C35B4.IP)
# [02:19] <philor> rnewman: what was that you said about having landed on aurora and everything was fine?
# [02:20] * Quits: garnacho (carlos@moz-EA2D0541.dyn.user.ono.com) (Ping timeout)
# [02:21] * nthomas is now known as nthomas|away
# [02:21] <rnewman> I am quite probably wrong!
# [02:21] <rnewman> kinda glaze over some android failures
# [02:21] <rnewman> fixing now
# [02:21] * Joins: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net)
# [02:21] <rnewman> (as in, thirty seconds)
# [02:22] <philor> charmingly, the difference between totally busted and completely normal is nearly impossible to see with the untrained eye
# [02:22] * Quits: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net) (Client exited)
# [02:22] * Joins: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com)
# [02:22] <rnewman> yeah, not really working in my favor :D
# [02:22] * AutomatedTester is now known as AutomatedTester|AFK
# [02:22] * Quits: birtles (chatzilla@moz-4B5534A0.mozilla.or.jp) (Ping timeout)
# [02:22] <rnewman> thanks for being ever-watchful!
# [02:22] <philor> or the trained eye, since I quite often just keep starring right through "the browser won't even start" for a push or two
# [02:23] * Quits: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com) (Input/output error)
# [02:23] * Joins: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com)
# [02:24] <rnewman> maybe we should hire some people to sit at a bank of phones
# [02:24] <rnewman> manually testing every push
# [02:24] <rnewman> heh
# [02:24] <ekr> mechanical turk!
# [02:24] <rnewman> Mechanical Sucker
# [02:25] <philor> talos still has some porn in it, doesn't it?
# [02:25] <rnewman> how did I manage to do both two hours of triage *and* makefile hacking in one awful day?
# [02:25] <rnewman> heh
# [02:26] * Quits: krit (krit@moz-1FC1932F.adobe.com) (Quit: Leaving.)
# [02:27] <fryn> does anyone know what would cause an attempt to launch Firefox to result in "The program can't launch because mozglue.dll is missing from the program"
# [02:27] <fryn> aside from mozglue.dll simply being deleted by a user or application w/ access to the filesystem.
# [02:28] * philor backspaces through blaming it on antivirus
# [02:28] <philor> AV having it locked during startup?
# [02:28] <fryn> yeah, i immediately guessed antivirus too.
# [02:28] * Joins: victorporof (victorporo@5F6B43D3.1C233438.79933D60.IP)
# [02:29] * Quits: victorporof (victorporo@5F6B43D3.1C233438.79933D60.IP) (Client exited)
# [02:30] * Joins: victorporof (victorporo@5F6B43D3.1C233438.79933D60.IP)
# [02:32] * Joins: clokep (Instantbir@moz-69FB3955.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com)
# [02:35] <@dolske> apparently common enoguh that google autocomplete knows about it. http://cl.ly/1r1i000p0Y0R0D3e1a2W
# [02:37] <JonathanS> dolske, so google know everything?
# [02:37] <@dolske> fryn: have you tried this entirely legit and not at all sketchy advice? http://cl.ly/0t0w450L133p3L0R1z3j
# [02:37] * Quits: eflores (eflores@538BABFE.A073F3E.97BBD552.IP) (Client exited)
# [02:38] * Joins: eflores (eflores@538BABFE.A073F3E.97BBD552.IP)
# [02:38] <fryn> dolske: yikes! (about it being common enough)
# [02:38] * mdas is now known as mdas|afk
# [02:38] * Quits: akeybl (akeybl@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving...)
# [02:38] <fryn> dolske: it's occurring on a UR participant's machine, so, no, i will not suggest that.
# [02:39] * Joins: ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP)
# [02:39] * ChanServ sets mode: +o ehsan
# [02:41] * Quits: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [02:42] * Quits: @ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP) (Ping timeout)
# [02:43] * Joins: ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP)
# [02:43] * ChanServ sets mode: +o ehsan
# [02:44] * Joins: ehsan_ (ehsan@BEDFFF26.EE049E25.8B035CD7.IP)
# [02:45] * Joins: akeybl (akeybl@moz-BBE3ABD.mv.mozilla.com)
# [02:45] * Quits: @ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP) (Ping timeout)
# [02:46] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [02:46] <njn> is it possible to run |hg qref -e| on a patch that's not at the top of your stack?
# [02:47] <RyanVM> i keep racing with philor
# [02:47] * Quits: akeybl (akeybl@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving...)
# [02:47] * Quits: ehsan_ (ehsan@BEDFFF26.EE049E25.8B035CD7.IP) (Input/output error)
# [02:48] * Quits: ircloggr (nodebot@moz-9B5D748D.compute-1.amazonaws.com) (Client exited)
# [02:48] <philor> RyanVM: I usually give up pretty quickly, though ;)
# [02:49] * Joins: ircloggr (nodebot@moz-E155841C.compute-1.amazonaws.com)
# [02:49] <RyanVM> heh
# [02:49] <RyanVM> i'm only watching inbound, fwiw
# [02:49] * Quits: jammink (textual@moz-BBE3ABD.mv.mozilla.com) (Quit: Textual IRC Client: http://www.textualapp.com/)
# [02:50] * Joins: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [02:50] * Quits: AaronMT (AaronMT@moz-E26428A8.cpe.net.cable.rogers.com) (Quit: Textual IRC Client: http://www.textualapp.com/)
# [02:51] * Joins: dseif (dseif@CE526453.531FD64C.630E4E47.IP)
# [02:53] * Quits: smooney (smooney@moz-BBE3ABD.mv.mozilla.com) (Quit: smooney)
# [02:54] * mdas|afk is now known as mdas
# [02:54] <Mook_as> njn: seems easier to make a new one, move things around, and qfold.
# [02:55] * Joins: jet (junglecode@moz-79F891EE.hsd1.ca.comcast.net)
# [02:55] * Quits: Mook_as (mook@moz-1FCC0032.activestate.com) (Quit: gone)
# [02:55] <njn> Mook_as: that sounds just as sub-optimal as qpop'ing down :(
# [02:56] * Quits: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com) (Quit: bmoss)
# [02:57] * Joins: bc_ (bc@moz-E31CD2CB.mozilla.org)
# [02:57] * Quits: bc (bc@moz-E31CD2CB.mozilla.org) (Ping timeout)
# [02:58] * Quits: azakai (alon@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [02:59] * edransch is now known as erick
# [02:59] * Joins: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
# [02:59] <Unfocused> njn: no, you can only operate on the top-most patch
# [03:00] <njn> Unfocused: |hg qref -e| should really be a separate command, IMO
# [03:01] <njn> oh well
# [03:01] * Joins: Boriss_ (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [03:01] <Unfocused> yea, i've thought the same thing :\
# [03:02] * Quits: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Connection reset by peer)
# [03:02] * Joins: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
# [03:02] * Quits: mfinkle (mfinkle@moz-8CB7201C.hsd1.pa.comcast.net) (Quit: ZNC - http://znc.sourceforge.net)
# [03:02] * Quits: Boriss (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [03:02] * Boriss_ is now known as Boriss
# [03:04] * nthomas|away is now known as nthomas
# [03:04] * Joins: ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP)
# [03:04] * ChanServ sets mode: +o ehsan
# [03:05] <@bz> njn: I just edit the diff file
# [03:05] <@bz> njn: in .hg/patches
# [03:05] <@bz> njn: yes, that's kinda bogus
# [03:06] * Quits: yuan (ywang@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: yuan)
# [03:06] * Joins: zzzzz_ (chatzilla@moz-107FCDBA.hfc.comcastbusiness.net)
# [03:07] * Quits: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [03:08] <njn> bz: huh, I don't think it's so bad
# [03:08] * njn goes and tries it
# [03:08] * Quits: myk (Instantbir@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: Instantbird -- http://www.instantbird.com)
# [03:11] <njn> bz: hmm, the updated string didn't show up in |hg out|
# [03:12] <njn> bz: but when I did |hg qref -e|, the updated string *was* there
# [03:12] * fabrice is now known as fabrice|afk
# [03:13] <Unfocused> hg out uses the commit history
# [03:13] <Unfocused> it doesn't care about mq
# [03:13] * Quits: @ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP) (Input/output error)
# [03:14] <Unfocused> so if you manually edit the patch file,you need to qpop and qpush it again, to edit the actual commit
# [03:14] * Quits: bent (chatzilla@moz-C3562645.hsd1.ca.comcast.net) (Ping timeout)
# [03:14] <mattwoodrow> bz: ping
# [03:14] * Quits: juanb (jbecerra@moz-BBE3ABD.mv.mozilla.com) (Quit: juanb)
# [03:14] <njn> Unfocused: huh, thanks
# [03:14] * Joins: dwendorf (Mibbit@moz-6E3B32E8.lightspeed.mdsnwi.sbcglobal.net)
# [03:15] <RyanVM> philor: is there any way to get tbpl to try again if it times out generating a log?
# [03:15] <Bas> taras: I'll send something today.
# [03:15] * Joins: azakai (alon@moz-7B0110AD.mv.mozilla.com)
# [03:15] <taras> Bas: thanks
# [03:16] <taras> Bas: as long as it's before my tuesday morning, i'm happy
# [03:16] <taras> s/happy/really happy/
# [03:18] * Quits: loadbang (loadbang@moz-28BBF200.range86-162.btcentralplus.com) (Quit: loadbang)
# [03:18] * philor is now known as philor|away
# [03:18] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [03:19] <@bz> mattwoodrow: ACK
# [03:19] <@bz> er, ack
# [03:19] <mattwoodrow> bz: You worked on making sure adding table rows/columns didn't invalidate the whole table right?
# [03:19] * Quits: wesj (Instantbir@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Input/output error)
# [03:19] <mattwoodrow> I was wondering if you remembered the bug number for that, and if you had any test cases
# [03:20] <mattwoodrow> want to make sure I don't regress it with DLBI
# [03:20] <@bz> mattwoodrow: sure
# [03:20] <@bz> mattwoodrow: lemme pull it up
# [03:20] <@bz> mattwoodrow: however, that wasn't what I worked on. ;)
# [03:20] * Joins: bent (chatzilla@moz-C3562645.hsd1.ca.comcast.net)
# [03:21] <@bz> mattwoodrow: adding stuff invalidates the whole table
# [03:21] <@bz> mattwoodrow: what I did was make incremental reflow inside a cell not invalidate the whole table
# [03:21] <mattwoodrow> ah
# [03:21] <@bz> mattwoodrow: lemme look up the bu
# [03:21] <@bz> er, bug
# [03:22] * Joins: bjacob (bjacob@moz-ADCA75DC.cpe.net.cable.rogers.com)
# [03:22] * Quits: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com) (Quit: mccr8)
# [03:22] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [03:23] <@bz> So the original bug was https://bugzilla.mozilla.org/show_bug.cgi?id=414298
# [03:24] <@bz> And then some regression fixes in https://bugzilla.mozilla.org/show_bug.cgi?id=416073 and https://bugzilla.mozilla.org/show_bug.cgi?id=416073
# [03:24] <@bz> er, and https://bugzilla.mozilla.org/show_bug.cgi?id=421632
# [03:24] <@bz> And a few others; see the "depends on" list of that first bug
# [03:24] <mattwoodrow> thanks bz, I'll see how badly I regressed them :)
# [03:25] <@bz> https://bugzilla.mozilla.org/show_bug.cgi?id=377419 might have some of the original perf testcases
# [03:25] <@bz> maybe
# [03:25] <@bz> this was one of those "we've seen this in lots of perf bugs, let's just fix it" things
# [03:26] <@bz> one simple thing to test....
# [03:26] <@bz> If you have a table with some cells
# [03:26] * philor|away is now known as philor
# [03:26] <@bz> and one of the cells has a single child which is a fixed-size div
# [03:26] <@bz> and the div has overflow:hidden
# [03:26] <@bz> and you change the innerHTML of that div
# [03:26] <@bz> nothing outside that div should get invalidated
# [03:27] * Quits: cpeterson (cpeterson@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [03:27] * Joins: fs (Elchi3@D6F6ECA1.2E742F60.632B8C24.IP)
# [03:29] * ctalbert is now known as ctalbert|afk
# [03:29] <mattwoodrow> cool, I think I can write a mochitest for that too
# [03:29] <@bz> mattwoodrow: awesome
# [03:29] <@bz> mattwoodrow: tests++
# [03:30] <mattwoodrow> anyway, gotta run for lunch, thanks for your help
# [03:31] * Quits: zzzzz_ (chatzilla@moz-107FCDBA.hfc.comcastbusiness.net) (Quit: ChatZilla 0.9.87 [Firefox 12.0/20120420145725])
# [03:33] * Quits: dseif (dseif@CE526453.531FD64C.630E4E47.IP) (Input/output error)
# [03:34] * Joins: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP)
# [03:35] * Joins: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [03:35] * jmaher|afk is now known as jmaher
# [03:36] <@bz> no problem
# [03:36] * Joins: pranavrc (pranavrc@FEB17D08.6753A649.C28326FD.IP)
# [03:36] * Quits: cadecairos (cadecairos@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [03:37] * mdas is now known as mdas|afk
# [03:37] * Quits: rniwa (rniwa@5CA6DC39.C60FE7DC.4065847B.IP) (Quit: rniwa)
# [03:38] <@bz> hrm
# [03:38] * Joins: bdahl (bdahl@moz-E197F13B.hsd1.ca.comcast.net)
# [03:38] <philor> rnewman: guess what?
# [03:38] <@bz> how come https://crash-stats.mozilla.com/report/index/70268703-b0ea-4e1b-8e6b-710482120430 doesn't have the file+line stuff linkified?
# [03:39] * Quits: kumar (kmcmillan@moz-BBE3ABD.mv.mozilla.com) (Quit: kumar)
# [03:39] <rnewman> philor: yeah, I know, this doesn't make any sense
# [03:40] * Quits: gmoro__ (guilherme@84484EC9.1F1093E8.D159334F.IP) (Ping timeout)
# [03:40] <rnewman> like, not even slightly
# [03:40] <rnewman> do you know offhand what name Talos is using to search for processes?
# [03:40] * Quits: gmoro (guilherme@84484EC9.1F1093E8.D159334F.IP) (Ping timeout)
# [03:40] <rnewman> I would assume "org.mozilla.fennec"
# [03:40] <rnewman> but it doesn't log *anything*
# [03:41] * Quits: dbradley (dbradley@moz-80450F75.fuse.net) (Quit: )
# [03:41] <rnewman> the fun part is: on my machine Fennec doesn't even *launch* the CP process unless you're syncing
# [03:41] * joduinn-mtg is now known as joduinn
# [03:42] * Joins: gmoro__ (guilherme@84484EC9.1F1093E8.D159334F.IP)
# [03:42] * Joins: gmoro (guilherme@84484EC9.1F1093E8.D159334F.IP)
# [03:42] <KWierso> bz: that just shows the filename? my crash reports (on windows, at least) show the full path to each file?
# [03:42] * jaws|away is now known as jaws
# [03:42] <KWierso> s/full/most of/
# [03:42] * Quits: chewey (chewey@moz-407D4743.dip0.t-ipconnect.de) (NickServ (GHOST command used by chewey_))
# [03:43] * Joins: chewey (chewey@moz-BB6B7074.dip0.t-ipconnect.de)
# [03:43] <@bz> KWierso: hmm
# [03:43] <@bz> KWierso: interesting
# [03:44] <@bz> KWierso: I don't know whether to love our tools or hate them. ;)
# [03:44] * Joins: andy9330 (Mibbit@moz-C5B0F8E0.dhcp.bycy.mi.charter.com)
# [03:44] <KWierso> why not both?
# [03:44] <rnewman> jmaher: yt?
# [03:44] * @bz only has room for one emotion in his heart
# [03:44] <KWierso> "rage of a thousand suns", then
# [03:45] <rnewman> I have at least 500 suns' worth to share
# [03:45] * Quits: zuzelvp (zuzelvp@2112147D.C3507A2D.9A8C35B4.IP) (Ping timeout)
# [03:45] * Quits: mdas|afk (mdas@50AE257C.D30B51A1.412CF160.IP) (Quit: mdas|afk)
# [03:45] <jmaher> rnewman: yeah
# [03:45] <rnewman> what inputs and outputs is the talos harness using?
# [03:46] <rnewman> that is, is it searching for "fennec", or "org.mozilla.fennec", or "org.mozilla.fennec_aurora"?
# [03:46] <rnewman> or PID, or ...
# [03:46] <KWierso> bz: if you toggle the "show/hide other threads" link/button thing, you have one linked file about halfway down for Monitor.h:122
# [03:46] <rnewman> it doesn't seem to log anything
# [03:46] * joduinn is now known as joduinn-commute
# [03:46] <jmaher> rnewman: it is searching for org.mozilla.fennec, org.mozilla.fennec_aurora, org.mozilla.firefox_beta or org.mozilla.firefox depending on the branch
# [03:46] <rnewman> OK, so this is confusing
# [03:47] <rnewman> I downloaded the Aurora build that is failing in tbpl
# [03:47] <@bz> KWierso: heh
# [03:47] <jmaher> oh...there is a failure?
# [03:47] <rnewman> and it's definitely using org.m0zill4.fennec_aurora as the process package
# [03:47] * Quits: mats (chatzilla@E8A3702C.D5022173.8500CC29.IP) (Quit: ChatZilla 0.9.88.2 [Firefox 11.0/20120312181643])
# [03:47] <rnewman> so it shouldn't match
# [03:47] <rnewman> https://tbpl.mozilla.org/php/getParsedLog.php?id=11343227&tree=Mozilla-Aurora&full=1
# [03:47] <rnewman> but I can't see which process talos thinks is still running
# [03:47] <rnewman> search for FAIL: browser failed to close after being initialized
# [03:48] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [03:48] <jmaher> ok, I see that; talos starts up an instance to get information an initialize the profile
# [03:49] * Joins: lduros (lduros@moz-BED1C6A5.c3-0.rdl-ubr1.trpr-rdl.pa.cable.rcn.com)
# [03:49] <jmaher> classic, the only day in months that I don't have my tegra or another device available to test on
# [03:49] <rnewman> NOISE: browser_name:undefined ...?
# [03:49] <jmaher> rnewman: that is missing a value of 'firefox' or 'fennec'
# [03:49] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [03:49] <rnewman> you mean "it's weird that it's neither of those in this log"?
# [03:50] <jmaher> no, it is expected
# [03:50] <rnewman> ok
# [03:50] <rnewman> flailing for possibilities here
# [03:50] * Quits: harth (harth@moz-C2C235AC.bb.sky.com) (Input/output error)
# [03:51] <rnewman> I have two: (a) the CP process is running, and talos is only looking for "fennec", so my mangling isn't helping; (b) the CP process is not running, and I'm somehow provoking this error through a completely unrelated change
# [03:51] * Quits: azakai (alon@moz-7B0110AD.mv.mozilla.com) (Ping timeout)
# [03:51] <jmaher> so for some reason we received all the output from the first launch of the browser via getInfo.html, but we never saw the browser close
# [03:51] * Quits: gmoro__ (guilherme@84484EC9.1F1093E8.D159334F.IP) (Ping timeout)
# [03:51] * Quits: gmoro (guilherme@84484EC9.1F1093E8.D159334F.IP) (Ping timeout)
# [03:52] <rnewman> the change in question renames one of the android:process names in fennec's manifest from org.mozilla.gecko.* to org.m0zill4.fennec_aurora.*
# [03:52] <rnewman> that's literally the only thing changed
# [03:52] <rnewman> so the hypothesis was that talos was still seeing that process alive, because its name now collides with the package name/fennec
# [03:52] * Joins: jimb (user@moz-F4EC06CC.hsd1.or.comcast.net)
# [03:52] <rnewman> is there anything else that could be going on?
# [03:53] <jmaher> yeah, it could be
# [03:54] * rnewman preps a patch which does org.mozilla.f3nnec
# [03:55] * Quits: bjacob (bjacob@moz-ADCA75DC.cpe.net.cable.rogers.com) (Ping timeout)
# [03:55] <rnewman> the weird thing is, this process shouldn't even be running
# [03:55] <KWierso> rnewman: april 1 was a month ago, isn't it a bit late to be introducing a prank that turns all browser text into l33tsp34k?
# [03:55] <rnewman> :D
# [03:55] <jmaher> KWierso: this is android:)
# [03:55] <rnewman> poor blassey had to r+ it :D
# [03:55] * Quits: sewardj (sewardj@moz-40DEA2D4.dip.t-dialin.net) (Ping timeout)
# [03:56] <rnewman> wouldn't it be nice if try builds got preferential treatment if their try syntax was really minimal?
# [03:56] * Quits: tH (Rob@cpc16-seac19-2-0-cust234.7-2.cable.virginmedia.com) (Quit: ChatZilla 0.9.88-rdmsoft [XULRunner 1.9.0.1/2008072406])
# [03:56] * Joins: sewardj (sewardj@moz-BAB78205.dip.t-dialin.net)
# [03:56] * Quits: @dveditz (dveditz@moz-BBE3ABD.mv.mozilla.com) (Quit: dveditz)
# [03:56] <rnewman> "just talos tests on android-opt, plz"
# [03:56] <jmaher> rnewman++
# [03:57] * Parts: lduros (lduros@moz-BED1C6A5.c3-0.rdl-ubr1.trpr-rdl.pa.cable.rcn.com)
# [03:57] * Quits: Boriss (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: Boriss)
# [03:58] <darktrojan> alternatively, "going to bed now, do it when you get around to it"
# [03:58] <darktrojan> "but before next week preferably"
# [03:58] <rnewman> heh
# [04:01] * Quits: gozala (gozala@moz-D5BED6F9.dsl.dynamic.sonic.net) (Quit: Leaving.)
# [04:01] <Mossop> rnewman: Well the more minimal people's syntax is the faster everyone will get builds, no need to build something special for that
# [04:01] <rnewman> but... the incentive!
# [04:02] <Mossop> Yeah, the thing we need is an incentive to not doing full try tests before pushing to inbound :p
# [04:02] <rnewman> :D
# [04:02] <@bz> so if I asked for ux-review
# [04:02] <@bz> and crickets
# [04:02] <@bz> do I just land?
# [04:02] <rnewman> Mossop: you make a compelling point.
# [04:02] <rnewman> bz: LAND IT
# [04:02] <rnewman> the crickets liked it
# [04:02] * @bz will do just that
# [04:02] * Joins: larfdesk (Adam_Hinke@moz-F92153ED.longlines.com)
# [04:02] <Mossop> Bug the ux team directly?
# [04:03] <Mossop> Don't assume they read bugmail
# [04:03] * @bz shrugs
# [04:03] * Quits: ekr (ekr@moz-D7997EC8.rtfm.com) (Ping timeout)
# [04:03] <@bz> it's just not that important in this case
# [04:03] <@bz> it's not like they reviewed the bug that caused the breakage
# [04:03] <@bz> and as far as bugmail
# [04:04] <@bz> what's the point of having a ux-review@mozilla.com bugzilla account if it won't be read?
# [04:05] * Quits: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Ping timeout)
# [04:05] * Joins: mdas (mdas@50AE257C.D30B51A1.412CF160.IP)
# [04:05] * Quits: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP) (Ping timeout)
# [04:06] <philor> rnewman: ten hours later: in your log, python remotePerfConfigurator.py ... --sampleConfig remote.config, in http://mxr.mozilla.org/build/source/talos/talos/remote.config#16 process: fennec, some churning, browser_config['process'] == "fennec"
# [04:06] * Joins: dbradley (dbradley@moz-80450F75.fuse.net)
# [04:07] * philor closes 20 tabs
# [04:08] <rnewman> I love you, philor
# [04:08] <rnewman> with the flame of a trillion suns
# [04:08] <rnewman> you are my eyes when I cannot see
# [04:08] <rnewman> my ears when I cannot hear
# [04:09] <pcwalton> Mossop: just threw a FUEL patch into your review queue. feel free to kick it off to someone else, I have no idea who maintains FUEL
# [04:09] <RyanVM> njn: ping
# [04:09] <njn> RyanVM: pong
# [04:09] <rnewman> that means this try push should be happy, then
# [04:09] <bent> pcwalton, which bug ooc?
# [04:09] <pcwalton> bent: bug 750583
# [04:09] <RyanVM> njn: nevermind, red looks to be infra-related
# [04:09] <pcwalton> FUEL can cause multi-minute shutdown times
# [04:09] * Quits: billm (billm@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving)
# [04:09] * Quits: philor (philor@moz-638273A8.my-nick.name) (Quit: ZNC - http://znc.sourceforge.net)
# [04:09] <pcwalton> not sure which addon is using it, probably readability or ABP
# [04:09] <njn> RyanVM: \o/
# [04:10] <bent> pcwalton, weiiirrrrddd
# [04:10] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [04:10] <bent> pcwalton, i just filed 750454 today too
# [04:10] * Joins: philor (philor@moz-638273A8.my-nick.name)
# [04:11] <pcwalton> ahaha
# [04:11] <Mossop> pcwalton: You do now!
# [04:11] <pcwalton> today is FUEL day
# [04:11] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [04:11] * bc_ is now known as bc
# [04:12] <bent> pcwalton, we decided to try removing the shutdown mechanism entirely
# [04:12] <pcwalton> bent: I'll attach my patch to the bug in case nuking gShutdown turns the tree orange or whatever
# [04:12] <pcwalton> feel free to not use it though, nuking gShutdown is better
# [04:12] <bent> the entire thing is crazy
# [04:12] * Quits: clokep (Instantbir@moz-69FB3955.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com) (Quit: Instantbird -- http://www.instantbird.com)
# [04:14] * Quits: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP) (Quit: Computer has gone to sleep.)
# [04:14] * Quits: njn (chatzilla@moz-71689D29.dyn.iinet.net.au) (Ping timeout)
# [04:14] * Quits: brendan (brendaneic@moz-BBE3ABD.mv.mozilla.com) (Quit: brendan)
# [04:14] * Quits: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [04:15] * Joins: njn (chatzilla@moz-3AD84CB9.dyn.iinet.net.au)
# [04:15] * Quits: dria (dria@moz-2830A996.dhcp-dynamic.fibreop.nb.bellaliant.net) (Quit: dria)
# [04:15] <Mossop> So do I need to review that bug or not?
# [04:16] * Joins: nhirata (nhirata.bu@moz-585D5EFC.hsd1.ca.comcast.net)
# [04:16] <rnewman> is it safe to cancel try builds with the Red X?
# [04:17] * Joins: akeybl (akeybl@moz-52D39FF6.hsd1.ca.comcast.net)
# [04:17] * Quits: tonymec (tonymec@78D6A323.3F6FF523.277517C1.IP) (Input/output error)
# [04:17] <rnewman> I seem to recall consternation last time I tried
# [04:17] <bent> Mossop, i recommend duping
# [04:18] * Joins: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [04:18] * Quits: hub (hub@D459F0F4.D9251953.6E712CE2.IP) (Ping timeout)
# [04:18] <bent> pcwalton is going to move patch over
# [04:18] <pcwalton> yeah, just did that
# [04:18] <heycam> rnewman, I think it's ok, it's just cancelling builds on m-c and other trees that screws things up
# [04:18] <ddahl> hey - does anyone know where the WebRTC code is being checked in?
# [04:18] * Quits: njn (chatzilla@moz-3AD84CB9.dyn.iinet.net.au) (Ping timeout)
# [04:18] * Quits: tonymec|away (tonymec@78D6A323.3F6FF523.277517C1.IP) (Quit: Konversation terminated!)
# [04:19] <rnewman> kk
# [04:19] <rnewman> ta, heycam
# [04:19] * Joins: njn (chatzilla@moz-C2274D74.dyn.iinet.net.au)
# [04:19] <rnewman> ddahl: https://github.com/mozilla/webrtc ?
# [04:19] <ddahl> rnewman: really?
# [04:19] * Quits: andy9330 (Mibbit@moz-C5B0F8E0.dhcp.bycy.mi.charter.com) (Quit: http://www.mibbit.com ajax IRC Client)
# [04:19] <rnewman> last commit 11 hours ago
# [04:20] <ddahl> haha
# [04:20] <rnewman> most non-mc work takes place on github these days :)
# [04:20] <rnewman> wouldn't be surprised if that were the place
# [04:21] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [04:21] * Joins: azakai (alon@moz-BBE3ABD.mv.mozilla.com)
# [04:22] * Quits: KLB (Kenneth_Ba@moz-C233EF2B.maine.res.rr.com) (Quit: ChatZilla 0.9.88.2 [Firefox 12.0/20120420145725])
# [04:22] * Joins: variable (toor@495486BA.98B6D721.292D5CE7.IP)
# [04:22] * Joins: tonymec__ (tonymec@78D6A323.3F6FF523.277517C1.IP)
# [04:22] * Quits: jgriffin (jgriffin@moz-4FBFA41D.hsd1.wa.comcast.net) (Quit: jgriffin)
# [04:22] * tonymec__ is now known as tonymec|away
# [04:22] * Joins: m_kato (m_kato@moz-348F61F0.mozilla.or.jp)
# [04:23] <@roc> it is not
# [04:23] <@roc> you probably want the alder branch
# [04:24] * Quits: bonnie (bbsurender@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [04:25] * Quits: pcwalton (pcwalton@moz-7B0110AD.mv.mozilla.com) (Quit: nbvcx)
# [04:26] * khuey|away is now known as khuey
# [04:26] * heycam is now known as heycam|away
# [04:28] * Quits: Mossop (mossop@moz-347C3D31.dsl.dynamic.sonic.net) (Ping timeout)
# [04:28] * Joins: Mardak (Mardak@moz-9509DFBB.hsd1.ca.comcast.net)
# [04:29] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [04:29] <ddahl> roc! thanks!
# [04:29] <ddahl> I knew we had a build somewhere
# [04:30] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [04:30] * Quits: azakai (alon@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [04:30] * Quits: bb10 (bb10@moz-C7B05616.org) (Ping timeout)
# [04:31] * Quits: victorporof (victorporo@5F6B43D3.1C233438.79933D60.IP) (Connection reset by peer)
# [04:32] * Joins: bb10 (bb10@moz-C7B05616.org)
# [04:35] * Joins: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP)
# [04:37] * Quits: jet (junglecode@moz-79F891EE.hsd1.ca.comcast.net) (Quit: jet)
# [04:37] * Joins: tonymec (tonymec@78D6A323.3F6FF523.277517C1.IP)
# [04:38] * Quits: priya (Adium@moz-5843392D.hsd1.ca.comcast.net) (Quit: Leaving.)
# [04:39] * Quits: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com) (Ping timeout)
# [04:39] * Joins: mccr8 (mccr8@moz-5F92CD4B.dsl.dynamic.sonic.net)
# [04:40] * Quits: bb10 (bb10@moz-C7B05616.org) (Ping timeout)
# [04:40] * Quits: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [04:40] * Quits: faramarz (faramarz@moz-BBE3ABD.mv.mozilla.com) (Quit: faramarz)
# [04:40] <RyanVM> philor: what failed here? https://tbpl.mozilla.org/php/getParsedLog.php?id=11345326&tree=Mozilla-Inbound
# [04:40] * Quits: Mardak (Mardak@moz-9509DFBB.hsd1.ca.comcast.net) (Quit: Mardak)
# [04:41] * Joins: bb10 (bb10@moz-C7B05616.org)
# [04:41] <philor> RyanVM: ctrl+f, "results: 2"
# [04:41] <RyanVM> found it
# [04:42] * joduinn-commute is now known as joduinn-home
# [04:42] <philor> not quite all-powerful, since some steps that say it don't mean it because you can have a step ignore that it failed, but mostly results: n will find the badness
# [04:45] * Joins: anky (anky@C772EAAE.744E7036.A3D1B221.IP)
# [04:45] * Quits: akeybl (akeybl@moz-52D39FF6.hsd1.ca.comcast.net) (Quit: Linkinus - http://linkinus.com)
# [04:49] * Joins: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [04:49] * Joins: surkov (surkov@C1A6A50A.88321457.33A1AC3C.IP)
# [04:53] * joduinn-home is now known as joduinn-afk
# [04:54] * heycam|away is now known as heycam
# [04:55] * Quits: mccr8 (mccr8@moz-5F92CD4B.dsl.dynamic.sonic.net) (Quit: mccr8)
# [04:55] * Quits: lmandel (lmandel@FE1F74.86ED00A7.971E19F6.IP) (Quit: lmandel)
# [04:56] * Joins: brendan (brendaneic@moz-A286C218.hsd1.ca.comcast.net)
# [04:56] * Quits: bbondy (bbondy@moz-C9962B2.home.cgocable.net) (Ping timeout)
# [04:57] * Joins: Boriss (FlyingToas@4AB84D20.81CFD33B.11DF82C5.IP)
# [04:58] * Joins: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com)
# [04:59] * Quits: fs (Elchi3@D6F6ECA1.2E742F60.632B8C24.IP) (Quit: going for another walk to the top of Twin Peaks to see SF at night)
# [04:59] <rnewman> hmm, my try build seems to have stalled
# [04:59] <rnewman> come on, sunshine
# [05:00] * Quits: rjohnson19 (chatzilla@moz-9148485F.hsd1.ma.comcast.net) (Quit: ChatZilla 0.9.88-rdmsoft [XULRunner 1.9.0.17/2009122204])
# [05:02] * philor is now known as philor|away
# [05:03] * Quits: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: anant)
# [05:04] * Quits: robarnold (rob@moz-59F6B5.hfc.comcastbusiness.net) (Ping timeout)
# [05:04] * Joins: robarnold (rob@moz-59F6B5.hfc.comcastbusiness.net)
# [05:04] * Joins: ctopper (craig@C3495DA.BA3DBA56.AE2B2F80.IP)
# [05:06] * Joins: KaIRC (robert@moz-88135F74.adsl.highway.telekom.at)
# [05:06] * Quits: KaiRo_away (robert@moz-4059AC21.adsl.highway.telekom.at) (Ping timeout)
# [05:08] * Quits: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: Reconnecting…)
# [05:08] * Joins: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [05:08] * Quits: RyanVM (chatzilla@moz-D04D3C77.phlapa.fios.verizon.net) (Quit: ChatZilla 0.9.88.2 [Firefox 15.0a1/20120429155331])
# [05:11] * Quits: micahg (micahg@moz-6E8FD6B4.c3-0.arm-ubr1.chi-arm.il.cable.rcn.com) (Ping timeout)
# [05:11] * philor|away is now known as philor
# [05:12] * Quits: Jesse (jruderman@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [05:12] * philor is now known as philor|away
# [05:12] * Joins: mconley (mconley@moz-D640D16C.cable.teksavvy.com)
# [05:13] * Joins: dveditz (dveditz@moz-5051E786.dhcp.cruzio.com)
# [05:13] * ChanServ sets mode: +o dveditz
# [05:16] * Joins: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
# [05:16] * Joins: knelson (Adium@moz-9C879B7.dsl.pltn13.sbcglobal.net)
# [05:17] * Parts: knelson (Adium@moz-9C879B7.dsl.pltn13.sbcglobal.net)
# [05:17] * Joins: anant (Anant@moz-D3725328.dsl.dynamic.sonic.net)
# [05:19] * Joins: micahg (micahg@moz-6E8FD6B4.c3-0.arm-ubr1.chi-arm.il.cable.rcn.com)
# [05:20] * Quits: damons (gnubeard@moz-BBE3ABD.mv.mozilla.com) (Quit: damons)
# [05:20] * philor|away is now known as philor
# [05:20] * cjones is now known as cjones-dinner
# [05:28] * Quits: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
# [05:29] * Joins: twi (Adium@moz-5390D98D.cust.dsl.vodafone.it)
# [05:29] * Quits: mdas (mdas@50AE257C.D30B51A1.412CF160.IP) (Quit: mdas)
# [05:29] * Joins: anant_ (Anant@moz-488092DD.hsd1.ca.comcast.net)
# [05:30] * Quits: anant (Anant@moz-D3725328.dsl.dynamic.sonic.net) (Ping timeout)
# [05:32] * Joins: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [05:33] * Joins: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP)
# [05:34] * Joins: sfink (chatzilla@moz-7B7651CB.dsl.pltn13.sbcglobal.net)
# [05:34] * Joins: gwagner_ (idefix2@moz-B8B530C2.hsd1.ca.comcast.net)
# [05:36] * Joins: kvda (kvda@AFDD388F.C3893E67.923345AB.IP)
# [05:36] * Joins: priya (Adium@moz-5843392D.hsd1.ca.comcast.net)
# [05:38] <ddahl> roc: I am curious where the SRTP Encryption objects are implemented for the WebRTC code. Do you know what google is using?
