/irc-logs / mozilla / #developers / 2011-12-19 / end
Options:
- # Session Start: Mon Dec 19 00:00:00 2011
- # Session Ident: #developers
- # [00:00] <philor> and blaming it on that test started in comment 4, dunno why nobody has ever, you know, blamed it on that test
- # [00:00] <khuey> you want him to do that to you? :-)
- # [00:00] <khuey> you should be thankful he didn't
- # [00:00] <mcsmurf> heh, tried to open about:memory => browser crashes
- # [00:00] <ttaubert> khuey: I am, kind of :)
- # [00:00] <mcsmurf> time to make a new build I guess (my build is already a bit older)
- # [00:01] <philor> actually, we're bad enough on Linux64 right now I could probably see what happens on try, shouldn't take more than 10 to feel confident
- # [00:01] <khuey> ooh
- # [00:01] <khuey> now it's XHR doing main thread i/o
- # [00:01] * Quits: Sander (chatzilla@moz-B871F4D3.direct-adsl.nl) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
- # [00:03] <ttaubert> philor: if that solves the problem feel free to disable it and open a bug that says we should completely rewrite that test, that's a pretty old test...
- # [00:04] <smaug> khuey: what?
- # [00:05] <smaug> khuey: or XHR using necko which does i/o
- # [00:05] <khuey> smaug: xhr.send(file) stats the file file on the main thread before returning
- # [00:05] <smaug> khuey: can I blame you :)
- # [00:06] * Quits: ehsan (ehsan@F0B20A8D.8458880F.57F33CED.IP) (Input/output error)
- # [00:06] <khuey> smaug: no
- # [00:06] <jduell> khuey: file a bug for the stat
- # [00:06] <khuey> jduell: I'm filing them
- # [00:06] <smaug> khuey: where do we handle file in XHR
- # [00:06] <khuey> I've already done 3
- # [00:06] <khuey> smaug: we don't explicitly
- # [00:07] * heycam|away is now known as heycam
- # [00:07] <khuey> smaug: we call ->Available on the postdata input stream
- # [00:07] <khuey> and if that includes a file ...
- # [00:07] * Joins: dao1 (dao@moz-8103251.superkabel.de)
- # [00:07] * Quits: dao (dao@moz-8103251.superkabel.de) (Ping timeout)
- # [00:10] <khuey> jduell: 711881 if you're interested
- # [00:10] <smaug> khuey: tricky case, since progress events need that value
- # [00:10] <khuey> smaug: yeah
- # [00:11] <khuey> smaug: I think Necko needs to hand us back that value somehow
- # [00:11] <jduell> what progress events?
- # [00:11] <smaug> khuey: loadstart event happens synchronously
- # [00:11] <smaug> jduell: DOM Progress events
- # [00:11] <khuey> smaug: uh, really?
- # [00:11] <khuey> that's bad design :-(
- # [00:11] <smaug> loadstart, progress, loadend etc
- # [00:12] <khuey> loadstart is really fired during the send call?
- # [00:12] <smaug> khuey: well, progress events predate any blob/file sending
- # [00:12] <khuey> sure
- # [00:12] * Joins: Mook (mook@68E3C9C2.16C74E88.6F478678.IP)
- # [00:12] <khuey> but synchronously dispatching events is still kind of messy
- # [00:12] <khuey> at least IMO
- # [00:13] <Waldo> yess
- # [00:13] <smaug> khuey: file a spec bug. Currently XHR spec says in send() "fire a progress event named loadstart."
- # [00:14] <khuey> smaug: you can take solace in the fact that if XHR is fixed Necko causes a stat a few instructions later :-)
- # [00:14] <Waldo> heh
- # [00:14] <jduell> hey, but that stat is cached :)
- # [00:14] * khuey is enjoying his main thread i/o assert
- # [00:14] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [00:14] <khuey> jduell: for now ;-)
- # [00:14] * Waldo imagines jduell in a buddy christ pose
- # [00:15] * jduell just learned about buddy christ from google images. Thanks, bro!