# [05:39] <jesup> ddahl: libsrtp
# [05:39] <ddahl> jesup: thanks
# [05:39] * Joins: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP)
# [05:39] <kvda> Hi I've seem to run into a problem with the latest Aurora release
# [05:40] <jesup> ddahl: media/webrtc/trunk/third_party/libsrtp/* (FYI, I'm also a libsrtp maintainer, not that there's much traffic)
# [05:40] <jesup> ddahl: (on alder)
# [05:40] <ddahl> jesup: cool, thank you. I was curious how all of that is put together
# [05:41] <jesup> THe only issue of contention (lots of it) is keying
# [05:41] * Joins: cviecco_ (cviecco@moz-2D8CF2BF.hsd1.ca.comcast.net)
# [05:41] <ddahl> jesup: as in creating a master key to derive further keys from?
# [05:42] <jesup> DTLS-SRTP looks to be mandatory-to-implement. plain RTP is disallowed (though for testing with a cmdline param, a NULL SRTP cipher is allowed). SDES is the argument point
# [05:42] * Joins: fabrice (fabrice@moz-94F028C6.hsd1.ca.comcast.net)
# [05:42] <ddahl> ok, more stuff to read about
# [05:45] * Quits: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP) (Quit: Computer has gone to sleep.)
# [05:47] * Joins: mfinkle (mfinkle@moz-8CB7201C.hsd1.pa.comcast.net)
# [05:48] * Quits: Boriss (FlyingToas@4AB84D20.81CFD33B.11DF82C5.IP) (Ping timeout)
# [05:49] * Joins: mike5w3c (MikeS@moz-2E518994.hfc.comcastbusiness.net)
# [05:49] * Quits: gwagner_ (idefix2@moz-B8B530C2.hsd1.ca.comcast.net) (Quit: gwagner_)
# [05:51] * Joins: Boriss (FlyingToas@6748FAF5.A8E6DA2D.217C7FCC.IP)
# [05:51] * Quits: bent (chatzilla@moz-C3562645.hsd1.ca.comcast.net) (Quit: ChatZilla 0.9.88-rdmsoft [XULRunner 1.9.2.19/20110707195905])
# [05:51] * Joins: aja (chatzilla@3846FB25.AEC64F0A.7880DB15.IP)
# [05:51] * khuey is now known as khuey|away
# [05:51] * Quits: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP) (Ping timeout)
# [05:55] * Joins: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP)
# [05:55] * Joins: krit (krit@moz-3AE657AD.hsd1.ca.comcast.net)
# [05:56] * Joins: Mossop (mossop@moz-347C3D31.dsl.dynamic.sonic.net)
# [05:57] * Quits: espindola (espindola@moz-97E8149E.dsl.teksavvy.com) (Quit: Ex-Chat)
# [05:58] * Joins: faramarz (faramarz@moz-CD61C00E.hsd1.ca.comcast.net)
# [05:59] * Joins: gwagner_ (idefix2@moz-B8B530C2.hsd1.ca.comcast.net)
# [06:00] * Quits: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP) (Ping timeout)
# [06:02] * Quits: Havvy (Mibbit@moz-B7B34335.ptld.qwest.net) (Quit: Restarting computer.)
# [06:03] * Joins: damons (gnubeard@moz-A41E6911.hsd1.ca.comcast.net)
# [06:03] * Joins: kumar (kmcmillan@moz-4BE034AB.ptr.us.xo.net)
# [06:03] * Joins: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP)
# [06:05] * Joins: dseif (dseif@moz-B6BACC41.dsl.teksavvy.com)
# [06:06] * Quits: bdahl (bdahl@moz-E197F13B.hsd1.ca.comcast.net) (Client exited)
# [06:07] * Quits: allisterb (allisterb@2CE5E668.5725EB66.F43E9E20.IP) (Ping timeout)
# [06:09] * Joins: ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP)
# [06:09] * ChanServ sets mode: +o ehsan
# [06:09] <rnewman> yaaaay talos is working again
# [06:09] * Quits: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Ping timeout)
# [06:10] * Quits: KaIRC (robert@moz-88135F74.adsl.highway.telekom.at) (Input/output error)
# [06:11] * Joins: mjschranz (mjschranz@moz-6FE6B833.cpe.net.cable.rogers.com)
# [06:11] * Joins: jet (junglecode@moz-79F891EE.hsd1.ca.comcast.net)
# [06:11] * Quits: @ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP) (Connection reset by peer)
# [06:12] * Joins: ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP)
# [06:12] * ChanServ sets mode: +o ehsan
# [06:12] <philor> itym t4l0s
# [06:12] <rnewman> lol
# [06:12] <rnewman> srsly
# [06:12] * rnewman lands on inbound, having debugged on auror
# [06:12] <rnewman> DOING IT RONG
# [06:13] * Joins: dseif_ (dseif@moz-B6BACC41.dsl.teksavvy.com)
# [06:13] * Quits: dseif (dseif@moz-B6BACC41.dsl.teksavvy.com) (Connection reset by peer)
# [06:14] * Joins: Havvy (Mibbit@moz-B7B34335.ptld.qwest.net)
# [06:14] <philor> "fatal error C1002: compiler is out of heap space in pass 2"
# [06:15] <philor> hello, FUCKED, OH SO TOTALLY FUCKED
# [06:15] <nigelb> heh
# [06:15] * Quits: nrc (nrc@538BABFE.A073F3E.97BBD552.IP) (Ping timeout)
# [06:15] <philor> khuey|away, save us!
# [06:16] <philor> rnewman: if you're going to land on inbound, you better do it quick, because it's about to close over that
# [06:16] <nthomas> esr10 ?
# [06:16] <philor> inbound
# [06:16] <rnewman> philor: done, thanks philor
# [06:16] * Quits: anky (anky@C772EAAE.744E7036.A3D1B221.IP) (Client exited)
# [06:17] <nthomas> huh, in parser/html
# [06:17] * cjones-dinner is now known as cjones
# [06:17] <nthomas> oh it is in xul, like I thought it would be
# [06:17] * Quits: mjschranz (mjschranz@moz-6FE6B833.cpe.net.cable.rogers.com) (Client exited)
# [06:18] * Joins: sworkman_ (sworkman@moz-BBE3ABD.mv.mozilla.com)
# [06:18] <philor> can you retrigger just that build, or should I trigger PGO on every platform to get just another Win?
# [06:18] * Quits: sworkman_ (sworkman@moz-BBE3ABD.mv.mozilla.com) (Quit: sworkman_)
# [06:19] * larfdesk will wait for a post to planet.mozilla to feel educated and informed :D
# [06:19] <nthomas> I'll do just that one
# [06:19] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [06:19] <philor> thx
# [06:20] * Quits: Hendikins (wolfox@moz-16899DFF.hhui4.ken.bigpond.net.au) (Ping timeout)
# [06:20] * Joins: birtles (chatzilla@moz-348F61F0.mozilla.or.jp)
# [06:20] <nthomas> got the same slave, dunno if that's good or not
# [06:21] * Joins: Hendikins (wolfox@moz-16899DFF.hhui4.ken.bigpond.net.au)
# [06:21] <philor> I'll trigger a full round on the tip cset, that'll give us lots of confusing variability :)
# [06:23] * Quits: mconley (mconley@moz-D640D16C.cable.teksavvy.com) (Input/output error)
# [06:23] <philor> oh, 9 o'clock, there's already one running there
# [06:23] <philor> or 4 back, or somewhere
# [06:24] * Quits: Boriss (FlyingToas@6748FAF5.A8E6DA2D.217C7FCC.IP) (Quit: Boriss)
# [06:25] <philor> and hmm, the last green says "linker max virtual size: 3021185024"
# [06:26] * Joins: nrc (nrc@538BABFE.A073F3E.97BBD552.IP)
# [06:26] <nthomas> did we get that onto graphs ?
# [06:26] <philor> so maybe just like the single esr10 one, it just likes to watch me freak out occasionally
# [06:27] <philor> don't think so, I think we expect that edmorley will keep looking at it every so often
# [06:27] * Joins: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se)
# [06:28] * Quits: @ehsan (ehsan@BEDFFF26.EE049E25.8B035CD7.IP) (Input/output error)
# [06:31] * Quits: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se) (Ping timeout)
# [06:31] * Quits: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP) (Ping timeout)
# [06:33] * Quits: kvda (kvda@AFDD388F.C3893E67.923345AB.IP) (Quit: ( x____x))
# [06:34] * Quits: mwu (mwu@moz-59435430.hsd1.nj.comcast.net) (Quit: This computer has gone to sleep)
# [06:35] <philor> nice that we have two "do 32-bit Win builds on 64-bit Win" bugs, one a P5 enh and one a P3 nor
# [06:36] <Havvy> philor: Time to close one?
# [06:37] * Quits: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Input/output error)
# [06:38] * Joins: ehsan_ (ehsan@BEDFFF26.EE049E25.8B035CD7.IP)
# [06:39] * Joins: mdas (mdas@50AE257C.D30B51A1.412CF160.IP)
# [06:40] * Joins: luke_laptop (andhow@moz-205A2353.hsd1.ca.comcast.net)
# [06:41] * nthomas is now known as nthomas|away
# [06:41] * Joins: bdahl (bdahl@moz-E197F13B.hsd1.ca.comcast.net)
# [06:42] <luke_laptop> does anyone know how to mass-trigger a bunch of talos runs? e.g., in https://tbpl.mozilla.org/?tree=Try&rev=20892516ccc7, did bholley have to click + 1000 times?
# [06:44] * BenWa is now known as BenWa|sms
# [06:45] * philor doesn't see quite 1000 runs there
# [06:46] * Quits: faramarz (faramarz@moz-CD61C00E.hsd1.ca.comcast.net) (Quit: faramarz)
# [06:46] <luke_laptop> philor: well.. like 50
# [06:48] <luke_laptop> philor are there any hotkeys for moving between green jobs?
# [06:48] <philor> alas, no
# [06:48] * Quits: automata (automata@8D23278A.C27CA109.16867D26.IP) (Quit: Saindo)
# [06:48] * Joins: smontagu (chatzilla@moz-A6D65AB5.red.bezeqint.net)
# [06:49] * Joins: faramarz (faramarz@moz-CD61C00E.hsd1.ca.comcast.net)
# [06:50] * Joins: krit1 (krit@moz-1FC1932F.adobe.com)
# [06:50] <philor> great, there's Linux64 PGO broken
# [06:50] * Quits: faramarz (faramarz@moz-CD61C00E.hsd1.ca.comcast.net) (Quit: faramarz)
# [06:50] * Quits: krit (krit@moz-3AE657AD.hsd1.ca.comcast.net) (Ping timeout)
# [06:52] * Quits: larfdesk (Adam_Hinke@moz-F92153ED.longlines.com) (Quit: Leaving.)
# [06:53] * njn regrets being so leisurely in his patch landing schedule today
# [06:53] * Joins: ekr (ekr@moz-D7997EC8.rtfm.com)
# [06:54] * Quits: pranavrc (pranavrc@FEB17D08.6753A649.C28326FD.IP) (Quit: Ping timeout: ∞)
# [06:56] * Quits: kumar (kmcmillan@moz-4BE034AB.ptr.us.xo.net) (Quit: kumar)
# [06:57] * jhopkins is now known as jhopkins|afk
# [06:57] * Quits: mdas (mdas@50AE257C.D30B51A1.412CF160.IP) (Quit: mdas)
# [06:57] * Quits: cviecco_ (cviecco@moz-2D8CF2BF.hsd1.ca.comcast.net) (Input/output error)
# [06:57] * Joins: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se)
# [06:58] * Quits: dseif_ (dseif@moz-B6BACC41.dsl.teksavvy.com) (Input/output error)
# [07:02] * Quits: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se) (Ping timeout)
# [07:02] * Joins: hub (hub@moz-E2FCA694.figuiere.net)
# [07:04] <sfink> luke_laptop: theoretically, you ought to be able to use the self-serve API, though I don't know much about it
# [07:04] * ctalbert|afk is now known as ctalbert
# [07:04] <sfink> eg: curl -u sfink@mozilla.com -k -d request_id=10948671 https://build.mozilla.org/buildapi/self-serve/try/request
# [07:04] <luke_laptop> sfink: ah, good point. for now i'm just doing the dumb thing
# [07:05] <sfink> (I found the request ID by navigating https://build.mozilla.org/buildapi/self-serve/try/rev/XXXXX and looking for an appropriate one to rebuild)
# [07:05] <sfink> I don't know if it suppresses repeated rebuilds of the same job or whatever
# [07:06] * Quits: ekr (ekr@moz-D7997EC8.rtfm.com) (Ping timeout)
# [07:07] * ewong|sleep is now known as ewong
# [07:08] * Joins: vikram360 (vikram360@195736B2.76910E54.2A068A5E.IP)
# [07:14] * Quits: jgilbert (jgilbert@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [07:15] * Joins: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP)
# [07:15] * Quits: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com) (Input/output error)
# [07:22] * Joins: sicking (chatzilla@moz-7F871C5C.hsd1.ca.comcast.net)
# [07:22] * Quits: ehsan_ (ehsan@BEDFFF26.EE049E25.8B035CD7.IP) (Input/output error)
# [07:24] * Quits: terrence (terrence@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [07:25] * Joins: Mardak (Mardak@moz-9509DFBB.hsd1.ca.comcast.net)
# [07:25] * Quits: m_kato (m_kato@moz-348F61F0.mozilla.or.jp) (Quit: Leaving...)
# [07:25] * Quits: Mardak (Mardak@moz-9509DFBB.hsd1.ca.comcast.net) (Quit: Mardak)
# [07:26] * Joins: ekr (ekr@moz-D7997EC8.rtfm.com)
# [07:30] * Quits: ekr (ekr@moz-D7997EC8.rtfm.com) (Ping timeout)
# [07:30] * Joins: Boriss (FlyingToas@moz-62AAA429.hsd1.ca.comcast.net)
# [07:30] <smontagu> sigh
# [07:30] <smontagu> why does ubuntu ui get worse with every release
# [07:31] * Quits: krit1 (krit@moz-1FC1932F.adobe.com) (Connection reset by peer)
# [07:31] <smontagu> now they totally screwed up alt-tab
# [07:31] <darktrojan> worse than it was already?
# [07:31] <smontagu> you used to be able to see all the possibilities on screen at once
# [07:32] <smontagu> now you only see three
# [07:32] <darktrojan> wtf
# [07:32] <smontagu> they also put my printer out of action by removing csh
# [07:32] * Joins: Xaquseg (xaquseg@moz-D4633F43.pixelhunger.com)
# [07:32] <smontagu> that was fun working out
# [07:33] * joduinn-afk is now known as joduinn-home
# [07:34] <darktrojan> are they even more comically oversized to fill up the extra space created by not dsplaying everything?
# [07:35] * Joins: ekr (ekr@moz-D7997EC8.rtfm.com)
# [07:35] * Quits: Mossop (mossop@moz-347C3D31.dsl.dynamic.sonic.net) (Ping timeout)
# [07:35] <smontagu> i was assuming that cause and effect were the other way round, but yes
# [07:35] * Joins: kvda (kvda@AFDD388F.C3893E67.923345AB.IP)
# [07:35] <darktrojan> :(
# [07:35] * darktrojan doesn't upgrade
# [07:36] * Joins: WeirdAl (chatzilla@moz-5BDC219.hsd1.ca.comcast.net)
# [07:37] <darktrojan> would it be rude to create ubuntumemes.tumblr.com?
# [07:37] <Havvy> darktrojan: Only if you are unaffiliated with Ubuntu.
# [07:37] <darktrojan> yes then
# [07:37] <Havvy> File a bug against Ubuntu about it instead.
# [07:38] <darktrojan> heh
# [07:38] <smontagu> "cover switch" instead of "layout based switcher" in the settings is a bit more usable
# [07:38] <Havvy> Also, somebody should create an aggregator http://programersmem.es/
# [07:38] <darktrojan> RESO WONT, we don't listen to users
# [07:38] <Havvy> Without the middle 's'
# [07:39] <smontagu> and with a double m in programmers :-P
# [07:39] * Joins: Faramarz (Faramarz@moz-CD61C00E.hsd1.ca.comcast.net)
# [07:39] * smontagu assumes that that applies in en-US, even if program has a single m
# [07:39] <WeirdAl> "So what if I hack in a batcave?"
# [07:40] <Havvy> smontagu: Aye, it does.
# [07:41] <Havvy> A programer makes programes.
# [07:41] * Quits: ddahl (ddahl@moz-976797D6.hsd1.il.comcast.net) (Quit: Adios Amigos)
# [07:41] <glandium> philor: it's because of bug 674225
# [07:42] * Quits: ekr (ekr@moz-D7997EC8.rtfm.com) (Ping timeout)
# [07:42] * Joins: bonnie (bbsurender@moz-9AED106A.dsl.pltn13.sbcglobal.net)
# [07:43] * Quits: Boriss (FlyingToas@moz-62AAA429.hsd1.ca.comcast.net) (Quit: Boriss)
# [07:43] <philor> glandium: sweet, I was trying to figure out whether that was going into libxul instead of whatever the separate one is, and then I got confused and decided to just use retriggers, and then I got stuck since triggering PGO is busted
# [07:43] * philor waves goodbye to opus
# [07:43] * Quits: luke_laptop (andhow@moz-205A2353.hsd1.ca.comcast.net) (Quit: luke_laptop)
# [07:43] <glandium> philor: or maybe not
# [07:43] <philor> or, yeah :(
# [07:44] <glandium> that dep was added in https://hg.mozilla.org/integration/mozilla-inbound/rev/f8c7652a33bf
# [07:44] <glandium> that's 5 months ago
# [07:46] * Quits: cpearce (chatzilla@538BABFE.A073F3E.97BBD552.IP) (Ping timeout)
# [07:47] * Quits: karl (karl@538BABFE.A073F3E.97BBD552.IP) (Ping timeout)
# [07:47] * Joins: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net)
# [07:47] <philor> glandium: which are we talking about, the Linux64 or the Windows?
# [07:47] * Quits: bonnie (bbsurender@moz-9AED106A.dsl.pltn13.sbcglobal.net) (Quit: Ex-Chat)
# [07:47] <glandium> philor: linux64
# [07:47] <WeirdAl> ? Opus going bye-bye?
# [07:48] <glandium> philor: opus is in a separate lib on windows
# [07:49] <philor> bummer, since that was the only thing that looked at all good for anything other than a sudden need to switch to 64-bit Windows builders
# [07:49] * Quits: damons (gnubeard@moz-A41E6911.hsd1.ca.comcast.net) (Quit: damons)
# [07:49] * Joins: sworkman (sworkman@moz-825EC923.hsd1.ca.comcast.net)
# [07:49] <glandium> are pgo builds clobbers?
# [07:51] * Quits: bdahl (bdahl@moz-E197F13B.hsd1.ca.comcast.net) (Client exited)
# [07:51] <glandium> philor: do we have a graph of the linker memory usage?
# [07:51] <glandium> istr one was setup
# [07:51] <philor> I don't think it was
# [07:52] <glandium> damn. that would have been useful
# [07:53] <glandium> https://bugzilla.mozilla.org/show_bug.cgi?id=710840 :(
# [07:54] <philor> yep
# [07:54] * Quits: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net) (Quit: Mardak)
# [07:54] * Quits: josh (josh@moz-2EE66546.nyc.res.rr.com) (Ping timeout)
# [07:55] * Quits: dwendorf (Mibbit@moz-6E3B32E8.lightspeed.mdsnwi.sbcglobal.net) (Quit: http://www.mibbit.com ajax IRC Client)
# [07:55] <glandium> philor: someone with direct access to the logs could get the history pretty fast, though
# [07:55] <glandium> philor: as for the linux64 one, it doesn't make much sense :(
# [07:56] <philor> yeah, I was willing to call it need-for-a-clobber, once one of my retriggers none of which worked came up green
# [08:04] * glob|away is now known as glob
# [08:05] * Joins: josh (josh@moz-2EE66546.nyc.res.rr.com)
# [08:06] * bz is now known as bz_slee
# [08:06] * bz_slee is now known as bz_sleep
# [08:08] * Joins: Boriss (FlyingToas@moz-62AAA429.hsd1.ca.comcast.net)
# [08:11] * Quits: Philip (philip@moz-65AD4613.zaynar.co.uk) (Ping timeout)
# [08:11] * Quits: Nikopol (nikopol@moz-68B53A37.chaosorigin.com) (Ping timeout)
# [08:11] * Joins: Philip (philip@moz-65AD4613.zaynar.co.uk)
# [08:11] * Joins: gcp (gpascutto@moz-D0E475EA.access.telenet.be)
# [08:11] * Joins: Nikopol (nikopol@moz-68B53A37.chaosorigin.com)
# [08:13] * Quits: dbradley (dbradley@moz-80450F75.fuse.net) (Ping timeout)
# [08:14] * Joins: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net)
# [08:16] * Joins: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net)
# [08:18] * Quits: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net) (Quit: Mardak)
# [08:19] * Quits: anant_ (Anant@moz-488092DD.hsd1.ca.comcast.net) (Quit: Leaving)
# [08:20] * Joins: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net)
# [08:20] * Joins: Faramarz_ (Faramarz@moz-CD61C00E.hsd1.ca.comcast.net)
# [08:20] * Quits: Faramarz_ (Faramarz@moz-CD61C00E.hsd1.ca.comcast.net) (Connection reset by peer)
# [08:20] * glob is now known as glob|away
# [08:20] * Quits: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net) (Quit: Mardak)
# [08:26] * Quits: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP) (Ping timeout)
# [08:27] * Quits: WeirdAl (chatzilla@moz-5BDC219.hsd1.ca.comcast.net) (Quit: ChatZilla 0.9.88.2 [Firefox 11.0/20120313114517])
# [08:29] * Joins: pranavrc (pranavrc@1D7FB90B.8CAB716A.C28326FD.IP)
# [08:29] * Joins: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP)
# [08:30] * Quits: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net) (Quit: Leaving.)
# [08:30] * Joins: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net)
# [08:30] * philor really looks forward to dumping this tree in edmorley's lap
# [08:30] * Joins: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net)
# [08:30] * Joins: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP)
# [08:30] * Quits: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net) (Quit: Mardak)
# [08:30] <nigelb> philor: I see a meme there ;-)
# [08:31] * Quits: gwagner_ (idefix2@moz-B8B530C2.hsd1.ca.comcast.net) (Quit: gwagner_)
# [08:31] * Joins: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net)
# [08:31] * Quits: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net) (Quit: Mardak)
# [08:31] * Quits: Faramarz (Faramarz@moz-CD61C00E.hsd1.ca.comcast.net) (Quit: Leaving)
# [08:31] * Quits: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP) (Ping timeout)
# [08:32] * Quits: AryehGregor (AryehGrego@moz-83CA4768.red.bezeqint.net) (Connection reset by peer)
# [08:32] * Joins: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP)
# [08:33] * Joins: AryehGregor (AryehGrego@moz-83CA4768.red.bezeqint.net)
# [08:33] * Quits: hub (hub@moz-E2FCA694.figuiere.net) (Ping timeout)
# [08:35] <AryehGregor> Um . . . https://groups.google.com/forum/?fromgroups#!forum/mozilla.dev.tech.editor
# [08:35] <AryehGregor> If no one is paying any attention whatsoever to a group like that, shouldn't it be shut down?
# [08:35] * Joins: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net)
# [08:36] <philor> http://cdn.memegenerator.net/instances/400x/19768018.jpg
# [08:37] <@roc> AryehGregor: yes
# [08:37] <philor> well, since it has "reached the storage limit on your mailbox" it will be shut down, unless it clicks an unlinked link to upgrade its mailbox
# [08:38] * Quits: squib (squib@moz-B01B5D55.dhcp.mdsn.wi.charter.com) (Quit: Leaving)
# [08:39] * Joins: Ms2ger (Ms2ger@4E399CED.3067C44D.37724B0D.IP)
# [08:39] * AryehGregor likes how something just popped up a "Enter passphrase to unlock your private key" dialog and he reflexively typed in his password without even thinking about it
# [08:44] <AryehGregor> Wait, why is m-i closed? I don't understand the message: http://tinderbox.mozilla.org/Mozilla-Inbound/status.html
# [08:44] * Quits: priya (Adium@moz-5843392D.hsd1.ca.comcast.net) (Quit: Leaving.)
# [08:45] * Joins: gcp_ (gpascutto@moz-D0E475EA.access.telenet.be)
# [08:45] <nigelb> philor++
# [08:46] * Quits: gcp (gpascutto@moz-D0E475EA.access.telenet.be) (Ping timeout)
# [08:47] <heycam> AryehGregor, the builders are running out of memory while linking again
# [08:47] <AryehGregor> Okay.
# [08:48] <philor> AryehGregor: we used too much virtual memory building Win PGO, and fixed that with /3GB or whatever the commandline switch is to use 3GB, that was the first time; we used too much virtual memory building Win PGO, and fixed that by disabling some stuff, backing some stuff out, splitting some stuff out of libxul and switching to VS2010, that was Son of bustage; now we're (possibly) hosed, return of son of
# [08:48] * Fallen|away is now known as Fallen
# [08:49] * Quits: brambles (brambles@4CBAB088.F3076E90.1822ACA6.IP) (Ping timeout)
# [08:49] * Quits: gcp_ (gpascutto@moz-D0E475EA.access.telenet.be) (Quit: Make a new plan, Stan!)
# [08:49] * Joins: brambles (brambles@4CBAB088.F3076E90.1822ACA6.IP)
# [08:50] * Joins: gcp (gpascutto@moz-D0E475EA.access.telenet.be)
# [08:50] * Joins: graememcc (chatzilla@moz-DE5D49D8.range86-148.btcentralplus.com)
# [08:51] * Quits: gcp (gpascutto@moz-D0E475EA.access.telenet.be) (Quit: Make a new plan, Stan!)
# [08:51] * Quits: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP) (Ping timeout)
# [08:52] * Quits: jet (junglecode@moz-79F891EE.hsd1.ca.comcast.net) (Quit: jet)
# [08:53] <Callek> ooo we hit that limit _already_ wtf
# [08:53] <Callek> wow
# [08:53] <Callek> I'm tempted to say we need to stop doing PGO given this freaking limit
# [08:53] <nigelb> Ouch.
# [08:54] <Callek> not much else we can do until MSVC starts letting us link pgo in a much less memory intensive manner
# [08:54] <@roc> that would be a big performance regression
# [08:54] <Callek> roc: it would -- but, for example, Chrome does *not* build PGO anymore for this same linker reason
# [08:55] <Callek> we can gain 'some' leeway (with a bunch of untested pain) by building on win64 instead, but we'd be in this same point again in probably about a year, if we continue adding more code, -- and no reason to think we won't
# [08:56] * Quits: @roc (chatzilla@538BABFE.A073F3E.97BBD552.IP) (Ping timeout)
# [08:56] <Callek> and its only some leeway, given that we'd still only have a max of 4GB of RAM to use instead of 3GB
# [08:56] * Quits: sworkman (sworkman@moz-825EC923.hsd1.ca.comcast.net) (Quit: sworkman)
# [08:56] * glob|away is now known as glob
# [08:56] <AryehGregor> You can't cross-compile on Win64 so the linker can use however much memory it wants but you're still producing a 32-bit binary?
# [08:57] <Callek> AryehGregor: no 64-bit linker that supports 32-bit resulting binaries
# [08:57] <Callek> correct
# [08:57] <AryehGregor> :/
# [08:57] <Callek> if that existed, that *would* be the correct solution, if still a painful one
# [08:58] <AryehGregor> I would throw in an anti-Windows jab here, but I just had to restart my machine because I was getting weird errors about D-Bus not running and "sudo service dbus restart" didn't help.
# [08:58] <nigelb> hehe
# [08:58] * Joins: gcp (gpascutto@moz-D0E475EA.access.telenet.be)
# [08:58] <philor> took us two years to go from 2GB to 3GB, though the unoptimistic theory that 3GB to 4GB would only be a year is probably reasonable
# [08:58] * Quits: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net) (Quit: Leaving.)
# [08:59] <Callek> philor: given our amount of checkins-per day and the amount of incomming employees we have, people working on B2G and Web-App supporting code, etc. I just expect the actual binary code to increase that much in give or take a year if we switched to 64 bit builds
# [09:01] <philor> Nick's retrigger built, which raises the delightful question "now what?"
# [09:02] <tbsaunde> Callek: well, most 2g code should be linux only and so not really hurt, but your general point still stands I think
# [09:02] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
# [09:03] <Callek> tbsaunde: well a bunch of B2G work involves creating web API's/features that can be part of "web as the platform" as well, so it would also build with windows
# [09:03] <Callek> its just not the main B2G focus _yet_
# [09:04] <Callek> and then there is win8-metro work that will add to our PGO footprint
# [09:05] <@dolske> the port of MSVC to B2G will fix everything.
# [09:06] <Callek> dolske++
# [09:06] <nigelb> dolske++
# [09:07] * Joins: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net)
# [09:07] <@dolske> stop it
# [09:07] <@dolske> firebot doesn't even keep track! :P
# [09:07] <tbsaunde> Callek: yes, I wasn't disagreeing with your general argument
# [09:07] <nigelb> aww, but we do!
# [09:07] <@dolske> nigelb: rank
# [09:07] <firebot> dolske: Sorry, I've no idea what 'doesn't even keep track! :P' might be.
# [09:08] * Joins: atuljangra (Mibbit@4F882FCC.4850A4DD.1957C0DA.IP)
# [09:08] * @dolske hugs firebot
# [09:08] <firebot> I love you too, dolske.
# [09:08] <nigelb> dolske: Sorry, I've not idea what rank might be
# [09:08] * Quits: mike5w3c (MikeS@moz-2E518994.hfc.comcastbusiness.net) (Quit: mike5w3c)
# [09:09] <glob> dolske, 15
# [09:09] <glob> http://logbot.glob.com.au/?a=search&c=mozilla%23developers&q=dolske%2B%2B&ss=&;se=
# [09:11] <@dolske> 15?
# [09:12] <@dolske> _15?!_
# [09:12] <@dolske> that's hardly worth my time.
# [09:12] <@dolske> also, here is a kitten kissing a puppy. http://i.imgur.com/rHN6e.jpg
# [09:12] <glob> lol
# [09:12] <glob> @ kitten
# [09:12] * Quits: Stan (Stan@moz-B727CC82.dip.t-dialin.net) (Ping timeout)
# [09:14] * Joins: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP)
# [09:14] * Joins: maikmerten (maikmerten@moz-343F0059.dynamic.qsc.de)
# [09:16] <JonathanS> http://mozillamemes.tumblr.com/post/21924649342/more-helpfully-you-can-run-vim-inside-of-a oh my
# [09:18] * joduinn-home is now known as joduinn-afk
# [09:19] <smontagu> dholbert: ping
# [09:20] * Quits: @dveditz (dveditz@moz-5051E786.dhcp.cruzio.com) (Quit: dveditz)
# [09:20] <philor> a mere 2.81 GB on the retrigger, and we're open again, until the next time
# [09:21] <glandium> Callek: arguably, we could stop stucking everything and its friends in libxul
# [09:23] <Callek> glandium: I think its far easier to convince people to stop doing PGO than it is to *keep* things out of libxul (or move them out of libxul)
# [09:24] <Callek> glandium: feel free to prove me wrong though
# [09:24] <Ms2ger> Dammit, a bug bounty
# [09:24] <philor> because perf regressions are easy to swallow?
# [09:24] <philor> oh, wait, we ignore them now, don't we?
# [09:24] <Callek> philor: have you even seen .tree-mngmt for the past 4 days ;-)
# [09:24] <Callek> hehe yea
# [09:26] * philor is now known as philor|away
# [09:28] * Joins: karl (karl@moz-1EA98AF1.jetstream.xtra.co.nz)
# [09:29] * Quits: ctopper (craig@C3495DA.BA3DBA56.AE2B2F80.IP) (Quit: ctopper)
# [09:29] * Quits: Ms2ger (Ms2ger@4E399CED.3067C44D.37724B0D.IP) (Ping timeout)
# [09:31] * Quits: Havvy (Mibbit@moz-B7B34335.ptld.qwest.net) (Quit: Night)
# [09:33] * Quits: atuljangra (Mibbit@4F882FCC.4850A4DD.1957C0DA.IP) (Quit: http://www.mibbit.com ajax IRC Client)
# [09:35] <gaston> #error Target architecture was not detected as supported by Double-Conversion.
# [09:35] <gaston> grr.
# [09:35] * Quits: kvda (kvda@AFDD388F.C3893E67.923345AB.IP) (Quit: ( x____x))
# [09:39] * Joins: jet (junglecode@moz-79F891EE.hsd1.ca.comcast.net)
# [09:40] * Joins: Ms2ger (Ms2ger@4E399CED.3067C44D.37724B0D.IP)
# [09:42] * Quits: jduell (jduell@moz-2D9EDA98.lightspeed.sntcca.sbcglobal.net) (Ping timeout)
# [09:43] * Quits: Ms2ger (Ms2ger@4E399CED.3067C44D.37724B0D.IP) (Quit: bbl)
# [09:43] * Joins: kvda (kvda@AFDD388F.C3893E67.923345AB.IP)
# [09:45] * Quits: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP) (Quit: Computer has gone to sleep.)
# [09:54] <gaston> glandium: you might be interested in #750620
# [09:56] * Quits: njn (chatzilla@moz-C2274D74.dyn.iinet.net.au) (Client exited)
# [09:57] * Joins: victorporof (victorporo@5F6B43D3.1C233438.79933D60.IP)
# [09:57] * Joins: bholley (anonymous@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [09:57] * Quits: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP) (Quit: Leaving.)
# [09:58] * Quits: victorporof (victorporo@5F6B43D3.1C233438.79933D60.IP) (Quit: Linkinus - http://linkinus.com)
# [09:59] * Joins: cpearce (chatzilla@moz-510B10B9.xdsl.xnet.co.nz)
# [10:03] * Quits: sicking (chatzilla@moz-7F871C5C.hsd1.ca.comcast.net) (Ping timeout)
# [10:06] * Quits: maikmerten (maikmerten@moz-343F0059.dynamic.qsc.de) (Client exited)
# [10:09] * Joins: xakz (XaMaD@moz-34FBE388.fbx.proxad.net)
# [10:11] * Joins: martyn (martyn@moz-CABED5F2.range81-154.btcentralplus.com)
# [10:13] * Joins: Standard8Away (Standard8@B7F1AE36.48015583.54C3481B.IP)
# [10:13] * Standard8Away is now known as Standard8
# [10:15] * Quits: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk) (Quit: This computer has gone to sleep)
# [10:16] * Joins: edmorley (edmorley@moz-E455C402.range86-145.btcentralplus.com)
# [10:17] <nthomas|away> edmorley: ping
# [10:17] * nthomas|away is now known as nthomas
# [10:17] * Joins: maikmerten (maikmerten@moz-343F0059.dynamic.qsc.de)
# [10:18] <edmorley> nthomas: hi :-)
# [10:18] <nthomas> good morning! could I enlist your help to have a look at bug 750611 and help me figure out what to cancel ?
# [10:19] <nthomas> the original submission times are easy to get, if that helps
# [10:19] * Quits: brendan (brendaneic@moz-A286C218.hsd1.ca.comcast.net) (Quit: brendan)
# [10:20] <edmorley> nthomas: sure, I'll take a look
# [10:20] * nthomas posts those times
# [10:20] * Joins: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk)
# [10:21] * Joins: marco (Mibbit@DAD8D1D4.96B8F5F8.10DC0B64.IP)
# [10:21] <edmorley> nthomas: everything in comment 3 of that bug can be cancelled tbh
# [10:21] <marco> pure curiosity, what is DLBI?
# [10:22] <edmorley> periodic pgo will have run on inbound et al, whilst we were waiting for the triggered runs
# [10:22] <edmorley> nthomas: and it's more important that we can trigger now without a backlog :-)
# [10:22] * Quits: pranavrc (pranavrc@1D7FB90B.8CAB716A.C28326FD.IP) (Quit: Ping timeout: ∞)
# [10:23] * Quits: bnicholson (bnicholson@moz-BBE3ABD.mv.mozilla.com) (Input/output error)
# [10:23] <nthomas> edmorley: ok. We probably can't cancel PGO without needing to clobber though
# [10:23] <edmorley> nthomas: sorry I should have been more explicit; pending only
# [10:24] <nthomas> k. I'ma gonna cancel that oak nightly anyway
# [10:24] <edmorley> nthomas: I can always just clobber if needs be :-)
# [10:25] <Callek> edmorley: ooo since when were you employed btw
# [10:25] <Callek> nthomas: I CC'ed everyone on that list ;-)
# [10:27] <nthomas> bah, 27d8078fe517 is the tip of oak
# [10:27] <edmorley> Callek: 19th :-)
# [10:28] <Callek> congrats!
# [10:28] <edmorley> Callek: thank you
# [10:33] <edmorley> nthomas: oh except it was teasing me, clobberer still not coopoerating :-(
# [10:33] <nthomas> why did you want to clobber ?
# [10:34] * nthomas hasn't killed any pgo builds
# [10:34] <edmorley> nthomas: I was seeing if was working out of curiosity
# [10:34] <edmorley> in case it was needed
# [10:35] <edmorley> given its flakiness last week
# [10:35] * Quits: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP) (Ping timeout)
# [10:35] <nthomas> specifically https://build.mozilla.org/clobberer/?branch=mozilla-inbound ?
# [10:36] <edmorley> yeah
# [10:38] * Joins: chrisccoulson (chr1s@CF62D2D6.FC921812.EE27BB8D.IP)
# [10:40] <nthomas> loads for me in 49 seconds, but then bug 749081 implies I might be getting different treatment
# [10:41] <edmorley> nthomas: I'm happy for all of my self serve builds to be cancelled if that's ok, just to reduce the queue
# [10:41] * Joins: msucan (mihai@7A4CDDE7.7A418818.B4CEF140.IP)
# [10:41] * Joins: smagnin (pike@moz-1816673.w90-57.abo.wanadoo.fr)
# [10:42] <nthomas> I'm inclined to leave it right now
# [10:42] <edmorley> ok, I just suspect it will hold up the inbound merge later
# [10:42] <edmorley> since I'll need to diagnose the PGO bustage before I can merge it
# [10:43] <nthomas> won't clobber also hold you up ?
# [10:43] <edmorley> yup
# [10:43] <edmorley> happy days :-)
# [10:43] * Joins: squeakytoy (squeakytoy@moz-9E84F244.netset.se)
# [10:43] <KWierso> 500 error in 32 seconds for me when I try loading the inbound clobberer :)
# [10:43] <KWierso> for whatever that's worth :)
# [10:44] * Quits: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk) (Quit: This computer has gone to sleep)
# [10:44] <Callek> KWierso: that sounds like the timeout bumps didn't take, refresh or maybe needs to happen elsewhere
# [10:44] <Callek> KWierso: the bug implies it should timeout in 90 secs
# [10:44] * Joins: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk)
# [10:45] <KWierso> who clobbers the clobberer?
# [10:45] * Quits: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk) (Quit: This computer has gone to sleep)
# [10:45] <edmorley> I'm sure philor and a baseball bat are involved somewhere along the line...
# [10:46] <Standard8> lol
# [10:46] * NeilAway wonders why a printer would need csh
# [10:47] <smontagu> NeilAway: the driver seems to
# [10:47] <KWierso> Callek: I've refreshed (clearing cache) a number of times. all the timeouts hover at or just above 30 seconds
# [10:47] * Joins: jacek (jacek@moz-5D707D3B.psi.wroc.pl)
# [10:47] <Callek> KWierso: imo, mentioning that in bug would be helpful
# [10:48] <KWierso> Callek: done
# [10:50] * Quits: Boriss (FlyingToas@moz-62AAA429.hsd1.ca.comcast.net) (Quit: Boriss)
# [10:50] * Joins: Boriss (FlyingToas@moz-62AAA429.hsd1.ca.comcast.net)
# [10:55] * Joins: robhawkes (robhawkes@moz-33A339B7.dsl.cnl.uk.net)
# [10:57] * Quits: marco (Mibbit@DAD8D1D4.96B8F5F8.10DC0B64.IP) (Quit: http://www.mibbit.com ajax IRC Client)
# [10:59] <edmorley> nthomas: thanks for sorting that :-)
# [10:59] <nthomas> np
# [11:02] * Joins: pascalc (chatzilla@moz-446F98C8.fbxo.proxad.net)
# [11:02] * Quits: birtles (chatzilla@moz-348F61F0.mozilla.or.jp) (Quit: ChatZilla 0.9.88-rdmsoft [XULRunner 1.9.0.17/2009122204])
# [11:08] * Quits: smagnin (pike@moz-1816673.w90-57.abo.wanadoo.fr) (Quit: Quitte)
# [11:10] * Joins: alfredo- (Adium@88F51059.F3BBB17D.144F44FA.IP)
# [11:10] * Joins: Yoric (Yoric@moz-920DB13B.fbx.proxad.net)
# [11:11] * Joins: janv (varga@moz-5EF35A43.o2inet.sk)
# [11:16] * Joins: loadbang (loadbang@moz-28BBF200.range86-162.btcentralplus.com)
# [11:16] * Joins: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP)
# [11:18] * Joins: TheCrap (TheCrap@moz-A5E58BAF.dip0.t-ipconnect.de)
# [11:19] * AutomatedTester|AFK is now known as AutomatedTester
# [11:20] * Joins: allisterb (allisterb@2CE5E668.5725EB66.F43E9E20.IP)
# [11:21] * Quits: kvda (kvda@AFDD388F.C3893E67.923345AB.IP) (Quit: Computer has gone to sleep.)
# [11:26] * Joins: nestos (nestos@moz-72F38CF5.home.otenet.gr)
# [11:27] <gaston> mm when running jsapi-tests after mfbt/double-conversio landing i'm getting "lazy binding failed" on testConservativeGC
# [11:27] <gaston> and the trace comesfrom double_conversion::DoubleToStringConverter::EcmaScriptConverter();
# [11:27] <gaston> does that ring a bell to anyone ?
# [11:31] * Quits: kdc (chatzilla@moz-6176CD61.no.shawcable.net) (Ping timeout)
# [11:32] * Joins: dao (dao@moz-A0B4DC75.superkabel.de)
# [11:33] * Quits: loadbang (loadbang@moz-28BBF200.range86-162.btcentralplus.com) (Quit: loadbang)
# [11:33] * Joins: kdc (chatzilla@moz-6176CD61.no.shawcable.net)
# [11:33] * Joins: kaze (kaze@moz-7E0F0F9E.fbx.proxad.net)
# [11:33] * Quits: nestos (nestos@moz-72F38CF5.home.otenet.gr) (Quit: Leaving)
# [11:34] * Joins: dao1 (dao@moz-A0B4DC75.superkabel.de)
# [11:35] * Quits: dao (dao@moz-A0B4DC75.superkabel.de) (Ping timeout)
# [11:36] * Joins: kanru` (user@79E6271C.120775F3.B7D4332.IP)
# [11:36] * Quits: dao1 (dao@moz-A0B4DC75.superkabel.de) (Ping timeout)
# [11:37] * Joins: dao (dao@moz-A0B4DC75.superkabel.de)
# [11:39] * Quits: nrc (nrc@538BABFE.A073F3E.97BBD552.IP) (Ping timeout)
# [11:39] * Joins: lduros (lduros@moz-BED1C6A5.c3-0.rdl-ubr1.trpr-rdl.pa.cable.rcn.com)
# [11:40] * heycam is now known as heycam|away
# [11:40] * Joins: marco (Mibbit@DAD8D1D4.96B8F5F8.10DC0B64.IP)
# [11:41] * Joins: nestos (nestos@moz-72F38CF5.home.otenet.gr)
# [11:42] <marco> tn, do you know if the bug you talked about in https://bugzilla.mozilla.org/show_bug.cgi?id=728738#c37 was filed?
# [11:43] * Joins: dao1 (dao@moz-A0B4DC75.superkabel.de)
# [11:43] * Quits: dao (dao@moz-A0B4DC75.superkabel.de) (Ping timeout)
# [11:46] * Joins: joe_walker_ (joe_walker@moz-15405DDA.cable.virginmedia.com)
# [11:47] * Quits: cjones (cjones@moz-45913895.socal.res.rr.com) (Quit: Leaving)
# [11:48] * Joins: jfkthame (jfkthame@15AA2040.A1C12133.9542EC20.IP)
# [11:48] * Joins: tH (Rob@cpc16-seac19-2-0-cust234.7-2.cable.virginmedia.com)
# [11:49] * AutomatedTester is now known as AutomatedTester|AFK
# [11:49] * Joins: dria (dria@moz-2830A996.dhcp-dynamic.fibreop.nb.bellaliant.net)
# [11:50] * Quits: vikram360 (vikram360@195736B2.76910E54.2A068A5E.IP) (Ping timeout)
# [11:51] * glob is now known as glob|away
# [11:52] * Quits: ericb2 (X@moz-9C4C3DED.fbx.proxad.net) (Ping timeout)
# [11:54] * sancus is now known as sancus_
# [11:54] * sancus_ is now known as sancus
# [11:55] * Quits: pascalc (chatzilla@moz-446F98C8.fbxo.proxad.net) (Ping timeout)
# [11:55] * Joins: pascalc (chatzilla@moz-446F98C8.fbxo.proxad.net)
# [11:56] * Joins: tchevalier (Instantbir@moz-BA7C7754.w90-0.abo.wanadoo.fr)
# [11:57] * Quits: tchevalier (Instantbir@moz-BA7C7754.w90-0.abo.wanadoo.fr) (Quit: Instantbird -- http://www.instantbird.com)
# [11:57] * Joins: tchevalier (Instantbir@moz-BA7C7754.w90-0.abo.wanadoo.fr)
# [12:01] * nthomas is now known as nthomas|away
# [12:02] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
# [12:03] * Quits: kanru` (user@79E6271C.120775F3.B7D4332.IP) (Client exited)
# [12:10] * Quits: karl (karl@moz-1EA98AF1.jetstream.xtra.co.nz) (Ping timeout)
# [12:14] * Quits: a-865 (fmcz@moz-A5D13CA.cable.mindspring.com) (Quit: ChatZilla 0.9.88.2 [SeaMonkey 2.9/20120422224137])
# [12:19] * Quits: dao1 (dao@moz-A0B4DC75.superkabel.de) (Ping timeout)
# [12:19] * Joins: dao (dao@moz-A0B4DC75.superkabel.de)
# [12:20] * Parts: marco (Mibbit@DAD8D1D4.96B8F5F8.10DC0B64.IP)
# [12:21] * Joins: harth (harth@moz-C2C235AC.bb.sky.com)
# [12:25] * Joins: clokep (Instantbir@moz-69FB3955.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com)
# [12:26] * Joins: TheLink (TheLink@moz-2C115FCB.pools.arcor-ip.net)
# [12:30] * Joins: icaaq (Adium@moz-718D9655.cust.bredbandsbolaget.se)
# [12:34] * NeilAway thwaps hixie
# [12:35] <nigelb> ...
# [12:39] * Joins: smagnin (pike@moz-1816673.w90-57.abo.wanadoo.fr)
# [12:43] * Joins: a-865 (fmcz@moz-A5D13CA.cable.mindspring.com)
# [12:44] * Joins: Ms2ger (Ms2ger@moz-997EEB55.access.telenet.be)
# [12:45] <edmorley> Ms2ger: good morning :-)
# [12:45] <Ms2ger> Morning :)
# [12:46] <Callek> NeilAway: whatd he do?
# [12:46] * Quits: decoder (quassel@moz-216446B9.own-hero.net) (Ping timeout)
# [12:48] * Quits: @ted (luser@moz-123C3802.scr.east.verizon.net) (Ping timeout)
# [12:48] <Ms2ger> edmorley, http://cdn.memegenerator.net/instances/400x/19768018.jpg was for you
# [12:48] <AryehGregor> Can someone explain to me why there's no automatic conversion between nsINode and nsIDOMNode and so on? Why can't it call do_QueryInterface or something automatically?
# [12:49] * Joins: decoder (quassel@moz-216446B9.own-hero.net)
# [12:49] <AryehGregor> Also, since Ms2ger is here: [120429 19:35:40] <Ms2ger> address_of(*outNode)
# [12:49] * Quits: janv (varga@moz-5EF35A43.o2inet.sk) (Quit: This computer has gone to sleep)
# [12:49] <AryehGregor> What does address_of do anyway?
# [12:49] <AryehGregor> As opposed to, say, &?
# [12:50] * sancus is now known as sancus_
# [12:50] * sancus_ is now known as sancus
# [12:50] <Ms2ger> I dunno
# [12:50] <Ms2ger> & might give a Foo**?
# [12:50] <Ms2ger> AryehGregor, because QI is expensive :)
# [12:51] <AryehGregor> So we follow the C philosophy of "if something is expensive, make it require lots of typing to discourage programmers from doing it, and hope they don't just write macros to get around it"?
# [12:51] <Ms2ger> :)
# [12:51] <Ms2ger> Really, though, history
# [12:52] <edmorley> Ms2ger: ha :-)
# [12:52] * Joins: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk)
# [12:52] <Ms2ger> And the fact that those conversions wouldn't have been terribly useful, with nsIDOM3Node and nsIDOMNSNode and friends
# [12:52] * Joins: eflores_ (eflores@538BABFE.A073F3E.97BBD552.IP)
# [12:52] * Quits: eflores (eflores@538BABFE.A073F3E.97BBD552.IP) (Connection reset by peer)
# [12:53] <tbsaunde> AryehGregor: ssuming people don't apply crazy macros it also makes it more obvious something heavy weight is being done
# [12:53] <AryehGregor> Well, people write methods all the time that overload nsIDOMNode* and nsINode* and just have one do do_QueryInterface and pass the result to the other.
# [12:53] <AryehGregor> . . . at least in editor/ code.
# [12:53] <Ms2ger> Well
# [12:54] <Ms2ger> That's because editor used to just use nsIDOMNode everywher
# [12:54] <Ms2ger> e
# [12:54] <Ms2ger> And that's dog slow
# [12:54] <Ms2ger> So we're transitioning to nsINode, but it's a lot of code
# [12:55] * Joins: dbradley (dbradley@moz-80450F75.fuse.net)
# [12:55] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
# [12:56] <Ms2ger> And this is better than having only methods that take nsIDOMNode and immediately QI to nsINode :)
# [12:56] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [12:56] * Quits: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net) (Quit: Leaving.)
# [12:57] * Quits: a-865 (fmcz@moz-A5D13CA.cable.mindspring.com) (Quit: ChatZilla 0.9.88.2 [SeaMonkey 2.9/20120422224137])
# [12:57] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [12:57] * Quits: cpearce (chatzilla@moz-510B10B9.xdsl.xnet.co.nz) (Ping timeout)
# [12:58] <AryehGregor> Why is nsIDOMNode slow?
# [12:58] <Ms2ger> Everything you call on it is a virtual call
# [12:59] <tbsaunde> everything on it is virtual?
# [12:59] <tbsaunde> f
# [12:59] * Quits: dao (dao@moz-A0B4DC75.superkabel.de) (Ping timeout)
# [12:59] * Joins: dao (dao@moz-A0B4DC75.superkabel.de)
# [12:59] <Ms2ger> And editor then checks the return value, even if it's pointless
# [13:00] <tbsaunde> and if it gives you an object it addres itso you have extra refcounting
# [13:01] <Ms2ger> Consider the iterating over children in https://bugzilla.mozilla.org/attachment.cgi?id=619409&action=diff :)
# [13:02] <edmorley> sad faces https://bugzilla.mozilla.org/show_bug.cgi?id=710840#c8
# [13:03] * Ms2ger poofs
# [13:04] * Joins: ted (luser@moz-2854A494.scr.east.verizon.net)
# [13:04] * ChanServ sets mode: +o ted
# [13:05] * Quits: Ms2ger (Ms2ger@moz-997EEB55.access.telenet.be) (Ping timeout)
# [13:06] * Quits: clokep (Instantbir@moz-69FB3955.c3-0.smr-ubr1.sbo-smr.ma.cable.rcn.com) (Quit: Instantbird -- http://www.instantbird.com)
# [13:07] * Joins: a-865 (fmcz@moz-A5D13CA.cable.mindspring.com)
# [13:13] <NeilAway> Callek: nothing important
# [13:13] * Quits: dao (dao@moz-A0B4DC75.superkabel.de) (Quit: Leaving.)
# [13:14] <Callek> heh was just curious :-)
# [13:14] * khuey|away is now known as khuey
# [13:15] <khuey> philor|away: if I post a captain hindsight meme does that count as saving us?
# [13:15] <edmorley> lol
# [13:15] * Fallen is now known as Fallen|away
# [13:15] <Callek> khuey: YES
# [13:15] <edmorley> khuey: I'm just filing a new bug to track the failures
# [13:15] <Callek> khuey: and if not, post it anyway
# [13:16] <khuey> or maybe super cool ski instructor is more appropriate
# [13:17] <khuey> "if you don't do anything with the five months of runway we bought, you're going to have a bad time"
# [13:17] <Callek> edmorley, khuey: I *really* am beggining to think we're better of dropping PGO, and just trusting -O2 or some such overall
# [13:17] <Callek> s/of dropping/off dropping/
# [13:18] * Quits: jfkthame (jfkthame@15AA2040.A1C12133.9542EC20.IP) (Quit: jfkthame)
# [13:19] * Joins: harsh (Mibbit@1A4071A.15C6DF7B.1551A00F.IP)
# [13:20] <khuey> edmorley: I would think we could look at the nightly build logs and narrow down the regressions pretty quickly
# [13:20] <khuey> edmorley: unless it's just been a slow gradual rise :-/
# [13:21] * Quits: maikmerten (maikmerten@moz-343F0059.dynamic.qsc.de) (Client exited)
# [13:21] <edmorley> yeah just doing that
# [13:21] * mattwoodrow is now known as mattwoodrow|away
# [13:22] * aja hopes it wasn't khuey's leak fix
# [13:23] <Callek> aja: haha :-)
# [13:23] * aja ducks
# [13:23] <khuey> ha
# [13:24] <aja> nice work, btw
# [13:27] * Quits: jet (junglecode@moz-79F891EE.hsd1.ca.comcast.net) (Quit: jet)
# [13:27] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [13:28] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [13:28] * Joins: JeroenDeDauw (j@moz-95881CDD.dip.t-dialin.net)
# [13:29] * edmorley unleashes the bugmail
# [13:30] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [13:30] * Joins: sgautherie (chatzilla@moz-D7B69DC4.fbxo.proxad.net)
# [13:31] <khuey> srsly
# [13:32] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [13:33] <gaston> arghhh
# [13:33] <gaston> media/libvpx/vp8/common/postproc.c:869: error: 'VP8_COMMON' has no member named 'rtcd'
# [13:34] * khuey is now known as khuey|away
# [13:34] * khuey|away is now known as khuey
# [13:34] <khuey> there's a bug
# [13:34] <gaston> m-c, y u do not want to complete a build ?
# [13:34] <gaston> ah, olready? nice
# [13:34] <edmorley> glandium: sorry, was trying to remember everyone who might be interested and failed :-)
# [13:37] <gaston> khuey: remember the # ? search in bugzilla doesn't yield much
# [13:38] <khuey> gaston: gmail says 750447
# [13:38] <gaston> thanks
# [13:38] * Quits: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com) (Ping timeout)
# [13:39] <gaston> (there's a problem when gmail search is better than bugzilla search...)
# [13:39] <Callek> gaston: it helps when you organize your bugzilla e-mail certain ways
# [13:39] <Callek> ;-)
# [13:40] <gaston> thank $deity i dont receive mail for every open bug
# [13:42] * Quits: smontagu (chatzilla@moz-A6D65AB5.red.bezeqint.net) (Ping timeout)
# [13:42] <edmorley> khuey: I believe we should close the trees, would you agree?
# [13:43] <Callek> edmorley: until you figure out if its a small range or a gradual increase I would agree, fwiw
# [13:44] <khuey> edmorley: is one out of every couple pgo builds dying, or are they all dying?
# [13:44] * Quits: bholley (anonymous@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Ping timeout)
# [13:44] * Joins: bholley (anonymous@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [13:44] <edmorley> khuey: buildapi retriggers was only just fixed this morning, so bit of a backlog, so not quite sure yet
# [13:45] <khuey> edmorley: ok
# [13:45] <edmorley> khuey: seems maybe 50%
# [13:45] <edmorley> https://tbpl.mozilla.org/?tree=Mozilla-Inbound&jobname=WINNT%205.2%20mozilla-inbound%20pgo-build
# [13:46] <edmorley> (might have to page down to see the first failure)
# [13:46] <khuey> edmorley: so if I'm reading this correctly the last green PGO build on inbound used 30 MB more memory than the PGO build on central
# [13:47] <edmorley> ah so I'll look at the range since last merge
# [13:47] * Quits: nestos (nestos@moz-72F38CF5.home.otenet.gr) (Input/output error)
# [13:48] * Quits: harsh (Mibbit@1A4071A.15C6DF7B.1551A00F.IP) (Quit: http://www.mibbit.com ajax IRC Client)
# [13:49] * Joins: smontagu (chatzilla@moz-A6D65AB5.red.bezeqint.net)
# [13:49] * Quits: tonymec (tonymec@78D6A323.3F6FF523.277517C1.IP) (Ping timeout)
# [13:50] * Quits: tonymec|away (tonymec@78D6A323.3F6FF523.277517C1.IP) (Ping timeout)
# [13:50] <khuey> ted: we really should tinderboxprint linker max virtual size
# [13:51] <khuey> edmorley: so
# [13:52] <khuey> https://tbpl.mozilla.org/php/getParsedLog.php?id=11339562&tree=Mozilla-Inbound&full=1
# [13:52] <khuey> linker max virtual size: 2982510592
# [13:52] <khuey> https://tbpl.mozilla.org/php/getParsedLog.php?id=11345300&tree=Mozilla-Inbound&full=1
# [13:52] <khuey> linker max virtual size: 3021185024
# [13:52] * Joins: jfkthame (jfkthame@moz-984B1A01.eduroam.brookes.ac.uk)
# [13:52] <khuey> those are only a few hours apart
# [13:53] * Quits: lduros (lduros@moz-BED1C6A5.c3-0.rdl-ubr1.trpr-rdl.pa.cable.rcn.com) (Client exited)
# [13:53] <khuey> http://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=0e2658794e06&tochange=f8c388f622f1
# [13:54] * Joins: tonymec (tonymec@78D6A323.3F6FF523.277517C1.IP)
# [13:54] <khuey> there's not much in there that looks like it should affect windows
# [13:54] * Quits: Yoric (Yoric@moz-920DB13B.fbx.proxad.net) (Input/output error)
# [13:54] * Joins: tonymec|away (tonymec@78D6A323.3F6FF523.277517C1.IP)
# [13:55] <edmorley> hnmmm
# [13:55] * Joins: espindola (espindola@moz-97E8149E.dsl.teksavvy.com)
# [13:55] * Joins: testman42 (Mibbit@moz-665CC0CD.dial-up.dsl.siol.net)
# [13:57] <khuey> edmorley: also the last inbound merge to m-c seems to have added 30ish MB as well
# [13:57] <edmorley> khuey: I'm just quickly catching up with starring and I'll start collating figures from as many of the logs as I can
# [13:58] <khuey> hmm, though the one before that was higher
# [13:58] * khuey wonders what the normal fluctuation here is
# [13:58] <testman42> Yo. Is Firefox going to support more mouse buttons soon? I have 7 clicks on my mouse and only 3 work the way I want them to. Mouse Button 4 and 5 are used as Back and Forward navigation, but I would like them to switch between tabs. How do I do that?
# [13:59] <gcp> use your mouse's software?
# [13:59] <testman42> didnt get any
# [14:00] <testman42> got just mouse, no software
# [14:00] <testman42> and I am on Linux
# [14:00] <testman42> even if i got something it probably wouldnt run on linux
# [14:00] * Quits: KWierso (chatzilla@moz-DDEBF960.desm.qwest.net) (Ping timeout)
# [14:01] * Quits: chrisccoulson (chr1s@CF62D2D6.FC921812.EE27BB8D.IP) (Quit: Ex-Chat)
# [14:02] * Joins: chrisccoulson (chr1s@CF62D2D6.FC921812.EE27BB8D.IP)
# [14:02] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
# [14:04] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
# [14:04] <testman42> I checked about:config and found just a lot of "mouse wheel + some button" settings, but no any other mouse button settings.
# [14:05] * Joins: bjacob (bjacob@moz-ADCA75DC.cpe.net.cable.rogers.com)
# [14:05] <testman42> Is there any way I can bind some function to mouse button?
# [14:06] <testman42> say, mouse button 4 becomes [Ctrl+Shift+Tab] and mouse button 5 becomes [Ctrl+Tab] ?
# [14:07] * Joins: mike5w3c (MikeS@moz-2E518994.hfc.comcastbusiness.net)
# [14:08] * jhopkins|afk is now known as jhopkins
# [14:09] * glob|away is now known as glob
# [14:10] * Joins: KaiRo (robert@moz-88135F74.adsl.highway.telekom.at)
# [14:10] * Joins: vikram360 (vikram360@195736B2.76910E54.2A068A5E.IP)
# [14:10] <edmorley> khuey: nightly is appalling loading these logs :-(
# [14:11] <khuey> heh
# [14:11] <edmorley> before I could just balme it on my old desktop at home; now this is with 8gb ram, core i7 and ssd :-(
# [14:12] * Quits: dbradley (dbradley@moz-80450F75.fuse.net) (Quit: )
# [14:12] <jfkthame> edmorley: grab a profile?
# [14:12] <khuey> I wish we had someone who's job was to rip out dead code
# [14:12] <khuey> we've got a bunch of crap that's almost dead
# [14:12] <khuey> old html parser, rdf
# [14:13] <darktrojan> almost :/
# [14:13] <khuey> right
# [14:13] <edmorley> jfkthame: yeah I'll gladly do that later today :-) (wiki page?)
# [14:13] * Joins: dseif (dseif@78676049.531FD64C.630E4E47.IP)
# [14:13] <edmorley> jfkthame: profiling build + that addon right?
# [14:13] * Joins: Stan (Stan@moz-9E1FC00B.dip.t-dialin.net)
# [14:13] <jfkthame> edmorley: dunno offhand - what platform are you on?
# [14:14] <edmorley> win32
# [14:14] <edmorley> (and yeah, \o/)
# [14:14] <jfkthame> ah - it's a while since i tried profiling there, i don't remember the details
# [14:14] * Quits: bsmith (bsmith@A13161C7.979D6A3B.E017DF26.IP) (Ping timeout)
# [14:16] <edmorley> jfkthame: str being: open 2-3 of these type logs at once https://tbpl.mozilla.org/php/getParsedLog.php?id=11302882&tree=Mozilla-Inbound
# [14:17] <edmorley> and try to carry on scrolling down tbpl
# [14:18] <jfkthame> i'm used to tbpl getting dog-slow anyway - but i tend to assume that's at least partly because i'm normally running debug builds
# [14:19] * Joins: mjschranz (mjschranz@moz-6FE6B833.cpe.net.cable.rogers.com)
# [14:19] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
# [14:21] * Joins: mconley (mconley@moz-D640D16C.cable.teksavvy.com)
# [14:22] * Quits: mjschranz (mjschranz@moz-6FE6B833.cpe.net.cable.rogers.com) (Client exited)
# [14:23] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [14:24] * Quits: icaaq (Adium@moz-718D9655.cust.bredbandsbolaget.se) (Quit: Leaving.)
# [14:24] <edmorley> jfkthame: I've just resorted to Chrome for now :-/
# [14:24] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [14:25] * Joins: nattokirai (nattokirai@moz-6A258170.dynamic.ppp.asahi-net.or.jp)
# [14:25] * Joins: dwendorf (Mibbit@moz-6E3B32E8.lightspeed.mdsnwi.sbcglobal.net)
# [14:25] <jfkthame> edmorley: that's sad :(
# [14:26] <edmorley> (just for these logs)
# [14:26] <edmorley> I'm going a few weeks back x several pgo a day
# [14:26] * Quits: testman42 (Mibbit@moz-665CC0CD.dial-up.dsl.siol.net) (Quit: http://www.mibbit.com ajax IRC Client)
# [14:26] * sancus is now known as sancus_
# [14:26] * sancus_ is now known as sancus
# [14:26] * Joins: testman42 (Mibbit@moz-665CC0CD.dial-up.dsl.siol.net)
# [14:30] * Joins: jprmc (jprmc@78763BD4.3CBDB8AB.DA78B690.IP)
# [14:30] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [14:30] <testman42> Do I really have to hack into Firefox to change few mouse button settings? Is there no more simple way?
# [14:31] * rail_away is now known as rail
# [14:31] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [14:32] <edmorley> jfkthame: on the plus side, in page search is a lot slower in Chrome
# [14:32] <edmorley> for those logs at least
# [14:32] <jfkthame> i wonder why
# [14:32] <jfkthame> not that i'm complaining ;)
# [14:33] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
# [14:33] * Quits: gcp (gpascutto@moz-D0E475EA.access.telenet.be) (Quit: Make a new plan, Stan!)
# [14:37] * Quits: dseif (dseif@78676049.531FD64C.630E4E47.IP) (Input/output error)
# [14:39] <glob> hrm... [20:29:17.581] NS_ERROR_FAILURE: Failure @ null:4294967295
# [14:39] * Quits: mconley (mconley@moz-D640D16C.cable.teksavvy.com) (Input/output error)
# [14:41] * Joins: dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net)
# [14:41] * ChanServ sets mode: +o dbaron
# [14:42] * Joins: pranavrc (pranavrc@F7F17312.87C59D23.C28326FD.IP)
# [14:43] * Joins: aleth (Instantbir@moz-24EE697D.ictp.it)
# [14:45] <glob> anyone else seeing NS_ERROR_FAILURE: Failure @ null:4294967295 when loading planet?
# [14:47] <khuey> no
# [14:47] <zzzzz> not in the Error Console, I'm not running a debug build either -
# [14:48] * Quits: jprmc (jprmc@78763BD4.3CBDB8AB.DA78B690.IP) (Ping timeout)
# [14:48] <zzzzz> I do see Warning: Specified "type" of "video/mp4" is not supported. Load of media resource http://cf.cdn.vid.ly/0l4v5g/mp4.mp4 failed.