- # [00:15] <Waldo> *buddychrist*
- # [00:16] <khuey> hah
- # [00:16] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [00:16] <Waldo> if you haven't seen the movie dogma, it is cringeworthily worthwhile
- # [00:17] * njn agrees with the cringe
- # [00:17] * Joins: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com)
- # [00:18] * Quits: Waldo (waldo@moz-104CC309.mv.mozilla.com) (Quit: brb, likely)
- # [00:18] * Quits: dao1 (dao@moz-8103251.superkabel.de) (Quit: Leaving.)
- # [00:19] * Joins: mbrubeck (mbrubeck@moz-755AD63.hsd1.wa.comcast.net)
- # [00:19] * Joins: Waldo (waldo@moz-534B4EF1.hsd1.mi.comcast.net)
- # [00:21] <khuey> smaug: hmm, part of me thinks that XHR should get the length from the File object
- # [00:21] <khuey> and then we should fix Files to always know their size
- # [00:21] <smaug> yeah
- # [00:23] <khuey> this probably requires some changes to our internal plumbing too
- # [00:23] * Quits: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com) (Ping timeout)
- # [00:23] <mbrubeck> woot, I can connect to tbpl again
- # [00:25] <smaug> khuey: so is the problem with nsIDOMFile and nsIFormData ?
- # [00:25] * smaug can't remember what all XHR can send
- # [00:25] * Quits: pnemsak_ (Miranda@moz-7409BD53.orange.sk) (Quit: pnemsak_)
- # [00:26] * Parts: klugefoo (kahr@moz-F88DFDF5.lightspeed.sntcca.sbcglobal.net)
- # [00:27] <khuey> smaug: it's with nsIDOMFile, yes
- # [00:27] <khuey> or an nsIFormData that includes nsIDOMFiles
- # [00:28] * Joins: kaze (kaze@moz-6386E8B2.w90-9.abo.wanadoo.fr)
- # [00:29] <khuey> taras: when are snappy meetings?
- # [00:30] <khuey> taras: or maybe I should just talk to you one on one about how to implement some stuff :-)
- # [00:30] <njn> dholbert: ping
- # [00:31] <njn> "may be used uninitialised in this function [-Wuninitialized]"
- # [00:31] <njn> gotta love the consistency of the british vs. US spelling
- # [00:32] * bwinton_away is now known as bwinton
- # [00:32] * Quits: TheLink (TheLink@moz-753C1E34.pools.arcor-ip.net) (Client exited)
- # [00:33] * bwinton is now known as bwinton_away
- # [00:33] <heycam> khuey, philikon, string constants got dropped from webidl recently. some web perf specs started to use them but we convinced them otherwise. :)
- # [00:35] <heycam> khuey, philikon, thread discussing removal: http://lists.w3.org/Archives/Public/public-script-coord/2011OctDec/thread.html#msg40
- # [00:35] <khuey> heycam: sounds good to me
- # [00:35] <khuey> less work to implement in my parser
- # [00:35] <khuey> ooh
- # [00:35] <khuey> main thread i/o bug #5
- # [00:35] <khuey> this time it's sicking's fault
- # [00:36] <jduell> khuey: how are detecting the main thread IO? I should probably do the same for blocking I/O on the socket transport thread
- # [00:38] <khuey> jduell: well I was only interested in cases where we're setting DEFER_OPEN on file streams and then causing main thread stats, so I stuck a conditional abort there
- # [00:38] <khuey> probably not generally applicable
- # [00:38] <khuey> but I've found plenty of bugs!
- # [00:38] <jduell> indeed
- # [00:39] <khuey> jduell: taras might have tools, idk
- # [00:39] <philor> mmm, and there's another of the failures from the m-c nightly, only on inbound now
- # [00:39] <khuey> jduell: also if we're not supposed to do blocking i/o on hte socket thread you should probably r- my patch for 711877 ;-)
- # [00:39] * Quits: bwinton_away (bwinton@D7D3C4B5.88764A66.72A31D6.IP) (Quit: Coyote finally caught me)
- # [00:40] <jduell> khuey: well, it's not ideal, but a stat is probably not the first thing to go after. Right now we block the thread reading from files we're uploading, which is worse.