# [14:48] <zzzzz> Source file: http://planet.mozilla.org/
# [14:48] <glob> ok, extension fodder then, thanks :)
# [14:48] <glob> also, null has a lot of lines
# [14:49] <darktrojan> hmm
# [14:49] * Quits: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk) (Quit: This computer has gone to sleep)
# [14:50] * Joins: Mitch (chatzilla@moz-8D46E76F.sbr802.nsw.optusnet.com.au)
# [14:50] * Joins: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk)
# [14:51] * Joins: joey (chatzilla@moz-EFCB4CBF.princetowncable.com)
# [14:51] * Parts: aleth (Instantbir@moz-24EE697D.ictp.it)
# [14:53] * Joins: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
# [14:53] * Joins: dseif (dseif@moz-57B23647.senecac.on.ca)
# [14:54] * Joins: Enn (enn@F2D29657.F60B0462.67AC9B1.IP)
# [14:54] * Quits: joe_walker_ (joe_walker@moz-15405DDA.cable.virginmedia.com) (Ping timeout)
# [14:56] * Joins: SeoZ (DanielJuyu@E9B5BAA5.F5246840.EAF0BD7A.IP)
# [14:56] * Joins: dbradley (dbradley@moz-80450F75.fuse.net)
# [14:57] * Quits: bc (bc@moz-E31CD2CB.mozilla.org) (Ping timeout)
# [14:57] * Joins: bc (bc@moz-E31CD2CB.mozilla.org)
# [15:00] * Joins: gcp (gpascutto@moz-D0E475EA.access.telenet.be)
# [15:03] * Joins: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com)
# [15:03] * Joins: Yoric (Yoric@moz-920DB13B.fbx.proxad.net)
# [15:04] * Quits: kanha (quassel@F4BC79F1.8293D34.9105FBCF.IP) (Ping timeout)
# [15:05] * khuey is now known as khuey|away
# [15:06] * Joins: Neil (neil@moz-7E027EE5.cable.virginmedia.com)
# [15:06] * Joins: kanha (quassel@F4BC79F1.8293D34.9105FBCF.IP)
# [15:06] * sancus is now known as sancus_
# [15:06] * sancus_ is now known as sancus
# [15:07] * Quits: NeilZZZ (neil@moz-7E027EE5.cable.virginmedia.com) (Ping timeout)
# [15:07] * Neil is now known as NeilZZZ
# [15:07] * Joins: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se)
# [15:08] * sancus is now known as sancus_
# [15:08] * sancus_ is now known as sancus
# [15:09] * sancus is now known as sancus_
# [15:09] * Joins: mjschranz (mjschranz@F1451709.44D93D66.1139E686.IP)
# [15:09] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
# [15:10] * Joins: armenzg (armenzg@moz-ED3249A4.dmz.releng.scl3.mozilla.com)
# [15:11] * bwinton_away is now known as bwinton
# [15:13] * Joins: AaronMT (AaronMT@moz-E26428A8.cpe.net.cable.rogers.com)
# [15:14] * Joins: mwu (mwu@moz-59435430.hsd1.nj.comcast.net)
# [15:16] * Joins: jimm (jmathies@moz-7F164CA1.pn.at.cox.net)
# [15:17] * Quits: Yoric (Yoric@moz-920DB13B.fbx.proxad.net) (Input/output error)
# [15:17] * sancus_ is now known as sancus
# [15:19] * Quits: nhirata (nhirata.bu@moz-585D5EFC.hsd1.ca.comcast.net) (Quit: nhirata)
# [15:20] * Joins: mconley (mconley@F2D29657.F60B0462.67AC9B1.IP)
# [15:20] * Joins: bbondy (bbondy@moz-C9962B2.home.cgocable.net)
# [15:21] * Quits: testman42 (Mibbit@moz-665CC0CD.dial-up.dsl.siol.net) (Quit: http://www.mibbit.com ajax IRC Client)
# [15:21] * Quits: SeoZ (DanielJuyu@E9B5BAA5.F5246840.EAF0BD7A.IP) (Quit: Leaving)
# [15:23] * sancus is now known as sancus_
# [15:23] * sancus_ is now known as sancus
# [15:24] * Quits: harth (harth@moz-C2C235AC.bb.sky.com) (Input/output error)
# [15:24] * Joins: Cwiiis (cwiiis@88F51059.F3BBB17D.144F44FA.IP)
# [15:24] * Quits: darktrojan (geoff@moz-30B3CCFD.telstraclear.net) (Quit: darktrojan)
# [15:26] * jlund|away is now known as jlund
# [15:27] * Quits: TheLink (TheLink@moz-2C115FCB.pools.arcor-ip.net) (Client exited)
# [15:28] * coop|afk is now known as coop
# [15:29] * Joins: ajuma (ajuma@F2D29657.F60B0462.67AC9B1.IP)
# [15:29] * Quits: bc (bc@moz-E31CD2CB.mozilla.org) (Ping timeout)
# [15:30] * catlee-away is now known as catlee
# [15:30] * Quits: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Connection reset by peer)
# [15:30] * Joins: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
# [15:33] * Joins: bc (bc@moz-E31CD2CB.mozilla.org)
# [15:34] * Joins: armenzg_ (armenzg@moz-ED3249A4.dmz.releng.scl3.mozilla.com)
# [15:34] * Quits: armenzg (armenzg@moz-ED3249A4.dmz.releng.scl3.mozilla.com) (Ping timeout)
# [15:35] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [15:36] * Quits: jfkthame (jfkthame@moz-984B1A01.eduroam.brookes.ac.uk) (Quit: jfkthame)
# [15:36] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [15:38] * Joins: krit (krit@moz-3AE657AD.hsd1.ca.comcast.net)
# [15:38] * Joins: krit1 (krit@moz-3AE657AD.hsd1.ca.comcast.net)
# [15:38] * rail is now known as rail-brb
# [15:38] * Joins: kats (kats@moz-DE13FC16.compute-1.amazonaws.com)
# [15:38] * Parts: kats (kats@moz-DE13FC16.compute-1.amazonaws.com)
# [15:39] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
# [15:39] * Quits: krit (krit@moz-3AE657AD.hsd1.ca.comcast.net) (Ping timeout)
# [15:40] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
# [15:41] * Joins: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net)
# [15:43] * bear-afk is now known as bear
# [15:43] * Joins: KWierso (chatzilla@moz-DDEBF960.desm.qwest.net)
# [15:44] * Quits: sgautherie (chatzilla@moz-D7B69DC4.fbxo.proxad.net) (Client exited)
# [15:44] * Joins: madhava (madhava@F2D29657.F60B0462.67AC9B1.IP)
# [15:45] * Quits: squeakytoy (squeakytoy@moz-9E84F244.netset.se) (Quit: Leaving)
# [15:48] * armenzg_ is now known as armenzg_buildduty
# [15:48] * Joins: ehsan (ehsan@F2D29657.F60B0462.67AC9B1.IP)
# [15:48] * ChanServ sets mode: +o ehsan
# [15:48] * Joins: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net)
# [15:50] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [15:51] <@ted> whee more PGO fun
# [15:51] <@ted> khuey|away: i guess tinderboxprint would get it on tbpl?
# [15:52] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [15:53] <jaws> smontagu: ping?
# [15:54] <jaws> smontagu: for bug 750551, should the reftests explicitly set the text-align since RTL?
# [15:54] * jaws is now known as jaws|brb
# [15:55] * jaws|brb is now known as jaws
# [15:55] <edmorley> ted: if that's what gets it on tbpl, that would be awesome, loading each log manually is a pita
# [15:55] * Joins: lmandel (lmandel@F2D29657.F60B0462.67AC9B1.IP)
# [15:56] <@ted> edmorley: i'm not totally sure, you'd have to check how tbpl works
# [15:56] * Quits: martyn (martyn@moz-CABED5F2.range81-154.btcentralplus.com) (Connection reset by peer)
# [15:56] <@ted> but if that's all it takes, then rs=me to just change that
# [15:56] * gregglind_away is now known as gregglind
# [15:56] * Joins: Optimizer (Mibbit@6E78482A.3686F45E.89AC0F27.IP)
# [15:58] * Joins: vikash (vikash@C7D00EB1.38F0169.5D9ABA9F.IP)
# [15:59] <jlebar> bz_sleep, When you're awake: nsTArray::Swap explicitly lets you swap between arrays with different allocators. So for your fallible --> infallible thing, I'd try that and see if it's fast enough.
# [16:00] * Quits: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com) (Input/output error)
# [16:00] * Parts: Optimizer (Mibbit@6E78482A.3686F45E.89AC0F27.IP)
# [16:01] <smontagu> jaws: do you mean as future-proofing? atm it doesn't seem to make a difference
# [16:02] * Quits: Mardak (Mardak@moz-4FA48382.hsd1.ca.comcast.net) (Quit: Mardak)
# [16:02] * rail-brb is now known as rail
# [16:02] * Quits: dwendorf (Mibbit@moz-6E3B32E8.lightspeed.mdsnwi.sbcglobal.net) (Quit: http://www.mibbit.com ajax IRC Client)
# [16:02] * mcote|afk is now known as mcote
# [16:02] <jaws> smontagu: not as future proofing, i just thought it was needed, but if not then that's cool too :)
# [16:03] * Joins: dwendorf (Mibbit@moz-6E3B32E8.lightspeed.mdsnwi.sbcglobal.net)
# [16:07] * Joins: adeel (chatzilla@4D641B2E.1A901703.59D8DB68.IP)
# [16:07] * Joins: harth (harth@88F51059.F3BBB17D.144F44FA.IP)
# [16:08] * Joins: jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP)
# [16:09] * Joins: mcot (mcot@C4B02.F3C4E8F3.C8444B8.IP)
# [16:09] <smontagu> jaws: with |-moz-box-orient: vertical| apparently dir doesn't affect default alignement
# [16:09] * Quits: wlach|afk (wlach@moz-14AFFC14.vif.net) (Quit: Leaving)
# [16:09] * Quits: adeel (chatzilla@4D641B2E.1A901703.59D8DB68.IP) (Quit: ChatZilla 0.9.88.1 [SeaMonkey 2.8/20120312224753])
# [16:09] * Joins: mib_lg9bn1 (Mibbit@76298C95.818589BB.979C46D2.IP)
# [16:10] <@ehsan> edmorley: ping
# [16:10] <jaws> ok, i was seeing it when i reported the bug, but that could have been part of the bug
# [16:10] * Joins: KLB (Kenneth_Ba@moz-C233EF2B.maine.res.rr.com)
# [16:10] * Quits: dseif (dseif@moz-57B23647.senecac.on.ca) (Input/output error)
# [16:10] <smontagu> jaws: btw, I don't see the bug with about:home, which is interesting
# [16:11] <edmorley> ehsan: pong
# [16:11] * Joins: wlach (wlach@moz-14AFFC14.vif.net)
# [16:11] <@ehsan> edmorley: hey did you see my email?
# [16:11] <edmorley> ehsan: yeah
# [16:11] <edmorley> did you see their reply?
# [16:11] <jaws> smontagu: i saw the bug by using Force RTL and then hitting refresh
# [16:11] <@ehsan> edmorley: no, where?
# [16:11] <@ehsan> in the bug?
# [16:11] <edmorley> yeah
# [16:12] <@ehsan> oh
# [16:12] <edmorley> saying it isn't in libxul unfortunately
# [16:12] <@ehsan> well
# [16:12] <@ehsan> wait
# [16:12] <jaws> maybe Force RTL isn't working the same as a true RTL build :-/ ?
# [16:12] <smontagu> I have a kind of manual ForceRTL by setting intl.uidirection.en to RTL in about:config
# [16:12] <@ehsan> jaws: it does not :)
# [16:12] <@ehsan> it does its best
# [16:13] <smontagu> ehsan: what does it do exactly?
# [16:13] <@ehsan> smontagu: iirc it toggles the pref, and tries to handle some of the locale.dir stuff
# [16:13] <@ehsan> but it fails on some cases
# [16:13] * Joins: drice (derice@1606D15F.E628B196.8E155D4E.IP)
# [16:13] <@ehsan> edmorley: yeah, glandium is right, I misread the patch :(
# [16:13] <@ehsan> edmorley: anything big landed recently besides this?
# [16:14] <edmorley> ehsan: have you seen bug 750661 comment 2?
# [16:14] * Joins: anky (anky@F4051266.244268D6.A3D1B221.IP)
# [16:14] * Quits: anky (anky@F4051266.244268D6.A3D1B221.IP) (Quit: Leaving...)
# [16:14] * Joins: brendan (brendaneic@moz-A286C218.hsd1.ca.comcast.net)
# [16:14] <@ted> that's a bummer
# [16:14] <edmorley> (newest csets first in that list)
# [16:14] <@ted> did the opus codec stuff add a lot of other code to libxul?
# [16:14] * @ehsan looks
# [16:14] * BenWa|sms is now known as BenWa
# [16:15] <@ehsan> ted: no, it's in gkmedias
# [16:15] <@ted> right, but
# [16:15] <@ted> i mean the dom glue etc
# [16:15] <glandium> Here's a random thought from someone who is on a day off: maybe we can split libxul pgo. lib.exe, which creates static libs, has the ltcg option for PGO. Maybe we can use it on the intermediate libraries, and skip it for the final libxul linkage.
# [16:15] * Quits: wlach (wlach@moz-14AFFC14.vif.net) (Connection reset by peer)
# [16:15] <@ehsan> ted: could be, I only read the second patch, and I failed at doing that properly!
# [16:15] <decoder> ehsan: i just added a patch to bug 750232. when that has landed, we can try to build the new clang rev and see if it fixes the linker issue
# [16:16] <@ehsan> glandium: that sounds like something which takes more than a few hours to try :(
# [16:16] <@ted> glandium: possible, but you'd have to be really careful because that could eat into your PGO perf really quickly
# [16:16] <glandium> ehsan: but that's something which could help on the long term
# [16:16] <@ted> you lose the WPO bits
# [16:16] <smontagu> jaws: hmm, it's not impossible that my patches for bug 548206 and bug 562169 are having side-effects on this. let me test without them
# [16:17] <jaws> firebot: hello?
# [16:17] <firebot> jaws: hello
# [16:17] <@ehsan> edmorley: hmm, that tells me that we've been cutting it pretty close for some time :(
# [16:17] <edmorley> ehsan: yeah, not great :-(
# [16:17] * Joins: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com)
# [16:17] <@ehsan> glandium: true, it's definitely worth trying, but we also need a very short term solution
# [16:17] <@ehsan> which is the more urgent problem
# [16:18] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [16:18] <@ehsan> glandium: and ted's right, we need to measure and see how much it regresses perf benchmarks
# [16:18] <glandium> now, if only bug 710840 had been done...
# [16:19] * Joins: mdas (mdas@F2D29657.F60B0462.67AC9B1.IP)
# [16:19] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [16:19] <@ehsan> glandium: well, nobody really pays much attention to our other talos numbers any more... :/
# [16:20] <glandium> ehsan: it's not about paying attention, it's about having available data to help
# [16:20] <@ehsan> ted: do you know exactly where the code for opus landed? is it _all_ in libgkmedias?
# [16:20] <@ted> i have no idea
# [16:20] <@ted> hence why i asked
# [16:20] <@ehsan> glandium: edmorley has collected the data
# [16:21] <@ehsan> edmorley: just to make sure I don't get things wrong, can you give me the revision where we first saw the win pgo bustage?
# [16:21] * Quits: mib_lg9bn1 (Mibbit@76298C95.818589BB.979C46D2.IP) (Quit: http://www.mibbit.com ajax IRC Client)
# [16:21] * philor|away is now known as philor
# [16:21] <@ehsan> decoder: so, is that something that we have to fix on our end?
# [16:22] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [16:22] <mbrubeck> A chart of edmorley's data from bug 750661: https://dl.dropbox.com/u/2128410/linker-size.svg
# [16:22] * @ehsan closes inbound for now
# [16:22] <graememcc> edmorley, ted: just had a quick read of js/MachineResult php/getLogExcerpt and php/inc/TinderboxPrintFilter in the tbpl code. TinderboxPrint would work afaict for getting the figure into the TBPL summary
# [16:22] * Joins: panchot (panchot@2A266F1F.A09E2EBD.5039E5EA.IP)
# [16:22] <edmorley> graememcc: awesome, thank you (filed bug 750717 for adding)
# [16:23] * Parts: panchot (panchot@2A266F1F.A09E2EBD.5039E5EA.IP) (Leaving...)
# [16:23] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [16:24] * ehsan changes topic to 'mozilla-central and inbound are CLOSED, see bug 750661. || Next uplift for Fx15: 2012-06-05 || New/want to help? See irc://irc.mozilla.org/#introduction || http://krijnhoetmer.nl/irc-logs/'
# [16:25] <@ehsan> mbrubeck: so which revision corresponds to that significant jump?
# [16:25] <edmorley> ehsan: the full list of values is https://bugzilla.mozilla.org/attachment.cgi?id=619911
# [16:25] <edmorley> failures are listed there too
# [16:26] <edmorley> list is newest to old
# [16:26] <@ehsan> hmm
# [16:26] * Quits: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net) (Quit: Leaving.)
# [16:26] <@ehsan> so f8c388f622f1 seems to be the first one where there is a significant change
# [16:26] * Joins: jfkthame (jfkthame@15AA2040.A1C12133.9542EC20.IP)
# [16:26] * Joins: dseif (dseif@moz-57B23647.senecac.on.ca)
# [16:26] * Joins: wlach (wlach@moz-14AFFC14.vif.net)
# [16:26] * Quits: dwendorf (Mibbit@moz-6E3B32E8.lightspeed.mdsnwi.sbcglobal.net) (Quit: http://www.mibbit.com ajax IRC Client)
# [16:27] <edmorley> mbrubeck: thank you for charting that :-)
# [16:27] <edmorley> want me to start on some retriggers before that?
# [16:27] <edmorley> (f8c388f622f1)
# [16:27] <@ehsan> edmorley: no, wait
# [16:27] <edmorley> ok
# [16:28] <mbrubeck> ehsan, edmorley: Range for the significant jump is https://hg.mozilla.org/integration/mozilla-inbound/pushloghtml?fromchange=0e2658794e06&tochange=f8c388f622f1
# [16:28] <@ehsan> edmorley: I'm reading all of the patches between f8c388f622f1 and the last PGO build before it
# [16:28] <@ehsan> mbrubeck: yeah I'm reading those patches
# [16:29] * Quits: surkov (surkov@C1A6A50A.88321457.33A1AC3C.IP) (Quit: surkov)
# [16:30] <decoder> ehsan: that depends on how you see it. the llvm people say compiler-rt uses its own build system so it's not necessarily a bug... but even if you consider it a bug in the way they build compiler-rt, then it's unlikely that there will be a quick fix from their side. almost nobody will need this change most likely
# [16:30] <@ehsan> mbrubeck: edmorley: so all of the stuff there is either mobile/linux only, or very small patches :/
# [16:32] * Joins: beaufour (beaufour@18D5CC88.C7EE4FB2.ECED8BE3.IP)
# [16:32] <edmorley> hmmm :-(
# [16:33] * mbrubeck glances at "hg log --stat -r 0e2658794e06::f8c388f622f1"
# [16:33] * Quits: dseif (dseif@moz-57B23647.senecac.on.ca) (Connection reset by peer)
# [16:33] * @ehsan tries to think of more things to investigate
# [16:33] * Joins: dseif (dseif@moz-57B23647.senecac.on.ca)
# [16:33] <davidb> ehsan: is this about the talos regression?
# [16:33] <edmorley> davidb: win pgo OOM
# [16:33] <davidb> oof
# [16:33] <@ehsan> decoder: ok, this is unfortunate, but I guess kind of necessary
# [16:33] <edmorley> that's closed the tree again (bug 750661)
# [16:33] <@ehsan> decoder: I'll rebuild my clang when that patch lands
# [16:33] <decoder> ehsan: okay cool :) thx
# [16:34] <decoder> ehsan: id go with the old version on the linux slaves meanwhile, but it has been deleted on 32 bit
# [16:34] <decoder> so on 32 bit, i currently have no working clang on try
# [16:34] <@ehsan> decoder: well, I'm about to produce new clang binaries
# [16:34] <decoder> yea I know :)
# [16:34] <@ehsan> should be done in a couple of days I think
# [16:34] <decoder> cool
# [16:34] * Quits: bjacob (bjacob@moz-ADCA75DC.cpe.net.cable.rogers.com) (Ping timeout)
# [16:34] <@ehsan> decoder: (I'm working on this as a side project!)
# [16:35] <decoder> let me know if you need any help to push this :)
# [16:35] <decoder> im working on this as a main goal right now
# [16:35] <decoder> (asan builds)
# [16:35] <@ehsan> decoder: sure, but mostly I need to wait for builds in VMs etc :)
# [16:36] <@ehsan> decoder: it would be helpful if you could see if you can enable us use tooltool to download a clang package for our builds
# [16:36] <@ehsan> decoder: (ping jhford)
# [16:36] <@ehsan> cause I have not been able to look into that at all so far
# [16:39] * Quits: gcp (gpascutto@moz-D0E475EA.access.telenet.be) (Connection reset by peer)
# [16:39] * Joins: ddahl (ddahl@moz-976797D6.hsd1.il.comcast.net)
# [16:41] <@ehsan> ted: do you think we have large pieces of code which can be moved to gkmedias
# [16:41] <@ehsan> at least temporarily?
# [16:42] <jmaher> ajuma: I did the same command as you and it worked for me
# [16:42] <jmaher> wait, it is saving your profile to /data/local/tests instead of /mnt/sdcard/tests
# [16:42] <ajuma> jmaher: this tegra doesn't have an sdcard, so trying to write to /mnt/sdcard will be a problem
# [16:42] <jmaher> that hsould work, but it is a difference in our environments
# [16:42] * bear is now known as bear-afk
# [16:43] <jmaher> the permissions are different on /data/local vs /mnt/sdcard, that is the real difference
# [16:44] <decoder> ehsan: tooltool? and what package do you mean? instead of rpm building?
# [16:45] <ajuma> jmaher: let me make /data/local writeable and try again
# [16:45] <jmaher> ajuma: ok
# [16:45] <@ehsan> decoder: yeah, tooltool is a thing which we use for b2g which lets you download a tarball and extract it before the build, and then use it
# [16:45] <@ehsan> it would be an easier way for us to upgrade clang
# [16:45] <@ehsan> decoder: although we don't _have_ to use it for now I guess
# [16:45] * Joins: timdream (timdream@moz-4B9BEE5F.dynamic.kbtelecom.net)
# [16:46] * Joins: gcp (gpascutto@moz-D0E475EA.access.telenet.be)
# [16:46] <ajuma> jmaher: and reftest-sanity ran to completion :)
# [16:46] <jmaher> oh cool
# [16:47] <ajuma> jmaher: thanks for helping me get that sorted out
# [16:47] <jmaher> now, tweak the script to set browser.tabs.remote=false
# [16:47] * Joins: dveditz (dveditz@moz-5051E786.dhcp.cruzio.com)
# [16:47] * ChanServ sets mode: +o dveditz
# [16:47] * armenzg_buildduty is now known as armenzg_brb
# [16:47] <decoder> ehsan: i doubt it's a small thing to replace the rpm stuff with that :) besides we need the svn checkouts for clang/llvm/compiler-rt + our patches
# [16:47] <ajuma> jmaher: did that already
# [16:47] <@ehsan> decoder: ok, we can ignore that for now :/
# [16:48] <jmaher> ajuma: oh cool!
# [16:48] <ajuma> jmaher: i'll try border-radius now, since you saw problems there
# [16:48] <jmaher> yeah, that has blank tests and hanging ones
# [16:49] * Quits: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Quit: mkaply)
# [16:50] * Quits: aja (chatzilla@3846FB25.AEC64F0A.7880DB15.IP) (Input/output error)
# [16:50] * Joins: sgautherie (chatzilla@moz-D7B69DC4.fbxo.proxad.net)
# [16:50] * Quits: sgautherie (chatzilla@moz-D7B69DC4.fbxo.proxad.net) (Quit: ChatZilla 0.9.78.1 [SeaMonkey 1.5a/2007051508])
# [16:50] <ajuma> jmaher: it does hang on startup at the "do you want fennec to send your data…" prompt, and clicking yes/no gets in going again.
# [16:50] <ajuma> jmaher: then it ran 45 tests, and then crashed
# [16:51] * Joins: zwol (zack@moz-A5165AC6.lightspeed.sntcca.sbcglobal.net)
# [16:51] * Joins: bbondy2 (bbondy@moz-C9962B2.home.cgocable.net)
# [16:52] * Joins: Mardak (Mardak@moz-9D4EA709.sfo4.dsl.speakeasy.net)
# [16:52] * armenzg_brb is now known as armenzg
# [16:52] <edmorley> ehsan: the nightly inbound build has just failed too
# [16:52] <@ehsan> edmorley: I'm looking for things to rip out from libxul
# [16:53] * Quits: Mardak (Mardak@moz-9D4EA709.sfo4.dsl.speakeasy.net) (Quit: Mardak)
# [16:53] <jlebar> khuey|away, So...remote indexed db...
# [16:53] <@ted> ehsan: i have no idea :-/
# [16:53] <jfkthame> ehsan: could angle move to gkmedias, do you think?
# [16:53] <@ted> ehsan: it works best for things that are fairly standalone
# [16:53] <jmaher> ajuma: I have it run fine with that doorhanger, but we should disable that
# [16:53] <@ted> which is why all the media libs got moved
# [16:54] <@ted> jfkthame: i thought angle was already separate libs
# [16:54] <ajuma> jmaher: i don't know what to make of "PROCESS-CRASH | automation.py | application crashed (minidump found)" since this is after all the tests have passed
# [16:54] <ajuma> jmaher: and the crash dump it points to doesn't exist
# [16:55] <jmaher> ajuma: I recall there being a crash after the tests were done, during shutdown
# [16:55] <jmaher> ajuma: not sure why you don't have a dump available
# [16:56] * Quits: krit1 (krit@moz-3AE657AD.hsd1.ca.comcast.net) (Quit: Leaving.)
# [16:56] * Quits: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net) (Quit: jorendorff)
# [16:56] * Joins: garnacho (carlos@moz-EA2D0541.dyn.user.ono.com)
# [16:56] * Quits: mjschranz (mjschranz@F1451709.44D93D66.1139E686.IP) (Ping timeout)
# [16:56] <jfkthame> ted: ah, ok, looks like the main libs are separate, it's just gfx/angle/Makefile.in that says LIBXUL_LIBRARY=1 but that won't contribute much i guess
# [16:57] * Joins: mjschranz (mjschranz@F1451709.44D93D66.1139E686.IP)
# [16:57] * Quits: mike5w3c (MikeS@moz-2E518994.hfc.comcastbusiness.net) (Quit: mike5w3c)
# [16:57] * Quits: chrisccoulson (chr1s@CF62D2D6.FC921812.EE27BB8D.IP) (Quit: Ex-Chat)
# [16:58] <@ehsan> jfkthame: angle is already in gkmedias
# [16:58] <@ehsan> jfkthame: I was thinking about graphite
# [16:58] * Joins: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net)
# [16:58] <@ehsan> jfkthame: can we take it out/disable it?
# [16:59] <jfkthame> ehsan: that wouldn't be very popular - there are downstream gecko consumers relying on it
# [16:59] <ajuma> jmaher: do i need MINIDUMP_STACKWALK set in order to get a crash dump?
# [16:59] * Joins: jlebar|mac (~jlebarmac@moz-3F3A6302.dyn.columbia.edu)
# [16:59] <jfkthame> ehsan: but if we could move it to gkmedias i don't suppose that would matter much
# [16:59] <@ehsan> jfkthame: well, stopping the development on m-c isn't going to be too much popular either, we're talking about compromises
# [16:59] * Joins: chrisccoulson (chr1s@CF62D2D6.FC921812.EE27BB8D.IP)
# [16:59] * Quits: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com) (Input/output error)
# [16:59] <jmaher> ajuma: ahh, most likely
# [17:00] <edmorley> ehsan++
# [17:00] * Joins: martyn (martyn@moz-CABED5F2.range81-154.btcentralplus.com)
# [17:00] <@ehsan> jfkthame: the problem is that I don't quite know how to do that, but I do know how to disable it :)
# [17:00] <jfkthame> ehsan: i'll see if i can split it out without too much pain
# [17:00] <@ehsan> jfkthame: mostly one needs to know the APIs
# [17:01] <@ehsan> so that you can export them from gkmedias
# [17:01] <jfkthame> ehsan: yeah - i don't think that'll be too bad
# [17:01] <@ehsan> jfkthame: see https://bug709914.bugzilla.mozilla.org/attachment.cgi?id=581868 for an example of how we ripped out angle
# [17:01] * AutomatedTester|AFK is now known as AutomatedTester
# [17:01] <ajuma> jmaher: ok, trying again with that
# [17:01] <jfkthame> we did something similar with OTS iirc
# [17:01] <@ehsan> jfkthame: are you ok with me disabling it temporarily on inbound to see how much it helps?
# [17:02] <jfkthame> ehsan: yeah, i suppose so (sigh)
# [17:02] <@ted> ajuma: you need a couple of things
# [17:02] <@ted> a) to have run "make buildsymbols" in your objdir to prep the symbols
# [17:02] <@ted> b) to have MINIDUMP_STACKWALK set to a path to the minidump_stackwalk binary
# [17:02] <ajuma> ted: thanks, i had (b) but not (a).
# [17:03] <@ehsan> jfkthame: sorry about that, but PGO builds take 4 hours, and we need to know if this is going to help :(
# [17:03] <@ehsan> jfkthame: edmorley: filed bug https://bugzilla.mozilla.org/show_bug.cgi?id=750728
# [17:04] <jfkthame> ehsan: i know… it's ok, it hurts but we'd better try it
# [17:04] <@ehsan> yep
# [17:05] * ewong is now known as ewong|sleep
# [17:05] <@ehsan> ted: jfkthame: who wants to review a patch which disables graphite?
# [17:05] * Joins: joe_walk_ (joe_walker@moz-43315277.t-mobile.co.uk)
# [17:05] <@ted> whatev
# [17:05] * Quits: gcp (gpascutto@moz-D0E475EA.access.telenet.be) (Quit: Make a new plan, Stan!)
# [17:05] * Joins: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
# [17:05] <@ted> throw it at me or get a gfx peer or whatever
# [17:07] * Joins: gcp (gpascutto@moz-D0E475EA.access.telenet.be)
# [17:07] <@ehsan> ted: https://bugzilla.mozilla.org/attachment.cgi?id=619931&action=edit
# [17:07] * Quits: fabrice (fabrice@moz-94F028C6.hsd1.ca.comcast.net) (Quit: Leaving.)
# [17:07] * @ehsan starts to look for other stuff to rip out
# [17:08] * Quits: dseif (dseif@moz-57B23647.senecac.on.ca) (Input/output error)
# [17:08] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [17:08] * Quits: mjschranz (mjschranz@F1451709.44D93D66.1139E686.IP) (Client exited)
# [17:09] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [17:09] <espindola> Enn: ping
# [17:09] <Enn> espindola: hi
# [17:09] <espindola> so, I was under the impression that
# [17:09] <espindola> foo[i]->OnAssert...
# [17:09] <espindola> would keep a reference to foo[i] alive
# [17:09] * Quits: nattofriends (tsutsumi@moz-A369A791.warosu.org) (Ping timeout)
# [17:10] <espindola> your comment suggest that a race condition removing foo would make the ref count go to zero
# [17:10] <espindola> is that the case?
# [17:10] <espindola> removing the ith element from foo that is
# [17:10] <@ehsan> hmm
# [17:10] * Joins: nattofriends (tsutsumi@moz-A369A791.warosu.org)
# [17:11] * Joins: gmoro (guilherme@84484EC9.1F1093E8.D159334F.IP)
# [17:11] <@ehsan> does anyone know about libvpx
# [17:11] <@ehsan> ?
# [17:11] * Quits: brendan (brendaneic@moz-A286C218.hsd1.ca.comcast.net) (Quit: brendan)
# [17:11] * Joins: hessam (hessam@45968C53.9F7864B8.610F0005.IP)
# [17:12] * Joins: mjschranz (mjschranz@moz-57B23647.senecac.on.ca)
# [17:13] * Quits: mjschranz (mjschranz@moz-57B23647.senecac.on.ca) (Client exited)
# [17:13] * AutomatedTester is now known as AutomatedTester|AFK
# [17:13] <Enn> espindola: the observer has been removed from the list
# [17:13] <Enn> enn: see the second-to-last release
# [17:13] <espindola> Enn: well, we at least find it on the OnAssert call
# [17:14] <espindola> which suggest that that line is not holding a ref
# [17:14] <Enn> espindola: yes, it is found there and removed during that call.
# [17:14] * Joins: hub (hub@moz-E2FCA694.figuiere.net)
# [17:14] <Enn> perhaps a reference could be kept to the obsever before OnAssert is called
# [17:15] <espindola> oh well, I was expecting too much of nsCOMArray
# [17:15] <espindola> Enn: updating the bug
# [17:15] * armenzg is now known as armenzg_buildduty
# [17:16] <edmorley> ttaubert: xpcshell orange on fx-team
# [17:17] * Joins: sicking (chatzilla@moz-7F871C5C.hsd1.ca.comcast.net)
# [17:17] * Joins: mccr8 (mccr8@moz-5F92CD4B.dsl.dynamic.sonic.net)
# [17:17] <espindola> Enn: thanks!
# [17:17] * Quits: artur (artur@moz-6DFBD742.hsd1.vt.comcast.net) (Input/output error)
# [17:19] * Joins: kdcw (kdc@moz-F7413045.pk.shawcable.net)
# [17:20] * Quits: sicking (chatzilla@moz-7F871C5C.hsd1.ca.comcast.net) (Ping timeout)
# [17:21] * Joins: ericjung (Mibbit@moz-C6B344D.bstnma.fios.verizon.net)
# [17:21] * Joins: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP)
# [17:21] * Joins: automata (automata@8D23278A.C27CA109.16867D26.IP)
# [17:21] * Quits: msucan (mihai@7A4CDDE7.7A418818.B4CEF140.IP) (Quit: Leaving.)
# [17:23] * Joins: panchot (panchot@9B4DC816.AEDC213A.89E1DA2E.IP)
# [17:23] <ajuma> jmaher: looks like i'm going to rebuild with debug symbols (a simple "make buildsymbols" is crashing). i'll do that over lunch and try again this afternoon
# [17:24] <espindola> Enn: I think we need a lock anyway. Even if we create a nsCOMPtr temporary there would still be a race from the call to Count to the call to operator[].
# [17:24] <Enn> don't know
# [17:24] <jmaher> ajuma: cool; maybe we can figure this out in the afternoon
# [17:25] <@ted> ajuma: you need buildsymbols :-/
# [17:25] <@ted> debug symbols are enabled by default, but minidump_stackwalk needs them in a different format
# [17:25] * Joins: anant (anant@moz-BBE3ABD.mv.mozilla.com)
# [17:25] <@ted> what platform are you on?
# [17:25] <ajuma> ted: i'm on linux, but this is for android builds
# [17:25] <espindola> sewardj: yes, it would be really nice for firefox to be hellgrind clean :-(
# [17:26] * Parts: panchot (panchot@9B4DC816.AEDC213A.89E1DA2E.IP) (Leaving...)
# [17:26] <@ted> ajuma: hrm, that's not normal
# [17:26] <@ted> the tinderboxes run buildsymbols
# [17:26] <ajuma> ted: is adding "--enable-debug-symobls="-gdwarf-2"" to my mozconfig enough?
# [17:26] * Joins: Mardak (Mardak@moz-BBE3ABD.mv.mozilla.com)
# [17:27] <ajuma> (that's what the tinderboxes have)
# [17:27] * Joins: nhirata (nhirata.bu@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [17:27] <@ted> ajuma: that's the default
# [17:27] <sewardj> espindola: uh, is there some reason you say that at this point?
# [17:27] <@ted> sewardj: did you see we got the browser valgrind runs working again?
# [17:27] * Quits: joe_walk_ (joe_walker@moz-43315277.t-mobile.co.uk) (Input/output error)
# [17:27] <espindola> sewardj: the above discussion on bug 742455
# [17:28] <sewardj> ted: i saw that there was bugmail on it, but didn't see it had been fixed
# [17:28] <@ted> sewardj: they're running, they're just showing up red on tinderbox for silly automation reasons
# [17:28] <@ted> and they are showing a bunch of leaks, lots of which look like stdlib/gtk
# [17:28] <@ted> https://tbpl.mozilla.org/?noignore=1&jobname=Valgrind
# [17:29] * Joins: gfritzsche (gfritzsche@moz-255CC203.dynamic.qsc.de)
# [17:29] <@ted> but both the mozconfig and suppression files are in mozilla-central now
# [17:29] <@ted> so it should be pretty straightforward to get them in a good state
# [17:29] * jhford-work-away is now known as jhford-work
# [17:29] <@ted> https://tbpl.mozilla.org/php/getParsedLog.php?id=11356125&full=1&branch=mozilla-central
# [17:29] <@ted> was this morning's linux32 build
# [17:29] * Joins: msucan (mihai@7A4CDDE7.7A418818.B4CEF140.IP)
# [17:30] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [17:30] <gfritzsche> hi
# [17:30] <philor> s/silly automation reasons/because we said to run with a bunch of options that produce an exit code of 1 and then never did anything about suppressing them/
# [17:30] * khuey|away is now known as khuey
# [17:31] * Joins: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP)
# [17:31] <khuey> did the shit hit the fan yet?
# [17:31] <khuey> jlebar|mac: yeah ....
# [17:31] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [17:31] <sewardj> espindola: making Fx helgrind-clean is probably doable (the Chrome folks do that iirc) but would need some management-level support
# [17:31] <jlebar|mac> khuey: Well, b2g doesn't start up with OOP mozbrowser now, due to indexeddb. :-/
# [17:31] <sewardj> espindola: I have tried a couple of times in the past, but it is too much for one person
# [17:31] <jlebar|mac> It did a week or two ago, so I presume they just changed how they're using it in Gaia.
# [17:31] <@ted> philor: yes, silly automation reasons :-P
# [17:33] * Joins: evilpie (chatzilla@moz-451AEED8.pools.arcor-ip.net)
# [17:33] <sewardj> espindola: is there something you want me to helgrind for 742455 ? i have a hg-enabled build lying around here somewhere.
# [17:33] <espindola> sewardj: very similar logistic problems to switching to clang on OS X
# [17:34] <espindola> sewardj: no, I think the problem has been found
# [17:34] <espindola> helgrind would have found it way earlier
# [17:34] <espindola> but thanks
# [17:34] <evilpie> oh it's this time of the year again
# [17:34] <edmorley> evilpie: yup
# [17:35] <edmorley> s/year/quarter/
# [17:35] <@ted> we are going to be up shit's creek when we use up all 4GB of address space
# [17:35] <@ted> i guess "disable PGO" is the only option at that point
# [17:35] <khuey> we could stop adding code
# [17:35] <evilpie> would building on x64 work?
# [17:36] * Joins: joe_walker (joe_walker@moz-15405DDA.cable.virginmedia.com)
# [17:36] <@bsmedberg> josh, today is gfritzsche's first day. Do you have any particular plugin bugs you'd like him to take a look at?
# [17:36] <edmorley> ted: or 64bit linker, but that means 64bit builds right?
# [17:36] <philor> building on doesn't work, building for works
# [17:36] <evilpie> whatever we discussed that like a millions of times already
# [17:37] <evilpie> macosx throws up at uint64_t constant 1000000000000
# [17:37] <philor> so we have to keep ripping stuff out of libxul until Win32 dies
# [17:37] * Quits: smagnin (pike@moz-1816673.w90-57.abo.wanadoo.fr) (Quit: Quitte)
# [17:37] <@ted> edmorley: right
# [17:37] <sewardj> ted: what workload is it that runs on the V browser runs?
# [17:37] <evilpie> good reason to remove e4x, too
# [17:37] <@ted> MS doesn't provide a 64-bit linker that can produce 32-bit builds
# [17:37] <evilpie> can't we hack something together?
# [17:38] <@ted> sewardj: it runs profileserver.py, which is the same thing we profile our PGO builds on
# [17:38] <@ted> sewardj: loads sunspider and some CSS layout tests
# [17:38] <@ted> http://mxr.mozilla.org/mozilla-central/source/build/pgo/index.html?force=1#151
# [17:38] <sewardj> ted: ok ..
# [17:38] <sewardj> ted: I do complete mochitest runs from time to time
# [17:39] <@ted> that's obviously more thorough
# [17:39] <@ted> http://hg.mozilla.org/build/tools/file/default/scripts/valgrind/valgrind.sh
# [17:39] <@ted> is the full script that the valgrind builds execute
# [17:39] <sewardj> yeah, and you have to put the computer in a tank of cold water to stop it catching fire.
# [17:39] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [17:39] <@ted> hah
# [17:39] <@ted> well the plus side of running it on RelEng machines is that you don't have to put out the fire yourself
# [17:39] <philor> we talk about adding to the workload sometimes, then we remember that we've been too lazy to suppress the stuff we already need to grandfather in, without adding more
# [17:40] * joduinn-afk is now known as joduinn-coffee
# [17:40] <sewardj> ted: actually it's not so bad. latest V can do it in about 7 CPU hours and < 4GB mem
# [17:40] * Quits: xakz (XaMaD@moz-34FBE388.fbx.proxad.net) (Ping timeout)
# [17:40] <@ted> cool
# [17:40] * coop is now known as coop|afk
# [17:40] <sewardj> and it picks up new fluff all the time :)
# [17:40] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [17:41] <edmorley> philor: I've got the two new inbound oranges open ready to file if that helps
# [17:41] <philor> edmorley: that's a different help, I'm talking about the Valgrind suppression file
# [17:42] * @ted lunches
# [17:42] <edmorley> philor: yeah I meant generically / plus to make sure we weren't going to mid-air
# [17:43] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [17:44] * Quits: gerv (gerv@moz-8E68CF56.in-addr.arpa) (Ping timeout)
# [17:44] * Joins: panchot (panchot@9B4DC816.AEDC213A.89E1DA2E.IP)
# [17:45] * Joins: gozala (gozala@moz-D5BED6F9.dsl.dynamic.sonic.net)
# [17:45] * Parts: panchot (panchot@9B4DC816.AEDC213A.89E1DA2E.IP) (Leaving...)
# [17:45] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [17:46] * Joins: gerv (gerv@moz-8E68CF56.in-addr.arpa)
# [17:46] * Joins: jduell (jduell@moz-2D9EDA98.lightspeed.sntcca.sbcglobal.net)
# [17:47] * bear-afk is now known as bear
# [17:47] * Joins: bdahl (bdahl@moz-BBE3ABD.mv.mozilla.com)
# [17:48] * Quits: gerv (gerv@moz-8E68CF56.in-addr.arpa) (Ping timeout)
# [17:48] * Joins: xakz (XaMaD@moz-34FBE388.fbx.proxad.net)
# [17:48] * Joins: wesj (Instantbir@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [17:49] * jlund is now known as jlund|lunch
# [17:49] * Joins: Waldo (waldo@moz-BBE3ABD.mv.mozilla.com)
# [17:49] * catlee is now known as catlee-lunch
# [17:50] * armenzg_buildduty is now known as armenzg_lunch
# [17:51] * Quits: mccr8 (mccr8@moz-5F92CD4B.dsl.dynamic.sonic.net) (Quit: mccr8)
# [17:52] * Joins: gerv (gerv@moz-8E68CF56.in-addr.arpa)
# [17:52] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [17:53] * Joins: mayhemer__ (Miranda@B3D46202.F87A741B.F23860FD.IP)
# [17:53] * Quits: vikram360 (vikram360@195736B2.76910E54.2A068A5E.IP) (Ping timeout)
# [17:54] * Quits: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk) (Ping timeout)
# [17:54] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [17:54] * Joins: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk)
# [17:54] <@ehsan> ok, next victim: libpng (bug 750747)
# [17:55] * Joins: vikram360 (vikram360@195736B2.76910E54.2A068A5E.IP)
# [17:55] * Callek misread the bug as filed at first and thought you meant to "kill" libpng
# [17:56] <mbrubeck> back to GIF!
# [17:56] <Callek> was gonna scream, "but the web needs png"
# [17:56] * Joins: myk (Instantbir@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [17:56] <@ehsan> Callek: I'd like to kill it in fact, but people might be upset
# [18:00] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [18:01] <philor> fun, Profiling nightly died in IMAGE::BuildImage too
# [18:01] <philor> moves the window back a bit
# [18:02] * Mark_Capella is now known as Mark_Capella|inninout
# [18:02] <@ehsan> philor: do you have any new suspicions?
# [18:02] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [18:02] <philor> oh, profiling probably adds a bit, and it actually died yesterday too
# [18:02] <philor> so, no help there
# [18:03] * Fallen|away is now known as Fallen
# [18:03] * Joins: nihsanullah (naveed@moz-ECC7C78.bstnma.fios.verizon.net)
# [18:03] * Mark_Capella|inninout is now known as Mark_Capella
# [18:03] * Joins: jhammel (jhammel@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [18:04] * Quits: nihsanullah_ (naveed@moz-ECC7C78.bstnma.fios.verizon.net) (Ping timeout)
# [18:04] * @bz_sleep kicks bugzila
# [18:04] <philor> profiling's window is pretty much https://hg.mozilla.org/projects/profiling/pushloghtml?startID=673&endID=674
# [18:04] <@bz_sleep> y u not let me unsubscribe from bug?
# [18:04] * bz_sleep is now known as bz
# [18:05] * jimm is now known as jimm-lunch
# [18:05] <edmorley> philor: oh just seen your comment above
# [18:05] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [18:05] * mcote is now known as mcote|afk
# [18:05] <@ehsan> hmm
# [18:05] * @ehsan looks at the range
# [18:06] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [18:07] * Joins: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com)
# [18:08] <edmorley> ehsan: I'm going to push bug 750717 to inbound if that's ok; will at least mean we get the values in a more accessible manner
# [18:08] * Joins: jammink (textual@moz-BBE3ABD.mv.mozilla.com)
# [18:08] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
# [18:08] * Quits: gozala (gozala@moz-D5BED6F9.dsl.dynamic.sonic.net) (Quit: Leaving.)
# [18:09] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [18:09] * Joins: jgriffin (jgriffin@moz-4FBFA41D.hsd1.wa.comcast.net)
# [18:09] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [18:09] * Joins: faramarz (faramarz@moz-BBE3ABD.mv.mozilla.com)
# [18:10] * Joins: fabrice (fabrice@moz-BBE3ABD.mv.mozilla.com)
# [18:10] * Quits: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se) (Ping timeout)
# [18:11] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [18:11] <@ehsan> edmorley: yeah, sure
# [18:12] <fabrice> dougt: are you in MV today?
# [18:12] * Joins: joe_walker_ (joe_walker@moz-15405DDA.cable.virginmedia.com)
# [18:13] * Quits: pranavrc (pranavrc@F7F17312.87C59D23.C28326FD.IP) (Ping timeout)
# [18:13] * Quits: joe_walker (joe_walker@moz-15405DDA.cable.virginmedia.com) (Input/output error)
# [18:13] * Quits: joe_walker_ (joe_walker@moz-15405DDA.cable.virginmedia.com) (Quit: Leaving)
# [18:14] * Joins: joe_walker (joe_walker@moz-15405DDA.cable.virginmedia.com)
# [18:14] * davidb is now known as davidb|w3
# [18:16] * joduinn-coffee is now known as joduinn-commute
# [18:16] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [18:16] * Joins: pcwalton (pcwalton@moz-7B0110AD.mv.mozilla.com)
# [18:17] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [18:20] <edmorley> I hate line endings
# [18:20] <zwol> remove them all! write all your code on one line
# [18:21] <evilpie> so nobody bothered reporting it to mozilla? http://arstechnica.com/business/news/2012/05/firefox-security-add-in-exposes-users-web-browsing-history.ars
# [18:21] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [18:22] * Joins: addoffs (blah@83A2A2D8.ED7DE540.79496794.IP)
# [18:22] * Joins: rillian_lime (rillian@moz-DBEBB2B4.bc.hsia.telus.net)
# [18:23] * rillian_lime is now known as rillian
# [18:24] <@bz> evilpie: bug filed?
# [18:25] <zzzzz> the addon mentioned has reviews pointing out that's it spyware/malware - why is it even on AMO still ?
# [18:25] * Joins: mike5w3c (MikeS@454A1A4F.FF5089F.1771B9DE.IP)
# [18:25] <@bz> because no one filed a bug about it?
# [18:25] <evilpie> i am doing it right now
# [18:25] * Joins: knelson (Adium@moz-A2960B48.hsd1.ca.comcast.net)
# [18:25] * Parts: knelson (Adium@moz-A2960B48.hsd1.ca.comcast.net)
# [18:26] <gcp> https://addons.mozilla.org/en-US/firefox/addon/download-youtube-videos/
# [18:26] <bholley> edmorley: is the current PGO issue likely to be another week-long closure? I was hoping to land compartment-per-global tomorrow morning
# [18:26] <gcp> also has many comments claiming its malware
# [18:27] <addoffs> http://nakedsecurity.sophos.com/2012/05/01/privacy-concern-showip-firefox-add-on/
# [18:28] <addoffs> "The full URL of every webpage visited is sent to the Germany-based ip2info.org website, using unencrypted connections." "The ip2info.org website itself appears to be very new, having only been registered a month ago."
# [18:28] * Quits: ericjung (Mibbit@moz-C6B344D.bstnma.fios.verizon.net) (Quit: http://www.mibbit.com ajax IRC Client)
# [18:28] <addoffs> time for ban hammer
# [18:28] <edmorley> bholley: it's not looking great tbh, we've already made the switch to msvc2010 so that's out of the window; ehsan is looking at what else can be split out of libxul, there may also be some odd deadcode wins around (though some like the old parser & rdf still need more work before they can happen)
# [18:28] <@bz> erm
# [18:28] <Callek> edmorley: line ending fubar https://hg.mozilla.org/integration/mozilla-inbound/rev/e1f1d4f79b2d
# [18:28] <Callek> :-P
# [18:28] <@bz> we have pgo red again? :(
# [18:29] <NeilAway> espindola: the count changing during enumeration sounds more likely as the cause of the crash
# [18:29] <@bz> who added code?
# [18:29] <@ehsan> bz: me you and everybody else :(
# [18:29] <@ehsan> bholley: you can help by getting libjpg out of libxul :)
# [18:29] <edmorley> bholley: the other thing is trying to expedite bug 709480, but sadly it's been a P3 for a while
# [18:29] <bholley> ehsan: I am not the person to do that
# [18:30] <@ehsan> well until we get people to help with this, all trees will remain closed
# [18:30] <evilpie> bz: Blocklisting or Security?
# [18:30] <edmorley> bholley: although to be fair, it's non-trivial since I believe will require re-imaging a bunch of stuff
# [18:30] <@bz> evilpie: let's try Blocklisting, and cc jorge?
# [18:30] * Joins: rniwa (rniwa@454A1A4F.FF5089F.1771B9DE.IP)
# [18:31] <evilpie> who?
# [18:31] <bholley> ehsan: I just mean thank linkage issues and libxul aren't my area of expertise, and the fact that I'm an imagelib peer doesn't make it more so
# [18:31] <edmorley> Callek: yeah, hence my "I hate line endings" above :-) (though it would seem I actually got rid of the crlfs, not the other way around, so guess it's fine)
# [18:32] <@ehsan> bholley: ok, well let's hope that I will do a good enough job then
# [18:32] * Quits: evilpie (chatzilla@moz-451AEED8.pools.arcor-ip.net) (Quit: ChatZilla 0.9.88.2 [Firefox 15.0a1/20120501030504])
# [18:32] <bholley> ehsan: rs=bholley if that helps :-)
# [18:32] * Joins: gfritzsche_ (gfritzsche@moz-255CC203.dynamic.qsc.de)
# [18:33] * Joins: krit (krit@moz-1FC1932F.adobe.com)
# [18:33] <bholley> ehsan: but I think it's probably best to have a build person do the review
# [18:33] * Joins: evilpie (chatzilla@moz-451AEED8.pools.arcor-ip.net)
# [18:33] <@bz> evilpie: lemme look up the bug id
# [18:33] <evilpie> bug 750761
# [18:33] <@ehsan> bholley: the build system part is easy, the hard part is figuring out how imglib uses libjpeg etc
# [18:34] <@ehsan> same for libpng
# [18:34] <@ehsan> which is what I'm struggling with right now
# [18:34] <bholley> ehsan: how can I help?
# [18:34] <bholley> ehsan: it should be entirely in decoders/nsJPEGDecoder
# [18:34] <@bz> evilpie: jorge@mozilla.com
# [18:34] <bholley> ehsan: and encoders/nsJPEGEncoder
# [18:34] * Joins: priya (Adium@moz-BBE3ABD.mv.mozilla.com)
# [18:34] <evilpie> bz: cc'd
# [18:35] * Quits: brambles (brambles@4CBAB088.F3076E90.1822ACA6.IP) (Quit: leaving)
# [18:35] * davidb|w3 is now known as davidb
# [18:35] <gcp> what's the policy about add-ons that bundle toolbars and such?
# [18:35] <@ehsan> bholley: basically I don't know what stuff in image/ links to these libs, and which functions are used, etc
# [18:35] <@ehsan> and if there are other dependencies
# [18:35] <@ehsan> I'm just trying things out blindly here
# [18:35] <@ehsan> and so far nothing seems to work :(
# [18:35] * Joins: brambles (brambles@4CBAB088.F3076E90.1822ACA6.IP)
# [18:36] <bholley> ehsan: are we trying to prune functions we don't use? Or offload the whole library?
# [18:36] * Quits: gfritzsche (gfritzsche@moz-255CC203.dynamic.qsc.de) (Connection reset by peer)
# [18:36] * gfritzsche_ is now known as gfritzsche
# [18:36] * Quits: automata (automata@8D23278A.C27CA109.16867D26.IP) (Ping timeout)
# [18:36] <@ehsan> bholley: offload the whole lib into libgkmedias
# [18:37] * Joins: Mossop (mossop@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [18:37] <bholley> ehsan: ok. Why does it matter which functions are used?
# [18:37] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
# [18:37] <@ehsan> nm, I need to focus on getting this right
# [18:37] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Quit: sworkman)
# [18:38] * Joins: rniwa_ (rniwa@60A74940.D6CCE4AE.77834EAA.IP)
# [18:38] * Joins: pranavrc (pranavrc@57CC0C5E.1137763F.C28326FD.IP)
# [18:38] <bholley> ehsan: ok, ping me if you need anything
# [18:39] * Quits: rniwa (rniwa@454A1A4F.FF5089F.1771B9DE.IP) (Ping timeout)
# [18:39] * rniwa_ is now known as rniwa
# [18:39] * Joins: supreet (quassel@9E3665C8.B91F5BC2.DF11F364.IP)
# [18:39] * Quits: joe_walker (joe_walker@moz-15405DDA.cable.virginmedia.com) (Quit: Leaving)
# [18:39] * Quits: supreet (quassel@9E3665C8.B91F5BC2.DF11F364.IP) (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
# [18:41] * Joins: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP)
# [18:41] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
# [18:41] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
# [18:41] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Input/output error)
# [18:41] * Quits: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP) (Input/output error)
# [18:41] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
# [18:41] * Joins: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP)
# [18:42] * Quits: bholley (anonymous@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Quit: bholley)
# [18:42] <@bz> dbaron: ping
# [18:43] * Joins: cviecco_ (cviecco@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [18:43] * coop|afk is now known as coop|mtg
# [18:44] * armenzg_lunch is now known as armenzg
# [18:44] * Joins: jandem (jandem@66C76B89.FB8EABAE.DF9376EA.IP)
# [18:44] * jimm-lunch is now known as jimm
# [18:44] * armenzg is now known as armenzg_buildduty
# [18:45] * Quits: fabrice (fabrice@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving.)
# [18:45] * Joins: fabrice1 (fabrice@moz-BBE3ABD.mv.mozilla.com)
# [18:45] * Joins: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [18:45] <nemo> so. mozilla guys.
# [18:45] <nemo> a lot of people I've mentioned it to, were overjoyed by:
# [18:45] <nemo> https://bitbucket.org/satyr/ubiquity/downloads (tip.xpi)
# [18:45] <nemo> https://ubiquity.mozilla.com/hg/ubiquity-firefox/
# [18:45] * Joins: naveed (Harrower@moz-ECC7C78.bstnma.fios.verizon.net)
# [18:45] <nemo> which I ran into from https://addons.mozilla.org/en-US/firefox/addon/mozilla-labs-ubiquity/reviews/311761/
# [18:45] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
# [18:46] <nemo> I'm curious - if Satyr is maintaining ubiquity, why is it not on AMO?
# [18:46] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [18:46] * Joins: mib_s2xoyl (Mibbit@76298C95.818589BB.979C46D2.IP)
# [18:46] <nemo> it does seem to work in latest firefox nightly
# [18:46] * Quits: mib_s2xoyl (Mibbit@76298C95.818589BB.979C46D2.IP) (Quit: http://www.mibbit.com ajax IRC Client)
# [18:47] * Joins: TheLink (TheLink@moz-2C115FCB.pools.arcor-ip.net)
# [18:47] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [18:48] * Joins: jet (junglecode@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [18:49] * Quits: evilpie (chatzilla@moz-451AEED8.pools.arcor-ip.net) (Input/output error)
# [18:49] * Joins: zzzzz_ (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [18:49] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [18:49] * zzzzz_ is now known as zzzzz
# [18:49] * Quits: gcp (gpascutto@moz-D0E475EA.access.telenet.be) (Connection reset by peer)
# [18:49] * Joins: jdm (jdm@moz-749CB30C.dsl.scarlet.be)
# [18:51] * Joins: gozala (gozala@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [18:51] * Joins: fabrice (fabrice@moz-BBE3ABD.mv.mozilla.com)
# [18:51] * Quits: fabrice1 (fabrice@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving.)
# [18:52] * Joins: sicking (chatzilla@454A1A4F.FF5089F.1771B9DE.IP)
# [18:53] * Joins: dholbert_ (dholbert@A01051A9.6F669779.3D1CA460.IP)
# [18:53] * Joins: bonnie (bbsurender@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [18:53] * Quits: dholbert_ (dholbert@A01051A9.6F669779.3D1CA460.IP) (Connection reset by peer)
# [18:54] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Ping timeout)
# [18:54] * Joins: azakai (alon@moz-BBE3ABD.mv.mozilla.com)
# [18:55] * Joins: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net)
# [18:55] * philor is now known as philor|away
# [18:55] <zzzzz> pfft, fastest WONTFIX ever - bug 750761 :(
# [18:56] * bear is now known as bear-afk
# [18:56] * Quits: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Ping timeout)
# [18:57] * Quits: Boriss (FlyingToas@moz-62AAA429.hsd1.ca.comcast.net) (Quit: Boriss)
# [18:57] * Joins: juanb (jbecerra@moz-BBE3ABD.mv.mozilla.com)
# [18:58] <beltzner> nah, I've filed a bug as WONTFIX before
# [18:58] * Quits: Waldo (waldo@moz-BBE3ABD.mv.mozilla.com) (Quit: ChatZilla 0.9.87-4.1450hg.fc15 [XULRunner 11.0/20120313114635])
# [18:58] * Joins: gcp (gpascutto@moz-D0E475EA.access.telenet.be)
# [18:59] * Joins: stevee (Miranda@moz-BEBDF855.cable.virginmedia.com)
# [18:59] * Joins: yuan (ywang@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [18:59] <nigelb> beltzner: To refer people to?
# [19:00] <nigelb> Because that's a brilliant idea :D
# [19:00] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [19:01] * Quits: cviecco_ (cviecco@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Input/output error)
# [19:01] * joduinn-commute is now known as joduinn-mtg
# [19:02] * catlee-lunch is now known as catlee
# [19:02] * Joins: icaaq (Adium@moz-718D9655.cust.bredbandsbolaget.se)
# [19:02] <taras> Bas: around?
# [19:02] * jlund|lunch is now known as jlund
# [19:03] * Joins: smooney (smooney@moz-BBE3ABD.mv.mozilla.com)
# [19:03] <Bas> taras: I am
# [19:03] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [19:03] <@dbaron> bz, pong... or was it the email I just replied to?
# [19:04] * Quits: fabrice (fabrice@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving.)
# [19:04] * rnewman is now known as rnewman|afk
# [19:06] * Joins: joe_walker (joe_walker@moz-15405DDA.cable.virginmedia.com)
# [19:06] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [19:07] * Quits: krit (krit@moz-1FC1932F.adobe.com) (Quit: Leaving.)
# [19:07] <@bz> dbaron: it was the email
# [19:07] <@bz> dbaron: well, I had another thought, but I gave up on it
# [19:07] * Quits: azakai (alon@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [19:07] <@bz> dbaron: or maybe not
# [19:07] <@bz> dbaron: how do you feel about adding a camelCase version of method_ to the prop list?
# [19:08] * Joins: akeybl (akeybl@moz-7B0110AD.mv.mozilla.com)
# [19:08] <@dbaron> bz, fine
# [19:08] <@bz> dbaron: (and presumably the pref string, if any, would go there too)
# [19:08] <@bz> dbaron: enum order would largely show up in devtools
# [19:08] <@bz> dbaron: I would think
# [19:08] <@bz> dbaron: unless those re-sort stuff anyway
# [19:08] <@dbaron> bz, though I wonder if we could avoid the non-camelCase version?
# [19:09] * fabrice|afk is now known as fabrice
# [19:09] <@bz> well
# [19:09] <jdm> hmm, has the platform meeting happened yet?
# [19:09] <@bz> so I can avoid the camelcase version too
# [19:09] <@bz> by the simple expedient of tossing in some python, not just cpp
# [19:10] <@dbaron> bz, but if we aren't implementing CSS2Properties in C++ anymore, I wonder if the things that use domProp now would all just be ok with camelCase?
# [19:10] * @bz thinks
# [19:10] <mccr8> jdm: platform meeting is in an hour, at 11am MV time
# [19:11] * Joins: bholley_mobile (bholley_mo@FF9F28.56A6027.D3EC63AB.IP)
# [19:11] <@bz> so it would be camelCase for the non-domprop-prefixed stuff
# [19:11] <jdm> heh, european timezones are weeeeeird
# [19:11] <@bz> and InterCaps for the prefixed stuff
# [19:11] <@bz> (because that's what JS sees)
# [19:11] <@bz> and also, if we want to be able to pref this thing off when it first lands....
# [19:11] <@bz> I'm tempted to just do the python+cpp thing for now
# [19:11] <@dbaron> bz, ah, well, that might require having both around for a bit
# [19:11] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [19:11] <@dbaron> bz, or the python thing is ok
# [19:12] <@bz> and when we drop the xpidl bits, if ever, we can revisit
# [19:12] * Parts: jandem (jandem@66C76B89.FB8EABAE.DF9376EA.IP)
# [19:12] * Quits: twi (Adium@moz-5390D98D.cust.dsl.vodafone.it) (Client exited)
# [19:13] * Joins: Mook_as (mook@moz-1FCC0032.activestate.com)
# [19:14] * Joins: azakai (alon@moz-BBE3ABD.mv.mozilla.com)
# [19:14] <@bz> dbaron: so I'm going to add a prefname to the proplist
# [19:15] <@bz> dbaron: what marker do you prefer for "not pref controlled"?
# [19:15] * Joins: krit (krit@moz-1FC1932F.adobe.com)
# [19:15] <@bz> dbaron: NULL and empty string are the obvious options
# [19:16] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [19:16] * Joins: twi (Adium@moz-5390D98D.cust.dsl.vodafone.it)
# [19:17] * Joins: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net)
# [19:18] * Quits: cers (textual@D5CF850E.567E557.FE16CD6C.IP) (Ping timeout)
# [19:19] * Quits: Asa (asa@D13E5E3F.A1EC5031.204CA821.IP) (Ping timeout)
# [19:19] * Joins: cpeterson (cpeterson@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [19:20] * Joins: bent (chatzilla@moz-C3562645.hsd1.ca.comcast.net)
# [19:21] * Quits: hub (hub@moz-E2FCA694.figuiere.net) (Ping timeout)
# [19:21] * Quits: dcamp (dave@moz-8EBEC133.hsd1.wa.comcast.net) (Quit: Leaving...)
# [19:21] <jviereck> dholbert: hi, do you have a minute?
# [19:22] * Quits: micahg (micahg@moz-6E8FD6B4.c3-0.arm-ubr1.chi-arm.il.cable.rcn.com) (Quit: Ex-Chat)
# [19:22] * Joins: bholley_mobile| (bholley_mo@FF9F28.56A6027.D3EC63AB.IP)
# [19:22] * Quits: bholley_mobile (bholley_mo@FF9F28.56A6027.D3EC63AB.IP) (Ping timeout)
# [19:23] * Joins: squib (squib@moz-B01B5D55.dhcp.mdsn.wi.charter.com)
# [19:24] * Quits: tchevalier (Instantbir@moz-BA7C7754.w90-0.abo.wanadoo.fr) (Ping timeout)
# [19:24] * Quits: naveed (Harrower@moz-ECC7C78.bstnma.fios.verizon.net) (Ping timeout)
# [19:24] * Quits: jlebar|mac (~jlebarmac@moz-3F3A6302.dyn.columbia.edu) (Quit: jlebar|mac)
# [19:25] <romaxa> bsmedberg: ping
# [19:26] * Quits: merinui (merinui@moz-61C7235E.osk2.eonet.ne.jp) (Quit: Leaving...)
# [19:27] * Joins: naveed (Harrower@moz-ECC7C78.bstnma.fios.verizon.net)
# [19:28] * Joins: gakiwate (gautam@3AD1CBD5.809D2ACB.314AAB94.IP)
# [19:28] * philor|away is now known as philor
# [19:29] * Joins: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se)
# [19:29] * Quits: addoffs (blah@83A2A2D8.ED7DE540.79496794.IP) (Quit: )
# [19:31] * Joins: cers (textual@D5CF850E.567E557.FE16CD6C.IP)
# [19:31] * Joins: Boriss (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [19:32] * Joins: jlebar|mac (~jlebarmac@moz-3F3A6302.dyn.columbia.edu)
# [19:33] * Quits: jgriffin (jgriffin@moz-4FBFA41D.hsd1.wa.comcast.net) (Ping timeout)
# [19:34] * Joins: jgriffin (jgriffin@moz-4FBFA41D.hsd1.wa.comcast.net)
# [19:34] * Joins: jgriffin_ (jgriffin@moz-4FBFA41D.hsd1.wa.comcast.net)
# [19:34] * Quits: mayhemer__ (Miranda@B3D46202.F87A741B.F23860FD.IP) (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
# [19:34] * Joins: knelson (Adium@moz-A2960B48.hsd1.ca.comcast.net)
# [19:35] * Joins: shank (matt@moz-AECDC97A.hsd1.wa.comcast.net)
# [19:36] * Quits: jgriffin (jgriffin@moz-4FBFA41D.hsd1.wa.comcast.net) (Ping timeout)
# [19:36] * jgriffin_ is now known as jgriffin
# [19:37] * Quits: Mitch (chatzilla@moz-8D46E76F.sbr802.nsw.optusnet.com.au) (Ping timeout)
# [19:37] * Parts: knelson (Adium@moz-A2960B48.hsd1.ca.comcast.net)
# [19:37] * glob is now known as glob|away
# [19:37] * armenzg_buildduty is now known as armenzg_brb
# [19:38] * Joins: Mitch (chatzilla@moz-8D46E76F.sbr802.nsw.optusnet.com.au)
# [19:40] * Joins: bnicholson (bnicholson@moz-BBE3ABD.mv.mozilla.com)
# [19:41] * Quits: zzzzz (chatzilla@moz-FC5E07A0.dhcp.embarqhsd.net) (Quit: ChatZilla 0.9.88-rdmsoft [XULRunner 1.9.0.17/2009122204])
# [19:41] * Quits: bholley_mobile| (bholley_mo@FF9F28.56A6027.D3EC63AB.IP) (Ping timeout)
# [19:42] <tn> marco, not other than what comment 39 said
# [19:42] * Quits: vikash (vikash@C7D00EB1.38F0169.5D9ABA9F.IP) (Quit: Leaving)
# [19:42] * Joins: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP)
# [19:42] * Joins: hub (hub@1063758.68078EAA.B97E6502.IP)
# [19:43] <@ehsan> joe: this is my libpng patch: http://pastebin.mozilla.org/1610364
# [19:43] * Quits: sfink (chatzilla@moz-7B7651CB.dsl.pltn13.sbcglobal.net) (Ping timeout)
# [19:43] <@ehsan> joe: and this is what got angle out of libxul: https://bug709914.bugzilla.mozilla.org/attachment.cgi?id=581868
# [19:43] <@ehsan> joe: the latter is probably a better example
# [19:43] * bear-afk is now known as bear
# [19:44] <Bas> taras: I'm about to go for dinner :)
# [19:44] <Bas> So if you wanted to ask me something you'll have to do it now or in a while.
# [19:46] * Joins: bholley_mobile (bholley_mo@FF9F28.56A6027.D3EC63AB.IP)
# [19:46] * Joins: brendan (brendaneic@AA16AC92.EF8E4D8F.43362C16.IP)
# [19:46] <@ehsan> glandium: ping
# [19:47] * Joins: kaie (kaie@moz-55C6BE11.dip.t-dialin.net)
# [19:47] * Parts: gakiwate (gautam@3AD1CBD5.809D2ACB.314AAB94.IP) (Leaving)
# [19:47] * Quits: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se) (Ping timeout)
# [19:48] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [19:49] * Joins: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se)
# [19:49] * Joins: cjones (cjones@moz-45913895.socal.res.rr.com)
# [19:49] <jfkthame> ehsan: yuck, do we really need so much listed in symbols.def.in? seems like most of them would be purely internal to libpng, no?