- # [00:43] <khuey> ah
- # [00:43] <khuey> yeah that's worse than a stat ;-)
- # [00:43] <khuey> jduell: does that mean that uploading a file effectively blocks other network traffic?
- # [00:44] * Joins: asac (asac@moz-FA2AA691.adsl.alicedsl.de)
- # [00:44] * Quits: whimboo (whimboo@moz-216F9AFA.superkabel.de) (Ping timeout)
- # [00:44] * Joins: whimboo (whimboo@moz-216F9AFA.superkabel.de)
- # [00:46] <jduell> khuey: yes. for each chunk of a file we read for HTTP. For websockets it's worse right now--we block for whole file to be read--but that'll be fixed soonishly
- # [00:46] <khuey> fun
- # [00:47] * Joins: Tobbi (Tobbi@16BAC97A.933EA279.AC7F8427.IP)
- # [00:47] <jduell> khuey: see bug 690633 for the HTTP case
- # [00:47] * Tobbi is now known as IRCMonkey58367
- # [00:48] <khuey> web 6.0++
- # [00:48] <khuey> heh
- # [00:49] <jduell> khuey: the nsMultiplexInputStream::IsNonBlocking inplementation is particularly perverse IMO
- # [00:50] <khuey> yeah
- # [00:50] * Quits: florian (florian@moz-87C33FDA.kimsufi.com) (Quit: Instantbird 1.2a1pre)
- # [00:50] <khuey> I think it is the way it is because afaict everything was originally sync
- # [00:50] <khuey> and then the async stuff got special handling
- # [00:50] <khuey> but it's definitely totally backwards :-)
- # [00:51] * Quits: whimboo (whimboo@moz-216F9AFA.superkabel.de) (Ping timeout)
- # [00:51] <njn> Waldo: ping
- # [00:51] <Waldo> njn: pong
- # [00:51] <njn> Waldo: any idea how to handle this:
- # [00:51] * Joins: whimboo (whimboo@moz-216F9AFA.superkabel.de)
- # [00:51] <njn> warning: format ‘%llx’ expects argument
- # [00:51] <njn> of type ‘long long unsigned int’, but argument 3 has type ‘nsFrameState {aka long unsigned int}’ [-Wformat]
- # [00:51] <njn> nsFrameState is a PRUint64
- # [00:51] <njn> :(
- # [00:52] <njn> which is |long unsigned int| on 64-bits, apparently
- # [00:52] <Waldo> njn: I have a partial patch to add mozilla/IntTypes.h, which would address that so long as the argument passed in is one of the <stdint.h> types
- # [00:52] <njn> Waldo: by getting rid of PRUint64?
- # [00:52] <Waldo> njn: other than that, I think the claimed answer is that NSPR has its own printf thing with its own printf macros of some sort
- # [00:52] <njn> or redefinig it?
- # [00:53] <njn> ugh
- # [00:53] <Waldo> njn: you'd have to pass uint64_t for the argument
- # [00:53] <derf> njn: For now, you can just cast the PRUint64 to unsigned long long.
- # [00:53] <Waldo> basically <inttypes.h> is the way you're supposed to do this
- # [00:53] <derf> It's guaranteed to be at least 64 bits by C99.
- # [00:53] <njn> derf: ok... sigh
- # [00:53] <Waldo> really printf is just bad design
- # [00:54] <derf> The C type system is bad design.
- # [00:54] <derf> But at least it's not as bad as the C++ type system.
- # [00:54] <Waldo> and this macro junk is just a hackaround
- # [00:54] <Waldo> :-)
- # [00:54] * Quits: jhopkins (jhopkins@moz-41E1D586.tb.shawcable.net) (Ping timeout)
- # [00:54] * Joins: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com)
- # [00:54] <njn> derf: do you know why gcc complains if you pass a non-void* pointer to a %p arg in printf?