# [19:49] <joe> jfkthame: that's the exposed api
# [19:49] <joe> it is not a small library
# [19:49] <@ehsan> jfkthame: I have no clue really, I just exported everything I could find
# [19:50] * Joins: knelson (Adium@moz-A2960B48.hsd1.ca.comcast.net)
# [19:51] * Joins: anky (anky@C0485CBE.72B2334B.A3D1B221.IP)
# [19:51] * Quits: jviereck (Adium@moz-C5992AE4.dip.t-dialin.net) (Quit: Leaving.)
# [19:52] * Joins: billm (billm@moz-BBE3ABD.mv.mozilla.com)
# [19:53] * Quits: anky (anky@C0485CBE.72B2334B.A3D1B221.IP) (Ping timeout)
# [19:53] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [19:53] * Quits: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Ping timeout)
# [19:54] * Joins: anky (anky@C0485CBE.72B2334B.A3D1B221.IP)
# [19:56] * Quits: mdas (mdas@F2D29657.F60B0462.67AC9B1.IP) (Quit: mdas)
# [19:56] * Quits: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net) (Quit: Eaten by grue.)
# [19:57] <taras> Bas: crap..missed you. i wanted to suggest that we profile stuff with borders/gradients implemented as no-ops
# [19:57] <taras> to get a baseline perf estimate
# [19:57] * Quits: timdream (timdream@moz-4B9BEE5F.dynamic.kbtelecom.net) (Quit: timdream)
# [19:58] * Parts: knelson (Adium@moz-A2960B48.hsd1.ca.comcast.net)
# [19:58] * Joins: dwendorf (Mibbit@moz-4314DA11.gandermountain.com)
# [19:59] * Quits: brendan (brendaneic@AA16AC92.EF8E4D8F.43362C16.IP) (Quit: brendan)
# [19:59] * Joins: terrence (terrence@moz-BBE3ABD.mv.mozilla.com)
# [19:59] <joe> taras: that is a really good idea
# [19:59] <joe> also, turn off -moz-element
# [20:00] <joe> because iirc it uses svg filters on some of it
# [20:00] <joe> er, I guess, turn off svg filters
# [20:00] <jlebar|mac> How do I get trace-malloc to complain about leaks?
# [20:00] <@ehsan> joe: taras: a better idea would be to remove them from the theme
# [20:00] * Quits: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP) (Quit: Computer has gone to sleep.)
# [20:00] <jlebar|mac> There's some environment variable...
# [20:00] <@ehsan> cause there's a ton of stuff happening for those
# [20:01] <@ehsan> so disabling them is not as easy as returning early from one function or something
# [20:01] <joe> well
# [20:01] <joe> if we just want to remove painting
# [20:01] <joe> we can probably do that pretty easily
# [20:01] <@bz> no painting would be a huge perf win
# [20:01] <@bz> but a bit of a usability regression
# [20:01] <joe> for sighted people maybe
# [20:01] <@bz> right
# [20:01] <@ehsan> but what about the layout stuff which happens
# [20:01] * Quits: jimb (user@moz-F4EC06CC.hsd1.or.comcast.net) (Ping timeout)
# [20:02] <@ehsan> painting is not the only expensive thing there
# [20:03] * Quits: wlach (wlach@moz-14AFFC14.vif.net) (Quit: Leaving)
# [20:05] * Joins: sworkman_ (sworkman@moz-BBE3ABD.mv.mozilla.com)
# [20:05] * Joins: bent_ (chatzilla@moz-C3562645.hsd1.ca.comcast.net)
# [20:05] * Quits: bent (chatzilla@moz-C3562645.hsd1.ca.comcast.net) (Ping timeout)
# [20:05] * Quits: bent_ (chatzilla@moz-C3562645.hsd1.ca.comcast.net) (Quit: Reconnecting…)
# [20:05] * Quits: nhirata (nhirata.bu@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: nhirata)
# [20:05] * Joins: bent (chatzilla@moz-C3562645.hsd1.ca.comcast.net)
# [20:05] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [20:05] * sworkman_ is now known as sworkman
# [20:06] * Joins: nhirata (nhirata.bu@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [20:06] * Joins: catalinb (ethereal@48049090.B4F52EC7.5BC07656.IP)
# [20:06] * armenzg_brb is now known as armenzg
# [20:07] * Quits: bholley_mobile (bholley_mo@FF9F28.56A6027.D3EC63AB.IP) (Ping timeout)
# [20:07] * Joins: bholley_mobile (bholley_mo@FF9F28.56A6027.D3EC63AB.IP)
# [20:07] <mccr8> jlebar: https://developer.mozilla.org/en/Debugging_memory_leaks#BloatView ?
# [20:07] * Joins: andreasn (andreasn@moz-436FB3D1.a336.priv.bahnhof.se)
# [20:07] * Quits: martyn (martyn@moz-CABED5F2.range81-154.btcentralplus.com) (Quit: Ex-Chat)
# [20:07] <mccr8> jlebar: or more specifically, the "How to turn on refcnt/memory logging" section.
# [20:08] <jlebar|mac> Aha, XPCOM_MEM_LEAK_LOG. Thanks!
# [20:09] * Joins: zzzzz (chatzilla@moz-107FCDBA.hfc.comcastbusiness.net)
# [20:09] <taras> ehsan: taking out panting will make other stuff easier to focus on
# [20:10] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [20:10] <@ehsan> taras: I'm not sure what you plan to do to make those other stuff faster
# [20:11] * Joins: Sander (chatzilla@moz-B871F4D3.direct-adsl.nl)
# [20:11] <mccr8> jlebar|mac: yeah, I can never remember it. I always have to look up that page.
# [20:11] * Quits: jacek (jacek@moz-5D707D3B.psi.wroc.pl) (Quit: Konversation terminated!)
# [20:11] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [20:12] <jlebar|mac> How do I do a search on all AMO add-ons? I've seen this special MXR floating around...
# [20:12] * Joins: maikmerten (maikmerten@moz-343F0059.dynamic.qsc.de)
# [20:12] * Parts: alfredo- (Adium@88F51059.F3BBB17D.144F44FA.IP)
# [20:14] <mbrubeck> jlebar|mac: File a bug to get access to https://mxr.mozilla.org/addons/ if you don't already have it.
# [20:14] * Quits: gcp (gpascutto@moz-D0E475EA.access.telenet.be) (Connection reset by peer)
# [20:14] <jlebar|mac> Aha.
# [20:14] <jlebar|mac> Thanks.
# [20:14] <taras> ehsan: i have lots of plans
# [20:15] <@ehsan> taras: can you tell us about them? :)
# [20:15] * Quits: anky (anky@C0485CBE.72B2334B.A3D1B221.IP) (Ping timeout)
# [20:15] * Joins: evilpie_ (evilpie@moz-451AEED8.pools.arcor-ip.net)
# [20:16] * Joins: anky (anky@815BE04E.A23A2186.A3D1B221.IP)
# [20:18] * Joins: dao (dao@moz-95A844B3.superkabel.de)
# [20:18] * Joins: gcp (gpascutto@moz-D0E475EA.access.telenet.be)
# [20:18] * Joins: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP)
# [20:18] <mbrubeck> edmorley, ehsan: Could we perhaps make m-c APPROVAL REQUIRED with blanket approval for patches that are NPOTB on Windows?
# [20:19] <@ehsan> mbrubeck: that's fine by me
# [20:20] <@ehsan> mbrubeck: but I don't want anything to land on inbound for now
# [20:20] * @bz loves how the section on Chrome's bundling stuff got whitewashed out of Wikipedia
# [20:20] <mbrubeck> yeah, inbound is already a goner. :(
# [20:20] <edmorley> mbrubeck: sure, sounds like a plan
# [20:20] * mbrubeck heads to https://wiki.mozilla.org/index.php?title=Tree_Rules&action=edit
# [20:20] * Quits: jhopkins (jhopkins@moz-216F9986.tb.shawcable.net) (Input/output error)
# [20:21] <glandium> ehsan: pong
# [20:21] <Bas> taras: That's exactly what I'm working on by coincidence :)
# [20:21] <edmorley> mbrubeck, ehsan: what do we think: NPOTB, mobile-only, test-only, non-windows only?
# [20:21] <Bas> taras: I'll have some results for you tomorrow.
# [20:22] <@ehsan> glandium: I know that you're on vacation, but I would really appreciate if you can take a quick glance at http://pastebin.mozilla.org/1610364 and tell me if I'm doing anything crazy
# [20:22] <@ehsan> edmorley: yeah
# [20:22] <Bas> taras: Rather, I was doing to implement gradients as solid fills and borders as rectangles.
# [20:22] <Bas> So as not to let the decrease in pixels to fill affect the result.
# [20:22] <@ehsan> edmorley: although I'll back out any patches which change a single line in windows' libxul without mercy ;)
# [20:22] <edmorley> ehsan: likewise :-)
# [20:23] <@ehsan> cool
# [20:23] <edmorley> ehsan, mbrubeck: are we going to a=<insert own reason>, or a=sheriff_on_irc_who_vetted etc?
# [20:23] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
# [20:23] <@ehsan> edmorley: either way is fine by me
# [20:24] <mbrubeck> edmorley: For now I went with a=reason on https://wiki.mozilla.org/Tree_Rules
# [20:24] <glandium> ehsan: I'm not on "vacation", it's just may day :)
# [20:24] <edmorley> let's try self-serve then, we can always switch if necessary
# [20:24] <jesup> joduinn-mtg: Much better than previous meetings! :-)
# [20:24] <mbrubeck> but I'm open to suggestions.
# [20:24] <@ehsan> heh, I see
# [20:24] <edmorley> mbrubeck: sounds good :-)
# [20:25] * joduinn-mtg tips hat of thanks to jesup for mic-debugging! :-)
# [20:25] <taras> Bas: would be cool if we could land this prefed off
# [20:25] <taras> even if temporarily
# [20:26] <taras> maybe on ux branch?
# [20:26] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
# [20:26] * Joins: wlach (wlach@moz-14AFFC14.vif.net)
# [20:26] <Bas> taras: Hmm, let me see what the final patch looks like and you can decide if you want it.
# [20:26] * Joins: mdas (mdas@50AE257C.D30B51A1.412CF160.IP)
# [20:27] <glandium> ehsan: I don't think you need to export all those symbols. We only use a few
# [20:27] <glandium> ehsan: also, why are the LOCAL_INCLUDES needed?
# [20:27] * Joins: Optimizer (Mibbit@6E78482A.3686F45E.89AC0F27.IP)
# [20:27] * mbrubeck changes topic to 'inbound CLOSED, m-c APPROVAL REQUIRED: bug 750661 || Next uplift for Fx15: 2012-06-05 || New/want to help? See irc://irc.mozilla.org/#introduction || http://krijnhoetmer.nl/irc-logs/'
# [20:27] <Optimizer> jaws: I cannot seem to see the task specific icons in the window 7 jumplist
# [20:29] * Quits: jdm (jdm@moz-749CB30C.dsl.scarlet.be) (Client exited)
# [20:29] * Quits: @bz (bzbarsky@moz-69B5879F.bstnma.fios.verizon.net) (Quit: Leaving)
# [20:31] * Joins: cadecairos (cadecairos@F1451709.44D93D66.1139E686.IP)
# [20:31] * Quits: dwendorf (Mibbit@moz-4314DA11.gandermountain.com) (Quit: http://www.mibbit.com ajax IRC Client)
# [20:31] * Joins: jgilbert (jgilbert@moz-BBE3ABD.mv.mozilla.com)
# [20:31] * Joins: smaug (chatzilla@moz-AD13B184.elisa-mobile.fi)
# [20:31] * ChanServ sets mode: +o smaug
# [20:32] * Quits: pranavrc (pranavrc@57CC0C5E.1137763F.C28326FD.IP) (Quit: Ping timeout: ∞)
# [20:33] * Quits: Stan (Stan@moz-9E1FC00B.dip.t-dialin.net) (Ping timeout)
# [20:33] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Input/output error)
# [20:33] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
# [20:33] <@ehsan> glandium: the compiler wouldn't find png.h without the LOCAL_INCLUDES
# [20:33] <zzzzz> Optimizer: are you running something other than the Aero them on Win7, i.e. Windowblinds or some such ? New Icons looks OK for me
# [20:34] <@ehsan> glandium: and I don't know which one of those symbols are used :(
# [20:34] <zzzzz> s/them/theme
# [20:34] <glandium> ehsan: note, the easy way to know what symbols you need to export is to not export them, get an error, and export those that the linker says are missing :)
# [20:34] <Optimizer> zzzzz: nopes, windows 7 Areo, 64 build
# [20:34] * Quits: bholley_mobile (bholley_mo@FF9F28.56A6027.D3EC63AB.IP) (Ping timeout)
# [20:34] <Optimizer> default theme, just changes wallpaper
# [20:34] <zzzzz> ok,
# [20:34] <@ehsan> glandium: neat!
# [20:34] <@ehsan> I'll try that
# [20:34] * Joins: Lucas (Lucas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [20:34] * Quits: davidillsley (chatzilla@moz-6FDCDDC6.range86-160.btcentralplus.com) (Client exited)
# [20:34] * Joins: Stan (Stan@moz-9E1FC00B.dip.t-dialin.net)
# [20:35] * Joins: bholley (anonymous@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [20:35] <glandium> ehsan: did you omit a part of the patch? because i don't see why png.h would be suddenly missing, expect if you move it, which i don't see in that patch
# [20:35] <Optimizer> do I have to clear some cache or something like that ?
# [20:35] * Joins: Asa (asa@moz-BBE3ABD.mv.mozilla.com)
# [20:35] <glandium> s/expect/except/
# [20:35] * Joins: cadecairos_ (cadecairos@moz-57B23647.senecac.on.ca)
# [20:36] * Joins: anky_ (anky@2935FD87.5111A3CC.A3D1B221.IP)
# [20:36] <@ehsan> glandium: no, it was puzzling to me as well, and I can't explain it
# [20:36] * Quits: cadecairos (cadecairos@F1451709.44D93D66.1139E686.IP) (Ping timeout)
# [20:36] * Quits: anky (anky@815BE04E.A23A2186.A3D1B221.IP) (Ping timeout)
# [20:36] <jaws> Optimizer: close firefox, go to task manager, kill explorer.exe, then restart explorer.exe and firefox.exe and you should be good to go
# [20:37] * Quits: azakai (alon@moz-BBE3ABD.mv.mozilla.com) (Connection reset by peer)
# [20:37] <jaws> killing explorer.exe with firefox closed should force a rebuild of the icon cache
# [20:37] * Joins: jhopkins (jhopkins@moz-216F9986.tb.shawcable.net)
# [20:37] * Joins: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net)
# [20:37] * Joins: bholley_mobile (bholley_mo@40D65E62.26DA4C22.B3F72630.IP)
# [20:37] * Joins: azakai (alon@moz-BBE3ABD.mv.mozilla.com)
# [20:38] <glandium> ehsan: it might be worth checking why, although it's just fine to add these LOCAL_INCLUDES
# [20:38] <Optimizer> I have restarted the computer also 2-3 times from yesterday
# [20:38] <Optimizer> jaws: ^
# [20:38] * Quits: cadecairos_ (cadecairos@moz-57B23647.senecac.on.ca) (Input/output error)
# [20:39] <Optimizer> can it be some windows cache problem ?
# [20:39] * Quits: ddahl (ddahl@moz-976797D6.hsd1.il.comcast.net) (Ping timeout)
# [20:39] <jaws> Optimizer: can you go to about:config and set browser.taskbar.lists.refreshInSeconds to something like 10?
# [20:39] <glandium> ehsan: the last chunk is going to break "native" png builds, since you remove PNG_LIBS
# [20:39] <glandium> unconditionally
# [20:39] * Joins: bholley_mobile| (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [20:39] * Joins: dcamp (dave@9E727688.17C2F9C.163DC5C6.IP)
# [20:39] * Quits: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net) (Ping timeout)
# [20:40] * Quits: bholley_mobile (bholley_mo@40D65E62.26DA4C22.B3F72630.IP) (Ping timeout)
# [20:40] <Optimizer> jaws: done
# [20:40] <glandium> ehsan: one possibility, for png.h, is it not being exported to dist/include anymore. Which is actually a good thing.
# [20:40] <@ehsan> glandium: ok, I'll make that depend on the MOZ_NATIVE thingy
# [20:40] <jaws> Optimizer: cool, then in 10 seconds it should rebuild those icons
# [20:40] * Joins: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net)
# [20:40] * Joins: tchevalier (Instantbir@moz-BA7C7754.w90-0.abo.wanadoo.fr)
# [20:41] * Joins: sfink (chatzilla@moz-BBE3ABD.mv.mozilla.com)
# [20:41] * Quits: Mitch (chatzilla@moz-8D46E76F.sbr802.nsw.optusnet.com.au) (Ping timeout)
# [20:41] * Joins: Mitch_ (chatzilla@moz-8D46E76F.sbr802.nsw.optusnet.com.au)
# [20:41] <Optimizer> jaws: not happening, infact my gmail icon is still showing 8 unread mails, if it would have updated, then there would have been 0 unread
# [20:41] * Mitch_ is now known as Mitch
# [20:41] <@ehsan> glandium: I'm honestly fine with those LOCAL_INCLUDES
# [20:41] <Optimizer> I think its some windows problem
# [20:42] * Mitch is now known as IRCMonkey55958
# [20:42] <jaws> Optimizer: do you have visual studio installed?
# [20:42] <zzzzz> Optimizer: have you tried Firefox safe-mode to make sure something like AdBlock or NoScript is not causing it, or maybe HWA ?
# [20:43] <Optimizer> HWA : Off, no adblock, noscript, Yes visual studio is installed, jaws , zzzzz
# [20:43] <jaws> Optimizer: if you have Visual Studio installed, you can open up the executable and look at the embedded icons to see if they are in your version
# [20:44] * Quits: maikmerten (maikmerten@moz-343F0059.dynamic.qsc.de) (Client exited)
# [20:44] <bholley> ehsan: are we looking to generate some negative diffs? I can see how much of caps I can rip out if need be
# [20:44] * Joins: Jake (Jake@moz-3284655B.resnet.drexel.edu)
# [20:44] * Quits: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net) (Ping timeout)
# [20:44] * bholley has no idea of the scale of what we need
# [20:45] <bholley> but the caps ripout is coming anyway
# [20:45] <edmorley> bholley: I think at this point, everything helps
# [20:45] * Joins: automata (automata@8D23278A.C27CA109.16867D26.IP)
# [20:45] <bholley> edmorley: ok, I'll see what i can do
# [20:45] <bholley> edmorley: also, FWIW, compartment-per-global is a negative diff
# [20:46] <Optimizer> jaws: all the three icons are there
# [20:46] <bholley> edmorley: but it's a major behavior change
# [20:46] * Joins: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net)
# [20:46] <bholley> edmorley: should we hold off on it?
# [20:46] <edmorley> bholley: if only because reopening the tree is one thing; giving us enough headroom for the next few weeks until the builders are changed is another
# [20:47] <bholley> edmorley: not sure what you mean
# [20:48] * Joins: ddahl (ddahl@moz-976797D6.hsd1.il.comcast.net)
# [20:48] <edmorley> bholley: I meant in reference to ripping as much out as possible
# [20:49] <bholley> edmorley: gotcha
# [20:49] <firebot> Check-in: http://hg.mozilla.org/mozilla-central/rev/9645b00119e4 - Margaret Leibovic - Bug 750130 - Telemetry doorhanger appears then disappears when onLocationChange is fired during startup. r=bnicholson a=android-only
# [20:49] <firebot> http://hg.mozilla.org/mozilla-central/rev/0d06d9f8e4f6 - Nick Alexander - Bug 713524 - Batch bookmark inserts. r=rnewman, a=android-only
# [20:49] <firebot> http://hg.mozilla.org/mozilla-central/rev/9c0d42c3ae49 - Matt Brubeck - Bug 746703 - Set targetSdkVersion to 14 in AndroidManifest.xml to disable legacy menu button [r=blassey, a=android-only]
# [20:50] <firebot> http://hg.mozilla.org/mozilla-central/rev/62a472b617de - Nick Alexander - Bug 743153 - Use ContentUris.parseId instead of RepoUtils.getAndroidIdFromUri. r=rnewman, a=android-only
# [20:50] <edmorley> bholley: as to CPG, I'll defer to ehsan and co
# [20:50] <firebot> http://hg.mozilla.org/mozilla-central/rev/74108dd3201d - Richard Newman - Bug 749857 - Don't upload history records without URI or visits. r=liuche, a=android-only
# [20:50] * jwir3 is now known as jwir3|lunch
# [20:51] * Joins: panchot (panchot@9B4DC816.AEDC213A.89E1DA2E.IP)
# [20:51] * edmorley is now known as edmorley|afk
# [20:51] * Quits: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP) (Max SendQ exceeded)
# [20:51] * Joins: carljm (carljm@moz-5C873C64.threepines.org)
# [20:51] * Quits: Optimizer (Mibbit@6E78482A.3686F45E.89AC0F27.IP) (Quit: http://www.mibbit.com ajax IRC Client)
# [20:52] * Joins: davidillsley (chatzilla@moz-6FDCDDC6.range86-160.btcentralplus.com)
# [20:52] * Quits: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net) (Quit: Eaten by grue.)
# [20:52] * Joins: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP)
# [20:53] <fabrice> smaug: It looks like my activity event constructor is not exposed properly (I get "MozActivityEvent is not a constructor" from a web page). Can you check what I'm missing in my patch?
# [20:53] * Joins: dwendorf (chatzilla@moz-6E3B32E8.lightspeed.mdsnwi.sbcglobal.net)
# [20:54] * Joins: Optimizer (Mibbit@6E78482A.3686F45E.89AC0F27.IP)
# [20:54] * Joins: Yoric (Yoric@moz-920DB13B.fbx.proxad.net)
# [20:54] * AutomatedTester|AFK is now known as AutomatedTester
# [20:55] <@smaug> fabrice: are you testing on the page
# [20:55] <@smaug> or console
# [20:55] * zpao is now known as zpao|detached
# [20:55] <fabrice> smaug: in a page
# [20:55] <@smaug> fabrice: did you add the same thing to nsDOMclassInfo as what other events do
# [20:55] <taras> ehsan: see dev.platform for details
# [20:56] * Quits: JeroenDeDauw (j@moz-95881CDD.dip.t-dialin.net) (Quit: Leaving.)
# [20:56] <fabrice> smaug: I think so, but I'll double check
# [20:56] <espindola> ted: sorry, any thoughts on 555727?
# [20:56] * armenzg is now known as armenzg_brb
# [20:56] * Joins: JeroenDeDauw (j@moz-95881CDD.dip.t-dialin.net)
# [20:57] * Joins: ferongr (ferongr@9461F8D8.3BB4FAB7.F5160715.IP)
# [20:57] * mcote|afk is now known as mcote
# [21:00] * Parts: @ted (luser@moz-2854A494.scr.east.verizon.net) (Ex-Chat)
# [21:00] * Joins: ted (luser@moz-2854A494.scr.east.verizon.net)
# [21:00] * ChanServ sets mode: +o ted
# [21:00] <@ted> oops
# [21:01] <@ted> espindola: probably the right patch, i just haven't had time to go through my queue
# [21:01] * Quits: @smaug (chatzilla@moz-AD13B184.elisa-mobile.fi) (Input/output error)
# [21:01] <smontagu> oof
# [21:01] <smontagu> what is the point of splitting a patch in two if they don't build separately?
# [21:01] * Joins: smaug (chatzilla@moz-AD13B184.elisa-mobile.fi)
# [21:01] * ChanServ sets mode: +o smaug
# [21:01] * Quits: JeroenDeDauw (j@moz-95881CDD.dip.t-dialin.net) (Input/output error)
# [21:02] <jfkthame> ehsan: would you be ok with me landing bug 750728 (the second patch, that is) on inbound?
# [21:03] * Quits: panchot (panchot@9B4DC816.AEDC213A.89E1DA2E.IP) (Client exited)
# [21:03] <froydnj> smontagu: sometimes it's easier to see the logic that way
# [21:04] <smontagu> I suppose
# [21:04] <jfkthame> though in that case it's helpful to fold them before landing, for the sake of future bisection
# [21:04] * armenzg_brb is now known as armenzg_buildduty
# [21:05] * Joins: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net)
# [21:05] <froydnj> ideally the patch submitter would call out non-buildability and explicitly state that the patches will be folded
# [21:05] <jfkthame> and then remember to do so :)
# [21:05] <@ehsan> jfkthame: has it passed try tests?
# [21:06] <froydnj> jfkthame: indeed :)
# [21:06] <jfkthame> ehsan: i didn't ask for tests, just builds - i figured any problems would show up as link failure
# [21:06] <@ehsan> bholley: any code removal will help
# [21:06] <@ehsan> jfkthame: ok then yes, please go ahead
# [21:06] * zpao|detached is now known as zpao
# [21:06] * Joins: JeroenDeDauw (j@moz-95881CDD.dip.t-dialin.net)
# [21:07] * Quits: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP) (Ping timeout)
# [21:08] * juanb is now known as juanb|lunch
# [21:08] * Quits: sfink (chatzilla@moz-BBE3ABD.mv.mozilla.com) (Client exited)
# [21:08] <glandium> ehsan: can you check whether png.h ends up in dist/include? if it doesn't, then I'm happy
# [21:08] * Quits: zwol (zack@moz-A5165AC6.lightspeed.sntcca.sbcglobal.net) (Quit: Ex-Chat)
# [21:09] * Quits: Cwiiis (cwiiis@88F51059.F3BBB17D.144F44FA.IP) (Quit: Leaving)
# [21:09] <@ehsan> glandium: it _is_ in dist/include
# [21:09] * Quits: kdcw (kdc@moz-F7413045.pk.shawcable.net) (Quit: HydraIRC -> http://www.hydrairc.com <- s0 d4Mn l33t |t'z 5c4rY!)
# [21:09] <@ehsan> glandium: although I don't know why anything in my patch should have any effect on that
# [21:10] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [21:10] <glandium> ehsan: some variables have some side effects
# [21:10] <@ehsan> hmm, yeah I bet
# [21:10] <glandium> ehsan: if it is in dist/include, then i don't see why you need the LOCAL_INCLUDES...
# [21:10] * Quits: JeroenDeDauw (j@moz-95881CDD.dip.t-dialin.net) (Quit: Leaving.)
# [21:10] <@ehsan> glandium: well, the build would fail without those
# [21:10] <@ehsan> wait, let me check somethign
# [21:11] <@ehsan> ok I wanted to check whether png.h is there on mac as well
# [21:11] * Quits: Yoric (Yoric@moz-920DB13B.fbx.proxad.net) (Input/output error)
# [21:11] <@ehsan> and it is
# [21:11] <kaie> glandium, re deb#670882 how can I install iceweasel? I have access to an ubuntu 11.10 system, but "apt-get install iceweasel" tells me "selecting firefox instead of iceweasel". is it necessary to uninstall firefox?
# [21:11] * Joins: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP)
# [21:11] * Quits: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP) (Ping timeout)
# [21:11] <glandium> kaie: ubuntu doesn't have iceweasel
# [21:11] * Joins: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP)
# [21:11] <kaie> glandium, could I manually install the deb, if you pointed me to a direct link? i386
# [21:12] <kaie> or do I really need a deb install?
# [21:12] <kaie> debian system
# [21:12] <glandium> kaie: it might work. check http://packages.debian.org
# [21:12] <glandium> kaie: you need iceweasel, xulrunner-10.0 and libmozjs10d
# [21:12] <kaie> glandium, ok, which distribution should I try?
# [21:13] * Joins: JeroenDeDauw (j@moz-95881CDD.dip.t-dialin.net)
# [21:13] <glandium> kaie: unstable
# [21:13] <kaie> ok
# [21:13] <glandium> kaie: you'll also need libnspr4-0d and libnss3-1d
# [21:13] <taras> Bas: can you turn your email from yesterday into a bug?
# [21:13] <AryehGregor> dbaron, would you prefer that I move the lines to the bottom even though they're going to be the only ones at the bottom? I can do that if you like.
# [21:14] <kaie> glandium, hmm, as I already have nspr/nss system libs those, maybe I should go for a VM.
# [21:14] * Quits: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se) (Ping timeout)
# [21:15] <glandium> kaie: might be safer
# [21:15] * khuey is now known as khuey|away
# [21:16] <Bas> taras: You mean a kind of 'general tracking' bug.
# [21:16] <taras> Bas: details up to you
# [21:17] <@dbaron> AryehGregor, the content ones are also there
# [21:17] * Joins: bz (bzbarsky@moz-69B5879F.bstnma.fios.verizon.net)
# [21:17] * ChanServ sets mode: +o bz
# [21:17] <AryehGregor> dbaron, okay, I'll move them.
# [21:17] <@dbaron> AryehGregor, I'd slightly prefer it, but you should go ahead either way
# [21:17] <jesup> ted: ping
# [21:17] <AryehGregor> dbaron, okay.
# [21:17] <@bz> taras: ping
# [21:17] * Quits: Sander (chatzilla@moz-B871F4D3.direct-adsl.nl) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
# [21:17] * Joins: roc (chatzilla@C0ACF8B.5E1E9EEA.613E47D1.IP)
# [21:17] * ChanServ sets mode: +o roc
# [21:18] <taras> bz: pong
# [21:18] <@ted> jesup: pong
# [21:18] * Joins: sfink (chatzilla@moz-BBE3ABD.mv.mozilla.com)
# [21:18] <jesup> So, did you see the OpenSLES question via email?
# [21:18] <jesup> What am I doing wrong? :-)
# [21:19] * Joins: johanc (chatzilla@moz-D8A1AA43.bredband.comhem.se)
# [21:20] <@smaug> josh: do we have any data how often that java stuff is used in the pages?
# [21:20] <jesup> ted: if it helps, here's the patch I'm using: http://pastebin.mozilla.org/1610639
# [21:20] <jesup> And the try failure: https://ftp.mozilla.org/pub/mozilla.org/firefox/try-builds/rjesup@wgate.com-50cb35ee3907/try-android/try-android-bm09-try1-build3307.txt.gz
# [21:20] <@ted> jesup: i saw but didn't digest it
# [21:20] <@ted> i don't actually know what SLES is
# [21:20] <jesup> Basically, I just need to know how to add an include directory for Android
# [21:21] <jesup> it's already in the NDK
# [21:21] <@bz> taras: so profiling tabstrip js stuff
# [21:21] <@bz> taras: and xpconnect
# [21:21] <AryehGregor> Bah, more closed tree! :(
# [21:21] * AryehGregor goes to bed, will try again tomorrow
# [21:21] <@bz> taras: I'd be happy to take a look; toss me str?
# [21:21] <jesup> ted: and this didn't work: + CPPFLAGS="-isystem $android_platform/usr/include -I$android_platform/usr/include/SLES $CPPFLAGS"
# [21:21] <josh> smaug: No telemetry, if that's what you're asking. From what I've seen over the years and my discussions with Oracle I think we'll be fine, and if not it isn't hard to back out.
# [21:22] <taras> bz: so far i've been holding ctrl+tab while profiler is on
# [21:22] <@bz> taras: ok
# [21:22] <@bz> taras: so just load two about:blank pages and go?
# [21:22] <taras> bz: oh that'sthe other detail
# [21:22] * Quits: anky_ (anky@2935FD87.5111A3CC.A3D1B221.IP) (Client exited)
# [21:22] <@ted> jesup: well, things are trying to include it via SLES/OpenSLES.h
# [21:22] <taras> bz: empty tabs seem to switch differently
# [21:22] <taras> bz: i go to news.google.com click on ~5-8 links to open in new tabs
# [21:23] <@smaug> josh: it is just that based on experience we may get the feedback too late
# [21:23] <taras> then use noscript or disable content js
# [21:23] <josh> smaug: I think the heaviest users are probably enterprise sites and we're not likely to get telemetry from them. They're also rarely on the fast upgrade path.
# [21:23] <taras> bz: to rule out event loop stuffing
# [21:23] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
# [21:23] <@smaug> josh: could we add a warning
# [21:23] <josh> smaug: I really doubt a warning is going to cause anyone to actually change something.
# [21:24] * Parts: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP)
# [21:24] * Joins: mjschranz (mjschranz@moz-57B23647.senecac.on.ca)
# [21:24] <@smaug> josh: have that even in FF14
# [21:24] <@smaug> josh: warning may hint someone to bug us
# [21:24] <josh> This functionality has been deprecated by Oracle and us for a while now, it's marked clearly in the Oracale docs.
# [21:24] <@smaug> josh: docs aren't enough
# [21:24] <josh> neither is a console warning
# [21:25] <@smaug> (no one read docs :p )
# [21:25] <@bz> taras: gotcha
# [21:25] <@smaug> that is possible
# [21:25] <@bz> taras: lemme pull up a profile
# [21:25] <josh> We're going to take a risk whenever we do this, there is no way around that.
# [21:25] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
# [21:25] <@smaug> josh: adding a warning to FF14 would be trivial
# [21:25] <josh> you mean Aurora?
# [21:25] * edmorley|afk is now known as edmorley
# [21:25] <josh> I'd be happy to do that.
# [21:25] * Quits: mjschranz (mjschranz@moz-57B23647.senecac.on.ca) (Client exited)
# [21:26] <@smaug> josh: yes
# [21:27] <josh> smaug: I can have a patch up today, it'll only take a few minutes to write
# [21:27] <jesup> ted: duh. :-) I was being distracted at the time. So, why does it not find it? Perhaps it's the 'android-9'? (I'm unclear how platform/android-N is resolved) $ find -name SLES --> ./platforms/android-9/arch-arm/usr/include/SLES
# [21:27] * Quits: catalinb (ethereal@48049090.B4F52EC7.5BC07656.IP) (Ping timeout)
# [21:27] <@ted> --with-android-platform=DIR
# [21:27] <@ted> location of platform dir, default NDK/build/platforms/android-5/arch-arm
# [21:27] <@smaug> josh: nsIDocument has very handy WarnOnceAbout
# [21:27] <jesup> That's for NDK r5c
# [21:28] <josh> smaug: Thanks, I didin't know about that. I'll request review from you.
# [21:28] <jesup> Aha. And, so, could I change that to 9? Without breaking 1/2 the world?
# [21:28] * Joins: Matt_ (matthewger@807A3136.9C68C162.6BF415CA.IP)
# [21:28] * Joins: armenzg__ (armenzg@moz-ED3249A4.dmz.releng.scl3.mozilla.com)
# [21:28] <mwu> jesup: SLES? I thought we were still supporting old androids without that
# [21:28] * jesup is still getting feet wet with Android
# [21:28] <Matt_> there's no way to intercept a request and redirect it somewhere else, right?
# [21:28] <@ted> jesup: no, our default android version target is 5
# [21:28] * Matt_ just wants to confirm before he does the heavy lifting with a traceable channel
# [21:29] <@ted> --with-android-version=VER
# [21:29] <@ted> but that changes our minimum target requirements
# [21:29] <jesup> mwu: the Webrtc code wants OpenSLES; I'll see if it has alternatives
# [21:29] <jesup> I think OpenSLES is important for low-latency audio, IIRC
# [21:29] * Quits: armenzg_buildduty (armenzg@moz-ED3249A4.dmz.releng.scl3.mozilla.com) (Ping timeout)
# [21:29] <mwu> worry not, android audio latency sucks no matter what api you use
# [21:29] <jesup> older android sucked for latency. (Still sucks some, not as much though IIRC)
# [21:30] <jesup> mwu: ++
# [21:30] <mwu> sles skips you past java
# [21:30] * armenzg__ is now known as armenzg_buildduty
# [21:30] <mwu> that's about it
# [21:30] <Matt_> ideally I'd change the request URL in http-on-modify-request but I know that's not allowed
# [21:30] <Matt_> unless something changed since I last looked into this
# [21:30] * Parts: Jake (Jake@moz-3284655B.resnet.drexel.edu)
# [21:30] * zpao is now known as zpao|detached
# [21:30] * Quits: vikram360 (vikram360@195736B2.76910E54.2A068A5E.IP) (Ping timeout)
# [21:30] <mwu> but it's still implemented using audiotrack which gets sent to audioflinger
# [21:30] <@ted> we're targeting eclair apparently?
# [21:30] <@ted> 2.0
# [21:30] <@ted> 9 is gingerbread (2.3)
# [21:31] <mwu> I was assuming 2.2 but uh, I dunno
# [21:31] * Joins: vikram360 (vikram360@D64D8E31.8274BA73.2A068A5E.IP)
# [21:31] * Quits: rniwa (rniwa@60A74940.D6CCE4AE.77834EAA.IP) (Ping timeout)
# [21:32] * Quits: sicking (chatzilla@454A1A4F.FF5089F.1771B9DE.IP) (Ping timeout)
# [21:32] <jesup> mwu: audioflinger - evil ;-)
# [21:32] * Quits: mike5w3c (MikeS@454A1A4F.FF5089F.1771B9DE.IP) (Ping timeout)
# [21:32] <mwu> yup.
# [21:33] <jesup> SO yeah, WebRTC only supports OpenSLES for audio
# [21:33] * Joins: Ally_ (textual@moz-26DFFD6.range109-152.btcentralplus.com)
# [21:33] * Joins: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [21:33] * Quits: bholley_mobile| (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Connection reset by peer)
# [21:33] * Quits: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Client exited)
# [21:33] <mwu> worst case, you can dynamically load opensles
# [21:33] <jesup> Which means 9 (2.3?) minimum
# [21:33] <jesup> mwu: any docs you know of? Or just google?
# [21:34] <romaxa> bsmedberg: ping
# [21:34] * Joins: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [21:34] <mwu> jesup: for what?
# [21:34] <@bz> taras: so this is interesting
# [21:34] <@bz> taras: I nuked all the painting from the profile
# [21:34] <cpeterson> ted, Fennec's android:minSdkVersion is 8 (Froyo), not 5.
# [21:34] <@bz> taras: then about 10% is mac system event loop crud
# [21:34] <firebot> Check-in: http://hg.mozilla.org/mozilla-central/rev/a53189537295 - Jonathan Griffin - Bug 750843 - Update marionette for ICS emulator, a=testonly, npotb
# [21:34] * Quits: vikram360 (vikram360@D64D8E31.8274BA73.2A068A5E.IP) (Ping timeout)
# [21:35] <@bz> some CC and refresh driver stuff
# [21:35] <jesup> mwu: on dynamically loading it (on older versions). Or were you meaning "just don't do audio if it doesn't load"? Ok for the hearing-impaired, not for most others. ;-)
# [21:35] <mwu> oh yeah, I meant just being mute on android 2.2
# [21:35] * Joins: jdm (jdm@moz-749CB30C.dsl.scarlet.be)
# [21:35] <mwu> unless you want to use audiotrack
# [21:35] <jesup> cpeterson: 8 is closer to 9 than 5! But still not there.
# [21:35] <@bz> taras: tons of event firing and event handlers
# [21:35] <mwu> I just meant dynamically loading it so you don't have to link against it
# [21:36] * zpao|detached is now known as zpao
# [21:36] <mwu> so it doesn't crash on 2.2
# [21:36] * Quits: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net) (Quit: jorendorff)
# [21:36] <@bz> taras: a bunch of restyling (why???)
# [21:36] * Joins: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net)
# [21:37] <@ted> cpeterson: oh, huh
# [21:37] <jesup> mwu: Yeah, if we have to we can do that so we have one package for all versions (though in <2.3 systems there will be a lot of not-so-usable code in the package...)
# [21:37] <@ted> wonder why configure lies
# [21:37] <mwu> jesup: btw, is opensles a requirement on all platforms?
# [21:37] <@ted> jesup: might be able to bump to 9 from 8
# [21:37] <mwu> or is it a convenient backend that's there
# [21:37] <@ted> if it makes life easier
# [21:37] <jesup> mwu: required for WebRTC on Android
# [21:37] <mwu> ah ok
# [21:38] <jesup> mwu: there's no fallback
# [21:38] * Quits: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net) (Quit: jorendorff)
# [21:38] <cpeterson> ted, some OpenGL code required SDK >= 8, but there is a (low-priority?) bug to try to include SDK >= 7 (Eclair).
# [21:38] <jesup> (unless we spend a lot of time writing one, or backport SLES if that's possible)
# [21:39] * Joins: mjschranz (mjschranz@moz-57B23647.senecac.on.ca)
# [21:39] * Quits: janv (varga@moz-6D58DC19.flarion.as5628.telecom.sk) (Ping timeout)
# [21:39] <mwu> I don't know what the numbers are like, but if there's enough people on 2.3 or newer, maybe you can get away with not supporting 2.2
# [21:39] * Quits: mjschranz (mjschranz@moz-57B23647.senecac.on.ca) (Client exited)
# [21:39] * Joins: brendan (brendaneic@AA16AC92.EF8E4D8F.43362C16.IP)
# [21:39] <mwu> it's possible to have a hack just for 2.2
# [21:39] <jesup> We can say "no WebRTC on < 2.3". Or we can limit FF to 2.3+. Or have "FF for old systems" and "FF for semi-current systems" ;-)
# [21:40] * Quits: brendan (brendaneic@AA16AC92.EF8E4D8F.43362C16.IP) (Quit: brendan)
# [21:40] <mwu> yeah I would want to do something like that
# [21:40] <cpeterson> jesup, 2.2 is about 21% of Android users, a significant number: https://developer.android.com/resources/dashboard/platform-versions.html
# [21:40] <jesup> which one?
# [21:40] <@ted> oof
# [21:40] <mwu> but, if you really want support
# [21:40] <@ted> i think not supporting webrtc on 2.2 would be okay
# [21:40] <mwu> you can build an opensles just for 2.2
# [21:40] <@ted> but you'd still have to rewrite the code somehow to not link directly to it
# [21:41] <mwu> the main worry is that the audiotrack api that opensles builds on top of may change from android to android
# [21:41] <mwu> but if it's for 2.2 only, it might be ok
# [21:41] <jesup> ted: right. Use a latebinding trick like we do for Alsa
# [21:42] <jesup> I wonder what usage FF has and what it will get in the future on 2.2
# [21:42] * Joins: mjschranz (mjschranz@moz-57B23647.senecac.on.ca)
# [21:42] * Joins: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net)
# [21:42] * merike|away is now known as merike
# [21:43] <mwu> there has also been libraries initially introduced in earlier versions of android and then officially supported in later versions
# [21:43] <mwu> like stlport
# [21:43] * Joins: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP)
# [21:43] <mwu> so maybe some android 2.2 builds have opensles. but I'd bet against it
# [21:44] <bjacob> where can i find build instructions for WebRT?
# [21:44] * Quits: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP) (Quit: brendan)
# [21:44] * Quits: jorendorff (jorendorff@moz-1747FB68.hsd1.tn.comcast.net) (Ping timeout)
# [21:44] <jesup> bjacob: what's WebRT? Do you mean WebRTC?
# [21:44] <bjacob> jesup: no, WebRT seems to be a thing to run web apps without the browser ui
# [21:44] <fabrice> bjacob: just do a trunk build
# [21:45] * Joins: WeirdAl (chatzilla@moz-D461843.ask.info)
# [21:45] <jesup> Aha, thanks
# [21:45] <bjacob> fabrice: what mozconfig
# [21:45] <fabrice> on windows or mac, since there is no linux support
# [21:45] <bjacob> ?
# [21:45] <bjacob> oh
# [21:45] <bjacob> no linux support?
# [21:45] <fabrice> bjacob: firefox
# [21:45] <taras> bz: restyling is probably focus?
# [21:45] * Quits: Enn (enn@F2D29657.F60B0462.67AC9B1.IP) (Ping timeout)
# [21:45] <fabrice> bjacob: not yet
# [21:45] <bjacob> argh
# [21:45] <taras> bz: ie borders changing, etc
# [21:45] <jesup> Ok, I'll push to 9 for now, and we can resolve what to do about it later
# [21:45] <bjacob> fabrice: i dont get it. firefox != webrt, right?
# [21:45] * Joins: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP)
# [21:46] * Quits: mjschranz (mjschranz@moz-57B23647.senecac.on.ca) (Client exited)
# [21:46] <fabrice> bjacob: the webrt stub is included in the fx build
# [21:46] <bjacob> fabrice: ah ok
# [21:46] <taras> bz: i saw some selector matching too
# [21:46] <WeirdAl> philor: ping re bug 735152: the latest reported failure does not look like the other failures on that bug... looks like a jsgc assertion failure instead
# [21:46] <bjacob> fabrice: not even a chance of some minimal linux support? i'm not going to reboot to windows and singletask on that just to debug webrt :-/
# [21:47] <firebot> Check-in: http://hg.mozilla.org/mozilla-central/rev/e2678ef38a0d - Malini Das - Bug 750553 - Rename getAttributeValue to getElementAttribute, r=jgriffin, a=testonly, npotb
# [21:47] * bjacob sucks it up
# [21:47] <fabrice> bjacob: not for now I'm afraid.
# [21:48] <bjacob> fabrice: please hear my plea to consider not only end-users but also developer productivity when deciding not to support linux
# [21:48] <@ted> oh boy
# [21:48] <@bz> taras: not sure
# [21:48] <@ted> i'm sure having both WebRT and WebRTC won't be confusing *at all*
# [21:48] <fabrice> bjacob: I was the first to complain (ask people in #openwebapps, they love me). There's a GSOC student doing it
# [21:48] * Joins: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP)
# [21:48] <@bz> taras: but it's taking a _long_ time
# [21:48] <bjacob> a gSOC? seriously?
# [21:49] * @bz woonders why the hell
# [21:49] <bjacob> so no chance at all to get it before august?
# [21:49] <mwu> lol gsoc
# [21:49] <jesup> ted: Need to rename it "WebRT-not-C"
# [21:49] <@bz> taras: so onLocationChange
# [21:49] <@bz> taras: in browser.js
# [21:49] <@bz> taras: runs on tab switch
# [21:49] <edmorley> ted: webrt, webrtc, winrt \o/
# [21:49] <@bz> taras: and does this:
# [21:49] <@bz> if (gFormSubmitObserver.panelIsOpen()) {
# [21:49] <@bz> gFormSubmitObserver.panel.hidePopup();
# [21:49] <@bz> }
# [21:49] * Joins: cadecairos (cadecairos@78D3F304.33EE9F8A.1139E686.IP)
# [21:49] <sfink> webr0t?
# [21:50] <taras> gavin: ^
# [21:50] <jhammel> webrtfm
# [21:50] * Joins: Ms2ger (Ms2ger@moz-997EEB55.access.telenet.be)
# [21:50] <gavin> ?
# [21:50] <@bz> taras: that's causing layout flushes that take some tme
# [21:50] <@bz> er, time
# [21:51] <bjacob> MozRT, but that would evocate music
# [21:51] <jhammel> webrotc for those of military inclination
# [21:51] * jaws is now known as jaws|brb
# [21:51] <@bz> taras: man
# [21:52] <@bz> taras: and then there's setMode for the identity box:
# [21:52] * Quits: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP) (Ping timeout)
# [21:52] <@bz> this._identityBox.className = newMode;
# [21:52] <@bz> this.setIdentityMessages(newMode);
# [21:52] <@bz> // Update the popup too, if it's open
# [21:52] <@bz> if (this._identityPopup.state == "open")
# [21:52] <@bz> this.setPopupMessages(newMode);
# [21:52] <@bz> Checking whether it's open flushes out the restyle from the className set, if any
# [21:52] <taras> that's awesome
# [21:53] <@bz> taras: our UI is like a typically-written web page
# [21:53] <gavin> bz: are you taking notes?
# [21:53] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [21:53] <@bz> gavin: trying to
# [21:53] <gavin> our UI doesn't get this way all at once :)
# [21:53] <@bz> gavin: got a place to dump them?
# [21:53] <taras> bz: i got etherpad
# [21:53] <gavin> bz: an etherpad? my inbox?
# [21:53] <@bz> gavin: sure, neither do web pages. ;)
# [21:53] <@bz> taras: linkme?
# [21:54] <gavin> someone needs to sit down and do some measuring/analysis
# [21:54] <taras> bz: https://etherpad.mozilla.org/SnappyTabstrip
# [21:54] <@bz> note, btw....
# [21:54] <@bz> that I'm trying to correlate a profile with what my debugger is telling me
# [21:54] <taras> gavin: that's what bz is doing, following up my dev.platform post
# [21:54] <gavin> taras: I mean someone who has more time and can write fixes :)
# [21:54] <@bz> so I know I have three separate calls to GetPopupState
# [21:54] <taras> gavin: oh that's ttaubert :)
# [21:54] <jdm> hmm, at what point in page load is nsGlobalWindow::SetDocShell called?
# [21:54] <gavin> but I'll take all the bz time I can get
# [21:54] <@bz> but I don't know which subset of them leads to the flushes
# [21:55] <jdm> I'm seeing a null docshell in a script block that is executed while loading the page
# [21:55] <@bz> the flushes that are 12% of my non-painting time
# [21:55] <@bz> jdm: SetDocShell is called before a page ever starts loading
# [21:55] <@bz> jdm: on the outer window
# [21:55] <@bz> jdm: how are you looking for the docshell?
# [21:56] * Joins: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP)
# [21:56] * Quits: bbondy2 (bbondy@moz-C9962B2.home.cgocable.net) (Quit: Going offline, see ya! (www.adiirc.com))
# [21:56] * Quits: jdm (jdm@moz-749CB30C.dsl.scarlet.be) (Ping timeout)
# [21:57] * Quits: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP) (Quit: Konversation terminated!)
# [21:57] * Joins: Mavericks (Mibbit@39785431.A962734.FDEA3160.IP)
# [21:58] * Joins: jdm (jdm@moz-749CB30C.dsl.scarlet.be)
# [21:59] * Joins: jviereck (Adium@moz-26045BE5.dclient.hispeed.ch)
# [21:59] <firebot> Check-in: http://hg.mozilla.org/mozilla-central/rev/3266a0e9f1b2 - William Lachance - Bug 750302 - DeviceManager files should have new-style license boilerplate, r=jmaher, a=android-only
# [21:59] <@bz> jdm: see above?
# [22:00] <jdm> bz: nope
# [22:00] <@bz> jdm: SetDocShell is called before a page ever starts loading
# [22:00] * Joins: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP)
# [22:00] <@bz> jdm: on the outer window
# [22:00] <@bz> jdm: how are you looking for the docshell?
# [22:00] <@ted> why are my youtube videos so janky lately :-(
# [22:00] <jdm> bz: mDocShell in nsGlobalWindow
# [22:00] <KWierso> ted: flash or html5?
# [22:00] <taras> ted: cache
# [22:00] <@bz> jdm: which nsGlobalWindow?
# [22:00] <edmorley> mbrubeck: looks like a lucky someone has a shiny new randomorange present :-)
# [22:00] <@bz> jdm: inner windows have a null mDocShell always
# [22:00] <@ted> KWierso: i think flash
# [22:00] <@ted> taras: :-(
# [22:00] <@bz> jdm: so how you're getting your nsGlobalWindow* matters
# [22:00] <@ted> taras: i tried to let it load all the way and it doesn't help
# [22:01] <@ted> the video janks
# [22:01] <@ted> ugh
# [22:01] <taras> ted: ah, then background tabs :)
# [22:01] <gaston> aaarrrrrghhh
# [22:01] <gaston> include/mozilla/a11y/Role.h:817: error: comma at end of enumerator list
# [22:01] <@ted> bleh
# [22:01] <jdm> bz: http://pastebin.mozilla.org/1610673
# [22:01] <taras> ted: and cache from them
# [22:01] <jhammel> why are my youtube videos all in flash now and the setting to promote webm has disappeared from the UI?
# [22:01] <jhammel> oh wait, i know why
# [22:01] <taras> ted: https://bugzilla.mozilla.org/show_bug.cgi?id=734979
# [22:02] <@bz> gaston: yeah, don't do that
# [22:02] <taras> ted: https://bugzilla.mozilla.org/show_bug.cgi?id=702485
# [22:02] <gaston> bz: don't "aaaarrrrrrghhhh" ? :)
# [22:02] <@bz> jdm: you have this in gdb right now?
# [22:02] <jdm> bz: yep
# [22:02] <@bz> gaston: don't leave trailing commas
# [22:02] * Quits: yuan (ywang@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: yuan)
# [22:02] <gaston> bz: i'm blaming whoever left it !
# [22:02] * Quits: Wevah (Wevah@moz-5435D49F.stcd.qwest.net) (Quit: Wevah)
# [22:03] * Quits: jammink (textual@moz-BBE3ABD.mv.mozilla.com) (Quit: Computer has gone to sleep.)
# [22:03] * Joins: jimb (user@9E727688.17C2F9C.163DC5C6.IP)
# [22:03] * Joins: catalinb (ethereal@805383FD.71B23C4A.5BC07656.IP)
# [22:03] * jaws|brb is now known as jaws|away
# [22:03] <@bz> ok
# [22:03] * Quits: jet (junglecode@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: jet)
# [22:03] <@bz> so your mDocShell is nill
# [22:03] <@bz> er, null
# [22:03] <gaston> and will report yet another bug from the "we're using a uber-strict compiler which makes kittens cry"-dept
# [22:03] <@bz> what's your mOuterWindow?
# [22:03] <@bz> Also, what's the value of mIsInnerWindow?
# [22:03] * Joins: fabrice1 (fabrice@moz-BBE3ABD.mv.mozilla.com)
# [22:03] <jduell> geoffbrown: ping
# [22:03] * @bz assumes null and false
# [22:03] <jdm> bz non-null and null, respectively
# [22:03] <jdm> er
# [22:04] <jdm> bz: true
# [22:04] <@bz> ok
# [22:04] <@bz> what's mOuterWindow->mDocShell ?
# [22:04] <jdm> non-null
# [22:04] <@bz> ok
# [22:04] <@bz> there you go
# [22:04] <@bz> everything working as expected
# [22:04] <jdm> should I be doing a forward-to-outer thing?
# [22:04] <@bz> doing it where?
# [22:05] <@bz> GetLocalStorage forwards to inner
# [22:05] <jdm> nevermind, I was confused
# [22:05] <jdm> yes
# [22:05] <@bz> GetDocShell() will get from the outer as needed
# [22:05] <jdm> oh wait
# [22:05] <jdm> yes
# [22:05] * Joins: cpearce (chatzilla@moz-510B10B9.xdsl.xnet.co.nz)
# [22:05] <jdm> there's even a docShell variable in the method already, I just used mDocShell instead :(
# [22:05] <jdm> sorry
# [22:05] <@bz> all good
# [22:06] * @bz goes back to his profile
# [22:06] * Quits: fabrice1 (fabrice@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving.)
# [22:06] * Joins: sworkman_ (sworkman@moz-BBE3ABD.mv.mozilla.com)
# [22:07] <@bz> taras: I really wish I had unified JS+C++ stacks
# [22:07] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [22:07] * sworkman_ is now known as sworkman
# [22:07] <jdm> I bet asuth's gdb extension to give you those is broken by now
# [22:07] * Quits: dwendorf (chatzilla@moz-6E3B32E8.lightspeed.mdsnwi.sbcglobal.net) (Quit: ChatZilla 0.9.88.2 [Firefox 15.0a1/20120501030504])
# [22:07] <jdm> although that wouldn't help with profiling
# [22:08] <jdm> that would be so nice
# [22:08] * Joins: jet (junglecode@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [22:08] * Joins: dwendorf (chatzilla@moz-6E3B32E8.lightspeed.mdsnwi.sbcglobal.net)
# [22:08] <jlebar> bz, Can I, from chrome JS, capture all events from an iframe and keep content from receiving them? (I'd settle even for capturing only 'load' events.)
# [22:08] * armenzg_buildduty is now known as armenzg_brb
# [22:08] * merike is now known as merike|away
# [22:09] * nthomas|away is now known as nthomas
# [22:09] <@bz> jlebar: I think so, yes
# [22:09] <@bz> jlebar: just add a capturing listener and call event.stopPropagation()
# [22:09] <@bz> jlebar: (modulo whatever e10s needs, of course)
# [22:09] <josh> sworkman: https://wiki.mozilla.org/ReleaseEngineering/DisposableProjectBranches
# [22:10] * Quits: evilpie_ (evilpie@moz-451AEED8.pools.arcor-ip.net) (Quit: ChatZilla 0.9.88.2 [Firefox 15.0a1/20120430030504])
# [22:10] * Quits: Hendikins (wolfox@moz-16899DFF.hhui4.ken.bigpond.net.au) (Ping timeout)
# [22:10] <@bz> gavin: is it expected that frames are recreated for <tab> elements?
# [22:10] <jlebar> bz, This is only in-process -- e10s just stops the events altogether, which is what I want. But how do I know that my capturing listener fires before anyone else's capturing listeners?
# [22:10] <@bz> gavin: on tab switch?
# [22:10] <Ms2ger> bz, I really wish we had split global window classes ;)
# [22:10] * Joins: Hendikins (wolfox@moz-16899DFF.hhui4.ken.bigpond.net.au)
# [22:10] <@bz> jlebar: capturing listeners fire from the top of the tree down
# [22:10] <@bz> jlebar: so a chrome one will always fire before content
# [22:10] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Input/output error)
# [22:10] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
# [22:11] <@bz> jlebar: if you need to fire before other _chrome_ listeners, then that's different
# [22:11] * Quits: Mavericks (Mibbit@39785431.A962734.FDEA3160.IP) (Quit: http://www.mibbit.com ajax IRC Client)
# [22:11] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [22:11] * Joins: rstrong (rstrong@moz-217F02CE.lightspeed.sntcca.sbcglobal.net)
# [22:11] * Joins: Wevah (Wevah@moz-5435D49F.stcd.qwest.net)
# [22:11] <taras> bz: and yeah, waiting on new jit, so we can do mixed stacks
# [22:11] <jlebar> bz, No, that's fine. But...my capturing listener can't be on the iframe, then. It needs to be on the chrome event handler, or the chrome window, or something? So it sees all 'load' events?
# [22:12] <taras> bz: or atleast a sane way to do js stacks :(
# [22:12] <@bz> taras: heh
# [22:12] <@bz> jlebar: chrome event handler, yep
# [22:12] <@bz> jlebar: if you want to see all load events everywhere
# [22:12] <@bz> jlebar: if you just want them for the tabs, <tabbrowser> would work fine
# [22:12] <gavin> bz: I imagine not
# [22:12] * gavin is in a meeting now, though
# [22:12] <gavin> bbl
# [22:13] <@bz> ok
# [22:13] * @bz will dig more
# [22:13] <espindola> who is is the "content team"? I was asked to cc someone in there in bug 742455, but I don't see anything in https://wiki.mozilla.org/Modules/Activities#Module_Ownership_System
# [22:13] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [22:14] <jlebar> bz, Okay, I think I have the general idea. Thanks.
# [22:14] * Quits: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Connection reset by peer)
# [22:14] * Joins: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [22:14] * Quits: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Quit: bholley_mobile)
# [22:14] * Joins: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [22:14] <jlebar> espindola, Now known as the "dom" team, for the sake of confusion.
# [22:15] * Quits: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP) (Quit: brendan)
# [22:15] <cjones> jlebar, "OOP iframes have null contentWindow's" --- you're talking about wrt chrome/embedding code in the parent process, right?
# [22:15] <espindola> jlebar: ah :-)
# [22:15] * Quits: jviereck (Adium@moz-26045BE5.dclient.hispeed.ch) (Quit: Leaving.)
# [22:15] <espindola> jlebar: still not in https://wiki.mozilla.org/Modules/Activities#Module_Ownership_System
# [22:15] <jlebar> cjones, Only the embedding code can touch the iframe element, so yes.
# [22:15] <jlebar> espindola, I have no idea why you're on the "Activities" page.
# [22:16] <jlebar> espindola, https://wiki.mozilla.org/Modules/All ?
# [22:16] <espindola> search failed
# [22:16] <espindola> found it, thanks
# [22:16] * Parts: ajuma (ajuma@F2D29657.F60B0462.67AC9B1.IP)
# [22:16] * Joins: mjschranz (mjschranz@moz-57B23647.senecac.on.ca)
# [22:16] * Joins: ajuma (ajuma@F2D29657.F60B0462.67AC9B1.IP)
# [22:16] <jlebar> cjones, I can hack something up which will make in-process mozbrowser look similar to OOP. I don't know if that will be good enough.
# [22:17] * bear is now known as bear-afk
# [22:17] * Joins: jviereck (Adium@moz-26045BE5.dclient.hispeed.ch)
# [22:17] <jlebar> cjones, To be truly the same, though, is complicated. It involves e.g. session history isolation, which I haven't looked into at all.
# [22:17] <@ehsan> glandium: ted: which one of you guys can review my libpng removal patch?
# [22:18] <@ted> depends on how complicated it is?
# [22:19] * Quits: jviereck (Adium@moz-26045BE5.dclient.hispeed.ch) (Quit: Leaving.)
# [22:19] * Joins: jorendorff (jorendorff@moz-BA7063A1.hsd1.tn.comcast.net)
# [22:19] <jlebar> cjones, If you're being embedded -- if you're inside the mozbrowser -- you can't touch the iframe element.
# [22:19] <@ehsan> ted: not that complicated: https://bugzilla.mozilla.org/attachment.cgi?id=620043&action=edit
# [22:20] <jesup> ted: Looks like android-9 breaks the fancy ELF loader...: build/mozglue/linker/ElfLoader.cpp:597: error: no matching function for call to 'sigaltstack::sigaltstack(NULL, stack_t*)'
# [22:20] <@ted> that's unfortunate
# [22:20] * Quits: dao (dao@moz-95A844B3.superkabel.de) (Quit: Leaving.)
# [22:20] * Quits: JeroenDeDauw (j@moz-95881CDD.dip.t-dialin.net) (Quit: Leaving.)
# [22:21] <bholley> ehsan: is there a tracking bug for code reductions?
# [22:21] * Joins: mayhemer__ (Miranda@B3D46202.F87A741B.F23860FD.IP)
# [22:21] * Quits: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP) (Ping timeout)
# [22:21] <@ehsan> bholley: not sure
# [22:21] <jesup> yes. This means I abandon WebRTC on Android until that gets fixed.
# [22:22] * Joins: ekr (ekr@moz-7B0110AD.mv.mozilla.com)
# [22:22] <bholley> ehsan: ok. I'll just make it block deadcode
# [22:22] <cjones> jlebar, i don't care all that much about it looking the same
# [22:22] <mwu> glandium: ^
# [22:22] <jlebar> cjones, Well, it matters to Gaia.
# [22:22] <cjones> jlebar, there are things that are off-limits for the cross-process case
# [22:22] <cjones> it shouldn't! :)
# [22:22] <jlebar> cjones, Like, Gaia touches iframe.contentWindow.
# [22:22] <cjones> those are things we need to fix
# [22:22] <jlebar> cjones, Exactly.
# [22:23] * Joins: sicking (chatzilla@454A1A4F.FF5089F.1771B9DE.IP)
# [22:23] <@ted> jesup: file a bug?
# [22:23] <cjones> gaia can't touch that
# [22:23] <cjones> do you mean b2g chrome?
# [22:23] <jlebar> cjones, So if you want to be able to "test" Gaia on Mac, it needs to look the same.
# [22:23] <jesup> ted: doing so. (know the correct component?)
# [22:23] * Joins: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
# [22:23] <@ehsan> bholley: can you CC me please?
# [22:23] <bholley> ehsan: sure thing
# [22:23] * @ehsan needs to attach libjpeg next
# [22:23] <cjones> jlebar, a good example here is the code we use to remote DOM APIs
# [22:24] <cjones> which has to run single-process and cross-process
# [22:24] <cjones> we could pass pointers everywhere
# [22:24] <geoffbrown> jduell; pong
# [22:24] <cjones> and even serialize them in IPDL
# [22:24] <cjones> (and folks have done that)
# [22:24] * Joins: bholley_mobile| (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [22:24] * Quits: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Connection reset by peer)
# [22:24] <cjones> but pointer passing is off-limits for dev
# [22:24] <@ted> not offhand
# [22:24] * Quits: mjschranz (mjschranz@moz-57B23647.senecac.on.ca) (Client exited)
# [22:24] <cjones> there's a bringup time when we have to smoke that code out
# [22:24] <jduell> geoffbrown: nothing--I figured it out and +r'd your patch
# [22:24] * Quits: bholley_mobile| (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Quit: Bye)
# [22:24] <cjones> and then it's review guidelines
# [22:24] * Joins: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP)
# [22:24] * Joins: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [22:24] <cjones> / testing on the realistic config
# [22:25] <edmorley> bholley++
# [22:25] <bholley> edmorley: :-)
# [22:26] <geoffbrown> jduell: thanks
# [22:26] * Quits: jdm (jdm@moz-749CB30C.dsl.scarlet.be) (Ping timeout)
# [22:26] <jduell> geoffbrown: np
# [22:26] * Quits: ekr (ekr@moz-7B0110AD.mv.mozilla.com) (Quit: ekr)
# [22:27] <jlebar> cjones, I think that's a recipe for wasted time for the people working on Mac -- they're going to write code in a more permissive environment than it will run. But I'm not one of those people, so I don't care, if you really think that's the way to go.
# [22:27] <jlebar> cjones, But it's more complicated than that...
# [22:27] * Joins: jdm (jdm@moz-749CB30C.dsl.scarlet.be)
# [22:27] <jlebar> cjones, Because there are things which will be *different* in in-process and oop. For example, session history.
# [22:27] * Quits: Ms2ger (Ms2ger@moz-997EEB55.access.telenet.be) (Ping timeout)
# [22:27] <@roc> woo bug 716439 is all r+ed up (and the last try run was green)
# [22:27] <@roc> now we just need the tree to open
# [22:28] <jdm> roc: pfft, just land with CLOSED_TREE
# [22:28] * Quits: Optimizer (Mibbit@6E78482A.3686F45E.89AC0F27.IP) (Quit: http://www.mibbit.com ajax IRC Client)
# [22:28] * Joins: anky (anky@2935FD87.5111A3CC.A3D1B221.IP)
# [22:29] * armenzg_brb is now known as armenzg_buildduty
# [22:29] * Joins: ekr (ekr@moz-7B0110AD.mv.mozilla.com)
# [22:30] * Joins: ericjung (Mibbit@moz-C6B344D.bstnma.fios.verizon.net)
# [22:30] * juanb|lunch is now known as juanb
# [22:30] * @bz finds why we reframe, thinks it might be mac-only
# [22:30] <@bz> taras: this is the other fun
# [22:30] <@bz> taras: different CSS on different platforms....
# [22:32] <cjones> jlebar, similarly, forcing OOP on mac users is a recipe for wasted time, since it's ~Tier III
# [22:32] <cjones> jlebar, what's the session history difference?
# [22:32] <Matt_> bz: ping
# [22:33] * Joins: karl (karl@F0AB138F.333C855E.C8A09C26.IP)
# [22:33] <@bz> Matt_: pong
# [22:33] <Matt_> bz: you probably know this off the top of your head
# [22:33] <jlebar> cjones, Session history sees iframe mozbrowser as a normal iframe. That means that <back> in the top-level frame can navigate the mozbrowser.
# [22:33] <Matt_> bz: I want to redirect certain responses to another URL
# [22:33] <jlebar> cjones, Last time I looked, we were using history.back in some places.
# [22:33] <@bz> responses or requests?
# [22:33] <Matt_> well
# [22:34] <Matt_> ideally requests but I'm trying to use a tracing listener
# [22:34] <@bz> Also, is this "I" a browser extension, browser core, or a web server? ;)
# [22:34] <Matt_> so at that point I already have the response
# [22:34] * Matt_ is a browser extension
# [22:34] <cjones> jlebar, whoa, really? how does that work?
# [22:34] <@bz> you can't do this with a tracing listener
# [22:34] <Matt_> yeah ok
# [22:34] <cjones> jlebar, (not being argumentative, i honestly don't know)
# [22:34] <Matt_> I was starting to come to that realization
# [22:34] * @bz is not sure we have a good way to do this at all, really
# [22:34] <jlebar> cjones, What exactly do you want to know?
# [22:34] <@bz> though we have several requests for it
# [22:34] <cjones> jlebar, "That means that <back> in the top-level frame can navigate the mozbrowser."
# [22:35] <jlebar> Just like for an iframe.
# [22:35] <Matt_> bz: I guess I could do it with a tracing listener but I'd have to read the data from the old URL and just send other data to the listener
# [22:35] * jlebar doesn't understand what cjones wants to know.
# [22:35] <@bz> Matt_: mmmm
# [22:35] <Matt_> bz: am I SOL?
# [22:35] <jlebar> cjones, e10s fixes all this.
# [22:35] <@bz> Matt_: and do various other stuff involving telling the listener about the different channel, etc
# [22:35] <cjones> jlebar, so here <back> is let's say an outer test.com/embed.html page
# [22:35] <cjones> er, the top-level frame is that
# [22:35] * Quits: kaie (kaie@moz-55C6BE11.dip.t-dialin.net) (Quit: Leaving)
# [22:35] <cjones> a tab in FF let's say
# [22:35] <Matt_> bz: well is there a better less hacky way?
# [22:35] <@bz> Matt_: I believe noscript might actually have some hackery for this
# [22:36] <bent> hm, anyone know, if i overlay browser.xul, does that mean i get loaded into every tab? or just every window?
# [22:36] <jlebar> sure, okay...
# [22:36] <@bz> Matt_: there is no non-hacky way
# [22:36] <@bz> Matt_: at the moment
# [22:36] <Matt_> bz: hmmm, ok… that's a good tip
# [22:36] <Matt_> I'll have a look at no script then
# [22:36] <@bz> Matt_: good luck
# [22:36] <Matt_> I was surprised how easy Chrome makes it to do that
# [22:36] <@bz> we should really just make it easy for extensions to do this
# [22:36] <Matt_> I thought it was semi-evil
# [22:36] <cjones> jlebar, oh, i think i see ... if i click a link in the inner frame, <back> takes me back to the previous location?
# [22:36] * Joins: kaie (kaie@moz-55C6BE11.dip.t-dialin.net)
# [22:36] <jlebar> Yes.
# [22:36] <@bz> e.g. via content policy
# [22:36] <cjones> in the inner frame
# [22:37] <jlebar> cjones, Correct.
# [22:37] <Matt_> yeah I thought this was going to be much easier than it is
# [22:37] <cjones> ok
# [22:37] <Matt_> bz: maybe I could register my own HTTP protocol handler?
# [22:37] <cjones> i was thinking some nav event of the *outer* frame would result in the inner being navigated
# [22:37] <@bz> well, yes
# [22:37] <@bz> you could
# [22:37] <Matt_> that's probably better than the tracing listener way which is truly terrible
# [22:37] <@bz> if you only need this for http
# [22:37] <Matt_> http and https I guess
# [22:37] <jlebar> cjones, well, history.back is an event of the outer iframe.
# [22:37] * Quits: harth (harth@88F51059.F3BBB17D.144F44FA.IP) (Input/output error)
# [22:37] <@bz> the problem is I suspect we precache the handler
# [22:37] <@bz> so overriding it is probably hard
# [22:37] * Quits: smontagu (chatzilla@moz-A6D65AB5.red.bezeqint.net) (Ping timeout)
# [22:37] <Matt_> pre cache it where?
# [22:37] * Joins: smagnin (pike@moz-1816673.w90-57.abo.wanadoo.fr)
# [22:37] <Matt_> in the IO service?
# [22:38] <@bz> yes
# [22:38] <cjones> jlebar, so two things --- i'm not sure things *don't* work like that in the cross-process case
# [22:38] * Matt_ checks
# [22:38] <cjones> jlebar, and if they don't, what's your concern?
# [22:38] <cjones> there a brazillion things that work differently in FF vs. b2g/gaia
# [22:38] <cjones> *there are
# [22:38] <jlebar> cjones, Supposing they don't -- I haven't tested, but I was pretty sure -- the concern is that code which uses history.back in Gaia will behave differently in OOP gaia vs. non-OOP gaia.
# [22:39] <jlebar> cjones, To the point that such code will not work properly in non-OOP Gaia.
# [22:39] * Joins: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP)
# [22:40] <cjones> jlebar, in theory yeah, but mozbrowser consumers don't use history.back
# [22:40] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
# [22:40] <Matt_> bz: so there's no way for me to override it since it'll be in the cache before I get a chance?
# [22:40] <Matt_> I'm a bootstrapped extension actually
# [22:40] <Matt_> so I can't even be sure I'll be there at startup
# [22:40] <cjones> if that indeed doesn't work the same as in-process already (not sure if we impl'd that for fennec), it's an interesting question of whether we *want* that to work
# [22:41] <Matt_> I could override the IO service <evil>
# [22:41] <jdm> shudder
# [22:41] <Matt_> that's probably what I'll have to do
# [22:41] <Matt_> or I'll just spoof the data in the tracing listener
# [22:42] <Bas> taras: Bug 750871
# [22:42] * Quits: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
# [22:42] <jlebar> cjones, You're right, they currently don't (if they ever did).
# [22:42] <jlebar> cjones, I am very happy to make Gaia developers' lives harder in exchange for making my life easier.
# [22:42] * Quits: @roc (chatzilla@C0ACF8B.5E1E9EEA.613E47D1.IP) (Ping timeout)
# [22:42] <jlebar> cjones, It sounds like that's what you'd like. No problem. :)
# [22:42] * Quits: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP) (Quit: brendan)
# [22:43] * Joins: yuan (ywang@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [22:43] <cjones> jlebar, i'm suggesting we land some hard work you've already done in exchange for unblocking other things, and moving other less-critical issues off the critical path
# [22:44] <jlebar> cjones, Sure.
# [22:44] * Quits: ekr (ekr@moz-7B0110AD.mv.mozilla.com) (Quit: ekr)
# [22:44] * Joins: mike5w3c (MikeS@454A1A4F.FF5089F.1771B9DE.IP)
# [22:44] <taras> Bas: ftw
# [22:45] * Quits: @dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net) (Quit: 8403864 bytes have been tenured, next gc will be global.)
# [22:46] <josh> didn't we switch to a Windows compiler with a much bigger address space limit?
# [22:46] * Joins: jammink (textual@moz-BBE3ABD.mv.mozilla.com)
# [22:46] <@bz> taras: ok, I dumped some data
# [22:46] <@bz> taras: in the etherpad
# [22:47] * Joins: dao (dao@moz-95A844B3.superkabel.de)
# [22:47] * Joins: jviereck (Adium@moz-26045BE5.dclient.hispeed.ch)
# [22:47] * Quits: sicking (chatzilla@454A1A4F.FF5089F.1771B9DE.IP) (Quit: Reconnecting…)
# [22:47] * Quits: jviereck (Adium@moz-26045BE5.dclient.hispeed.ch) (Client exited)
# [22:47] * Quits: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP) (Ping timeout)
# [22:48] * Joins: sicking (chatzilla@454A1A4F.FF5089F.1771B9DE.IP)
# [22:48] <taras> bz: thanks will go through it in a bit
# [22:49] * KaiRo is now known as KaiRo_away
# [22:49] * jlebar is now known as jlebar|mtg
# [22:49] * Joins: roc (chatzilla@C0ACF8B.5E1E9EEA.613E47D1.IP)
# [22:49] * ChanServ sets mode: +o roc
# [22:49] <@bz> "the most relevant comments are 297 and higher."
# [22:49] * @bz cries
# [22:49] <edmorley> bz: always reassuring to see!
# [22:50] * Quits: graememcc (chatzilla@moz-DE5D49D8.range86-148.btcentralplus.com) (Quit: ChatZilla 0.9.88.2 [Firefox 11.0/20120310193349])
# [22:50] <edmorley> ehsan++
# [22:50] * Quits: icaaq (Adium@moz-718D9655.cust.bredbandsbolaget.se) (Quit: Leaving.)
# [22:51] <edmorley> our libxul dissecting superstar :-)
# [22:51] * Joins: kinetik (kinetik@B0506AEA.F200EF31.613E47D1.IP)
# [22:53] * Joins: mcsmurf (mcsmurf@moz-6B14C802.dip.t-dialin.net)
# [22:53] <zzzzz> looks like your just barely under the wire at this point 2992783360
# [22:53] * Joins: Jesse (jruderman@moz-E31CD2CB.mozilla.org)
# [22:54] * Joins: kumar (kmcmillan@moz-4BE034AB.ptr.us.xo.net)
# [22:54] * Quits: dcamp (dave@9E727688.17C2F9C.163DC5C6.IP) (Quit: Leaving...)
# [22:54] * Quits: jhopkins (jhopkins@moz-216F9986.tb.shawcable.net) (Input/output error)
# [22:54] * Quits: @smaug (chatzilla@moz-AD13B184.elisa-mobile.fi) (Input/output error)
# [22:55] <philor> WeirdAl: yeah, but 302 smaug, and I'd bet since he retriggered it four times and hit it every time, he already knows it was him causing it
# [22:55] <Matt_> bz: change the content type to HTML in http-on-examine-response and then spoof the data using a meta redirect?
# [22:55] <Matt_> you think that could work?
# [22:55] <Matt_> spoof the data in the tracing listener I mean
# [22:56] * Joins: smaug (chatzilla@moz-AD13B184.elisa-mobile.fi)
# [22:56] * ChanServ sets mode: +o smaug
# [22:56] * Joins: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP)
# [22:56] <WeirdAl> philor: 302???
# [22:57] * Joins: dcamp (dave@9E727688.17C2F9C.163DC5C6.IP)
# [22:57] * Quits: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [22:57] * Quits: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP) (Quit: brendan)
# [22:57] * Quits: @dveditz (dveditz@moz-5051E786.dhcp.cruzio.com) (Quit: dveditz)
# [22:57] * joduinn-mtg is now known as joduinn-food
# [22:58] <@bz> Matt_: heh. That sorta works if you don't care about POST
# [22:58] <Matt_> bz: I don't
# [22:58] <Matt_> maybe I should try that then
# [22:58] <@bz> Matt_: if you _do_ care about POST life is really hard
# [22:58] <Mook_as> josh: no; the build machines were (at one point) modified to allow an extra gig (2->3), but the linker is still stuck being a 32 bit binary. there are plans, IIRC, to move to win64-hosted 32-bit toolchain, which gets ~4G or so.
# [22:58] <Matt_> bz: I just want to redirect JS
# [22:58] <Matt_> I mean requests for JS files
# [22:58] <Matt_> unlikely to be posts
# [22:58] * Quits: anant (anant@moz-BBE3ABD.mv.mozilla.com) (Connection reset by peer)
# [22:58] * Joins: anant (anant@moz-BBE3ABD.mv.mozilla.com)
# [22:59] * Matt_ wonders if there is a bug for this
# [22:59] * Joins: dveditz (dveditz@moz-5051E786.dhcp.cruzio.com)
# [22:59] * ChanServ sets mode: +o dveditz
# [22:59] <@bz> redirecting stuff?
# [22:59] <@bz> yes
# [22:59] <Matt_> looking for it...
# [22:59] <Matt_> that no script code is like 6500 lines in a huge XPCOM component
# [23:00] <philor> WeirdAl: HTTP redirect, he starred it
# [23:00] <Matt_> can't see any spoofing redirects… it looks like he just catches the redirects when they happen and does crazy stuff with them
# [23:00] <Matt_> but I might be missing it
# [23:00] * WeirdAl is not parsing philor's English very well
# [23:00] * Quits: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP) (Ping timeout)
# [23:01] * Joins: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP)
# [23:01] * Joins: dseif (dseif@moz-57B23647.senecac.on.ca)
# [23:01] * coop|mtg is now known as coop|afk
# [23:01] <philor> WeirdAl: lemme try again: "smaug starred it, so you'll want to ask him about it, but if you click the link in the bug comment to the log, then right at the top there's a link to "for push abc123whatever" that takes you to tbpl's results of that push, which will tell you he already saw it was really his new failure"
# [23:02] * Joins: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP)
# [23:02] <taras> Bas: looking forward to the border-hax0red build
# [23:03] * Quits: dao (dao@moz-95A844B3.superkabel.de) (Connection reset by peer)
# [23:03] <Matt_> bz: would my approach work for HTTPS?
# [23:03] * Joins: rjohnson19 (chatzilla@moz-9148485F.hsd1.ma.comcast.net)
# [23:04] * rnewman|afk is now known as rnewman
# [23:04] <Matt_> I guess the stream is decrypted by the time the tracing listener sees it
# [23:04] <decoder> ehsan: ping?
# [23:04] <@ehsan> decoder: hey
# [23:05] * Joins: dao (dao@moz-95A844B3.superkabel.de)
# [23:05] <WeirdAl> philor: so basically "smaug already knows about this" :)
# [23:06] <@smaug> WeirdAl: what did I do?
# [23:06] * khuey|away is now known as khuey
# [23:06] <Bas> taras: I'll have it for you tomorrow, okay?
# [23:06] <jdm> you knew about things
# [23:06] <Matt_> bz: is this the bug: https://bugzilla.mozilla.org/show_bug.cgi?id=421224 ?
# [23:06] <jdm> that was your first mistake
# [23:06] <WeirdAl> smaug: we're talking about a jsgc assert reported as an orange on test_xhr_timeout
# [23:06] <WeirdAl> bug 735152, last two comments
# [23:06] <taras> Bas: no rush, i'm just excited. Can't wait to blog about 'ugly, but fast firefox'
# [23:07] <@smaug> WeirdAl: sorry
# [23:07] <@smaug> my mistake
# [23:07] <Matt_> no activity since 2008, that's encouraging
# [23:07] <@smaug> and tbpl being silly
# [23:07] * Matt_ notes that WeirdAl wrote a patch
# [23:07] <firebot> Check-in: http://hg.mozilla.org/mozilla-central/rev/281574985410 - Margaret Leibovic - Bug 747023 - Install software notification doorhanger popup can appear from page in background tab. r=mfinkle a=android-only
# [23:07] <WeirdAl> Not for jsgc, I didn't
# [23:07] <WeirdAl> I implemented the timeout code on XHR
# [23:07] <Matt_> WeirdAl: https://bugzilla.mozilla.org/show_bug.cgi?id=421224
# [23:08] <Matt_> I guess that's what I'm looking for… a way for extensions to force URL redirects
# [23:08] <@smaug> I wish I could star tryserver builds without adding comments to bugs
# [23:08] <WeirdAl> oh, different subject
# [23:08] * philor is now known as philor|away
# [23:08] <Matt_> yeah sorry, I didn't want to distract you
# [23:08] <Matt_> anyway it doesn't really help me since it never got finished/landed
# [23:08] <Matt_> but I'll CC
# [23:08] <WeirdAl> Matt_: it's assigned to nobody
# [23:09] <Matt_> yeah that nobody guy is a slacker
# [23:09] <WeirdAl> if you want progress on it, find someone who's interested in fixing
# [23:09] <mbrubeck> smaug: You can
# [23:09] * Quits: joe_walker (joe_walker@moz-15405DDA.cable.virginmedia.com) (Input/output error)
# [23:09] * joduinn-food is now known as joduinn-mtg
# [23:09] <mbrubeck> smaug: If you just press "Add comment" (or type "C") and then type in a comment manually, then it will not post anything to bugzilla
# [23:09] * Quits: tonymec (tonymec@78D6A323.3F6FF523.277517C1.IP) (Ping timeout)
# [23:10] <@smaug> ah, true
# [23:10] <Matt_> WeirdAl: I doubt someones interested in fixing are that easy to come by
# [23:10] * Quits: dseif (dseif@moz-57B23647.senecac.on.ca) (Ping timeout)
# [23:10] <Matt_> anyway I'll find some kind of hack
# [23:10] <WeirdAl> hehe
# [23:10] * Quits: Standard8 (Standard8@B7F1AE36.48015583.54C3481B.IP) (Quit: ZNC - http://znc.in)
# [23:10] * mattwoodrow|away is now known as mattwoodrow
# [23:11] * Joins: dseif (dseif@moz-57B23647.senecac.on.ca)
# [23:11] * Quits: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [23:11] * armenzg_buildduty is now known as armenzg_afk
# [23:13] * Joins: dseif_ (dseif@moz-57B23647.senecac.on.ca)
# [23:13] <WeirdAl> Matt_: I think URI's are now immutable, which might make a fix for this bug significantly harder if you go the way I did
# [23:13] * Quits: edmorley (edmorley@moz-E455C402.range86-145.btcentralplus.com) (Quit: nn)
# [23:13] <khuey> URI's are immutable?
# [23:13] <khuey> since when?
# [23:13] * Quits: JonathanS (JonathanS@17EDFC35.8737F162.521902B0.IP) (Quit: Computer has gone to sleep.)
# [23:13] <WeirdAl> I think they are. I'm not sure.
# [23:13] <khuey> don't think so
# [23:13] <khuey> bz would know
# [23:14] * Quits: dseif (dseif@moz-57B23647.senecac.on.ca) (Ping timeout)
# [23:15] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
# [23:15] * Joins: ekr (ekr@moz-7B0110AD.mv.mozilla.com)
# [23:16] * Quits: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Connection reset by peer)
# [23:16] * Joins: bholley_mobile| (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [23:16] * Joins: bsmith (bsmith@moz-BBE3ABD.mv.mozilla.com)
# [23:17] * Quits: bholley_mobile| (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr) (Quit: Bye)
# [23:19] * Quits: bent (chatzilla@moz-C3562645.hsd1.ca.comcast.net) (Quit: Reconnecting…)
# [23:19] * Joins: bent (chatzilla@moz-C3562645.hsd1.ca.comcast.net)
# [23:19] <bsmith> hhuey: ekr and the WebRTC people would like to land 750,000 lines of code this summer. Can we stuff that in libxul?
# [23:19] <khuey> no
# [23:19] * khuey is now known as khuey|away
# [23:20] * Quits: tonymec|away (tonymec@78D6A323.3F6FF523.277517C1.IP) (Ping timeout)
# [23:20] <ekr> how about 749,999 lines?
# [23:20] <bholley> khuey|away: yt?
# [23:20] * Joins: cviecco_ (cviecco@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [23:20] <bsmith> We can move most of that to gkmedia. But, I heard that gkmedia doesn't have locking primitives and other necessary bits
# [23:20] * Joins: Pike (Pike@moz-EC415F92.dip.t-dialin.net)
# [23:20] * Quits: armenzg_afk (armenzg@moz-ED3249A4.dmz.releng.scl3.mozilla.com) (Input/output error)
# [23:20] * Joins: bholley_mobile (bholley_mo@moz-409EE9C9.net-81-220-20.rev.numericable.fr)
# [23:21] * Quits: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP) (Max SendQ exceeded)
# [23:21] * Quits: @roc (chatzilla@C0ACF8B.5E1E9EEA.613E47D1.IP) (Ping timeout)
# [23:21] * Joins: nigelb (Adium@99054AA0.28FB815D.274D17D6.IP)
# [23:22] * Quits: catalinb (ethereal@805383FD.71B23C4A.5BC07656.IP) (Ping timeout)
# [23:22] <bsmith> e.g. mozilla:Mutex. I don't know if that is true but in general, where should we put/move code that both libxul and gkmedia (or whatever) depend on?
# [23:22] <ekr> The thing I'm dealing with right this second is that I have some networking code that wants to use nsSocketTransportService, but that means I need to be calling into libxul from gkmedia, which I am told is a no-no.
# [23:22] <bsmith> And, is it wrong or bad to add NSPR as a dependency of gkmedia?
# [23:22] * Quits: drice (derice@1606D15F.E628B196.8E155D4E.IP) (Quit: Leaving.)
# [23:23] <rail> espindola: hey, do you mind if I upgrade clang to 155417 (bug 748208) for all platforms and remove older ones? (would need changes in your mozconfigs, I think)
# [23:24] * philor|away is now known as philor
# [23:24] * Quits: msucan (mihai@7A4CDDE7.7A418818.B4CEF140.IP) (Quit: Leaving.)
# [23:25] * Joins: harth (harth@moz-C2C235AC.bb.sky.com)
# [23:25] <taras> bz: still there?
# [23:25] <WeirdAl> bsmith: the reason against libxul is that we're hitting limits on profile-guided optimization - in fact, we just started hitting the 3GB ceiling on Windows again
# [23:26] * Joins: int3 (int3@DD99568F.EC8E31C5.474B6A5B.IP)
# [23:26] <bsmith> WeirdAl: Yes, I know.
# [23:26] <WeirdAl> oh, ok
# [23:27] <ekr> WeirdAl: we weren't really serious about pulling into libxul. But we are trying to figure out what the best way to proceed is given that we are likely to have some dependencies
# [23:27] * Quits: @smaug (chatzilla@moz-AD13B184.elisa-mobile.fi) (Ping timeout)
# [23:27] <ekr> between gkmedia and libxul
# [23:27] <@bz> taras: yes
# [23:27] <taras> bz: I wonder whether it would make sense to do the popup box oject and scrollbox stuff off refresh ticks?
# [23:28] <@bz> ekr: so..
# [23:28] <taras> refresh ticks being mozrequestanimation frame?..and how that help
# [23:28] * Joins: smaug (chatzilla@moz-AD13B184.elisa-mobile.fi)
# [23:28] * ChanServ sets mode: +o smaug
# [23:28] * Quits: @ehsan (ehsan@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [23:28] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
# [23:28] * Joins: tonymec|away (tonymec@78D6A323.3F6FF523.277517C1.IP)
# [23:28] <@bz> ekr: nsSocketTransportService implements interfaces, which expose virtual functions, which you can call, no?
# [23:29] <ekr> bz: This is something I'm still getting up to speed on.
# [23:29] <taras> bz: also where to set breakpoint to catch layout flushes
# [23:29] <@bz> ekr: which functions are you trying to call?
# [23:29] <bsmith> bz: I think I can help ekr out with that part.
# [23:29] <@bz> bsmith: awesome, thanks
# [23:29] <@bz> taras: mmmm
# [23:29] * bwinton is now known as bwinton_away
# [23:29] <bsmith> the part that is more problematic is mfbt and NSPR and other very basic building blocks that aren't available in gkmedia yet
# [23:30] <@bz> hrm
# [23:30] <@bz> mfbt should be
# [23:30] <@bz> if not, we should fix
# [23:30] <@bz> taras: PresShell::FlushPendingNotifications is an obvious option
# [23:30] * catlee is now known as catlee-away
# [23:31] <@bz> taras: though note that this will be called off the refresh driver too
# [23:31] * Quits: mconley (mconley@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [23:31] <taras> bz: PresShell::FlushPendingNotifications called on every refresh driver call?
# [23:31] <@bz> taras: no, only when there is dirty stuff in the presshell
# [23:32] <taras> right
# [23:32] * Quits: stevee (Miranda@moz-BEBDF855.cable.virginmedia.com) (Quit: Miranda IM - Multi protocol instant messenger @ www.miranda-im.org)
# [23:32] * Quits: bbondy (bbondy@moz-C9962B2.home.cgocable.net) (Ping timeout)
# [23:32] <@bz> taras: what are you trying to measure?
# [23:32] * Quits: cpearce (chatzilla@moz-510B10B9.xdsl.xnet.co.nz) (Ping timeout)
# [23:32] <taras> bz: i'm trying to parse your feedback
# [23:32] * Quits: ekr (ekr@moz-7B0110AD.mv.mozilla.com) (Quit: ekr)
# [23:33] * Joins: ekr (ekr@moz-7B0110AD.mv.mozilla.com)
# [23:33] <taras> bz: see etherpad
# [23:33] <taras> and the why there
# [23:33] * Quits: ekr (ekr@moz-7B0110AD.mv.mozilla.com) (Quit: ekr)
# [23:33] <@bz> ah
# [23:33] <@bz> so
# [23:33] <@bz> the issue with flushes
# [23:34] <@bz> is that if we flush multiple times
# [23:34] <@bz> and dirty stuff in between
# [23:34] <@bz> then we can end up doing work multiple times that could otherwise have been coalesced
# [23:34] * Joins: artur (artur@moz-6DFBD742.hsd1.vt.comcast.net)
# [23:34] * Quits: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [23:34] <taras> bz: is that re last point?
# [23:34] <@bz> so from the point of view of layout, the most performant usage pattern for consumers is:
# [23:34] <@bz> yes
# [23:34] <taras> ok
# [23:34] <@bz> 1) Get whatever layout info you need to compute stuff
# [23:35] <@bz> 2) Compute stuff and make all your changes without asking for more data
# [23:35] <@bz> (no step 3; then wait for the refresh tick)
# [23:35] <@bz> but in our case, even the "Get whatever layout info you need" step might be happening too often
# [23:35] <taras> bz: would help if we could do transactions
# [23:35] <taras> and have an exception thrown
# [23:35] <taras> on accidental reflows
# [23:35] <@bz> transactions in what sense?
# [23:35] <taras> err
# [23:35] <taras> flushes
# [23:36] * Quits: Ally_ (textual@moz-26DFFD6.range109-152.btcentralplus.com) (Quit: Textual IRC Client: http://www.textualapp.com/)
# [23:36] <@bz> So as a specific example
# [23:36] <taras> bz: window.executeWithoutFlushes(callback)
# [23:36] <@bz> say we can change tabs in 3ms
# [23:36] * Quits: mwu (mwu@moz-59435430.hsd1.nj.comcast.net) (Quit: This computer has gone to sleep)
# [23:36] <@bz> taras: that would not be hard to set up, actually
# [23:36] <taras> where callback would gurantee that we don't trip
# [23:36] <@bz> taras: might be hard to use, but....
# [23:36] <taras> cos debugging this stuff otherwise is nearly insane
# [23:36] <@bz> well
# [23:37] <@bz> it'll turn perf issues into exceptions
# [23:37] <@bz> right?
# [23:37] <taras> yes
# [23:37] <@bz> I guess that will make it easier to catch
# [23:37] <taras> if we had this feature, we could fix this kind of stuff quite quickly
# [23:37] <taras> across platforms
# [23:39] * Quits: gfritzsche (gfritzsche@moz-255CC203.dynamic.qsc.de) (Quit: Leaving)
# [23:39] * Quits: gozala (gozala@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: Leaving.)
# [23:39] <taras> should probably end up looking like window.assert*Stuff*(callback, [flushes, focus, whatever else we are trying to avoid])
# [23:40] * Joins: jgilbert_ (jgilbert@moz-7B0110AD.mv.mozilla.com)
# [23:40] * Quits: jgilbert (jgilbert@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
# [23:40] * Joins: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP)
# [23:40] * Joins: tonymec (tonymec@78D6A323.3F6FF523.277517C1.IP)
# [23:40] * Joins: gozala (gozala@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [23:41] <@bz> file bugs, please
# [23:41] * Quits: gozala (gozala@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: Leaving.)
# [23:41] <@bz> this could be done
# [23:41] <@bz> taras: I don't understand your GC question
# [23:41] <taras> bz: ok, i would appreciate a proof of concept patch someone could wrap up for you
# [23:41] * Quits: @smaug (chatzilla@moz-AD13B184.elisa-mobile.fi) (Ping timeout)
# [23:42] <taras> bz: it doesn't seem like there should be much js code run on tab switch
# [23:42] <taras> so why is there so much GC
# [23:42] * Joins: nrc (nrc@538BABFE.A073F3E.97BBD552.IP)
# [23:42] * Joins: smaug (chatzilla@moz-AD13B184.elisa-mobile.fi)
# [23:42] * ChanServ sets mode: +o smaug
# [23:42] * Quits: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [23:42] <@bz> there's a fair amount of code
# [23:43] <@bz> a bunch of events get fired, and their event handlers run, and some of them fire other events
# [23:43] <@bz> there is also, at least on mac, reframing of the <tab>
# [23:43] <@bz> which would reexecute the XBL binding stuff for it....
# [23:43] * Joins: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP)
# [23:44] * Quits: dao (dao@moz-95A844B3.superkabel.de) (Quit: Leaving.)
# [23:44] * Joins: gozala (gozala@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
# [23:44] * Quits: dseif_ (dseif@moz-57B23647.senecac.on.ca) (Input/output error)
# [23:45] * Joins: Yoric (Yoric@moz-920DB13B.fbx.proxad.net)
# [23:45] * Parts: nthomas (chatzilla@moz-A9A754FF.dsl.telstraclear.net)
# [23:45] * Joins: nthomas (chatzilla@moz-A9A754FF.dsl.telstraclear.net)
# [23:45] <bholley> bz: do I have your blessing to land CPG tomorrow morning given that it's a negative diff?
# [23:46] <bholley> (or at least a neutral one)
# [23:46] <@bz> hmm?
# [23:46] <@bz> bholley: why do you need my blessing?
# [23:46] <bholley> bz: mc is approval-required
# [23:46] <bholley> bz: khuey and ehsan aren't around - not sure who else to ask
# [23:46] * @bz is probably not the right person
# [23:46] <@bz> mail khuey+ehsan?
# [23:47] <bholley> bz: ok
# [23:47] * @bz is pretty sure they will still check mail tonight
# [23:47] <taras> bz: filed https://bugzilla.mozilla.org/show_bug.cgi?id=750892
# [23:47] * Quits: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
# [23:47] <@bz> also, which timezone are you in?
# [23:47] * Quits: twi (Adium@moz-5390D98D.cust.dsl.vodafone.it) (Quit: Leaving.)
# [23:47] <@bz> that is, when is "tomorrow morning"?
# [23:48] <bholley> bz: europe
# [23:48] * Quits: florian (Instantbir@moz-87C33FDA.kimsufi.com) (Quit: Instantbird -- http://www.instantbird.com)
# [23:48] * Joins: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP)
# [23:49] <@bz> bholley: ok. So yeah, I bet khuey will check mail in the next 8 hours or so
# [23:49] <bholley> bz: alright, sending mail. Thanks :-)
# [23:49] * Quits: Yoric (Yoric@moz-920DB13B.fbx.proxad.net) (Ping timeout)
# [23:51] * Quits: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Connection reset by peer)
# [23:51] * Joins: mkaply (Earlybird@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
# [23:51] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
# [23:51] * Joins: brendan (brendaneic@22E72D6E.615B56E6.43362C16.IP)
# [23:52] * Joins: roc (chatzilla@538BABFE.A073F3E.97BBD552.IP)
# [23:52] * ChanServ sets mode: +o roc
# [23:52] <@bz> ok
# [23:52] <@bz> dinnertime
# [23:52] * bz is now known as bz_dinner
# [23:52] * Quits: jfkthame (jfkthame@15AA2040.A1C12133.9542EC20.IP) (Quit: jfkthame)
# [23:53] * Joins: Cwiiis (cwiiis@moz-F15E698.croy.cable.virginmedia.com)
# [23:53] * Joins: dseif (dseif@78676049.531FD64C.630E4E47.IP)
# [23:53] * Quits: IRCMonkey55958 (chatzilla@moz-8D46E76F.sbr802.nsw.optusnet.com.au) (Quit: ChatZilla 0.9.88.1 [SeaMonkey 2.8/20120312220748])
# [23:54] * Joins: dao (dao@moz-95A844B3.superkabel.de)
# [23:56] * Quits: hessam (hessam@45968C53.9F7864B8.610F0005.IP) (Ping timeout)
# [23:57] * joduinn-mtg is now known as joduinn
# [23:58] * Quits: automata (automata@8D23278A.C27CA109.16867D26.IP) (Ping timeout)
# [23:58] * Joins: JeroenDeDauw (j@moz-15471F57.dip.t-dialin.net)
# [23:58] * Quits: anky (anky@2935FD87.5111A3CC.A3D1B221.IP) (Client exited)
# Session Close: Wed May 02 00:00:22 2012

The end :)