- # [00:55] <Waldo> possibly because you might have an operator void*(), but that seems wrong, because you could just look at the actual argument type to see if there is one
- # [00:55] * Joins: zandr (zandr@moz-891BD824.milewski.org)
- # [00:56] * Joins: rwaldron (rwaldron@moz-71B3012E.c3-0.abr-ubr1.sbo-abr.ma.cable.rcn.com)
- # [00:58] * Joins: birtles (chatzilla@moz-348F61F0.mozilla.or.jp)
- # [01:02] * Quits: whimboo (whimboo@moz-216F9AFA.superkabel.de) (Ping timeout)
- # [01:03] <derf> njn: Technically C allows pointers to different types to have different machine representations.
- # [01:03] <derf> And there's no implicit cast for pointers passed in variadic arguments (like there is for integer types).
- # [01:03] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [01:04] <njn> derf: egads
- # [01:04] <derf> Indeed.
- # [01:05] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [01:06] <lmandel> khuey: Snappy meetings are Thurs, 11am PST.https://wiki.mozilla.org/Performance/Snappy#Meetings
- # [01:06] * Quits: Mook (mook@68E3C9C2.16C74E88.6F478678.IP) (Quit: ChatZilla 0.9.87-rdmsoft [XULRunner 6.0/20110811165603])
- # [01:06] * Quits: bent (chatzilla@moz-C3562645.hsd1.ca.comcast.net) (Quit: ChatZilla 0.9.87-rdmsoft [XULRunner 1.9.2.19/20110707195905])
- # [01:07] * Quits: asac (asac@moz-FA2AA691.adsl.alicedsl.de) (Quit: leaving)
- # [01:07] * Joins: bsmith (bsmith@A2413EEC.695C1090.1F72B910.IP)
- # [01:09] * nthomas is now known as nthomas|away
- # [01:09] <bsmith> mayhemer: ping
- # [01:09] * Joins: KaIRC (robert@moz-CA29FA62.adsl.highway.telekom.at)
- # [01:09] * Quits: KaiRo (robert@moz-69A0E6F2.adsl.highway.telekom.at) (Ping timeout)
- # [01:10] <mayhemer> bsmith: pong
- # [01:11] * Quits: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com) (Ping timeout)
- # [01:11] <bsmith> mayhemer: Even if we create a fix/workaround for bug 711786 we still have https://bugzilla.mozilla.org/show_bug.cgi?id=704984 to worry about
- # [01:11] <bsmith> mayhemer: am I wrong in thinking that all of this network I/O was always supposed to be on the socket transport thread?
- # [01:12] <bsmith> my concern is that even if we fix these two bugs, how many more bugs are there?
- # [01:13] <mayhemer> bsmith: yes, the fix may take a lot of time, I'm worried too
- # [01:13] * Joins: fzzzy (donovan@moz-2B41AF9B.lightspeed.mtvwca.sbcglobal.net)
- # [01:13] <bsmith> mayhemer: but, I don't know if I am supposed to be working on a permanet fix, or a workaround until thunderbird can move its I/O to the socket transport thread
- # [01:13] <mayhemer> bsmith: AFAIK, socket transport and polling/reading/writting was always done on the socket thread only
- # [01:14] <mayhemer> bsmith: IsAlive seems to be an only exception
- # [01:14] * Joins: gal (gal@90E4D93F.93EC364C.1C63AAF3.IP)
- # [01:14] <bsmith> mayhemer: I wonder what happens if we just remove the IsAlive() from the IMAP code
- # [01:14] <mayhemer> bsmith: that's why I sent the email too.. I think there is no change to have a full fix in it one day
- # [01:14] <bsmith> mayhemer: it is inherently racy; even if the socket is alive at the time you call IsAlive(), it might not be by the time you send the IMAP command
- # [01:14] <mayhemer> bsmith: I don't IMAP protocol impl at all
- # [01:15] <mayhemer> bsmith: yes, the code should be able to handle restart or failure on a direct read/write failure
- # [01:15] <bsmith> mayhemer: for all practical purposes, you can think of it just like HTTP, for this discussion
- # [01:15] * Quits: gal (gal@90E4D93F.93EC364C.1C63AAF3.IP) (Quit: gal)
- # [01:16] <mayhemer> bsmith: yep, I suppose the system is quit the same
- # [01:16] <derf> In my experience the IMAP support is generally broken, and has been for as long as I've used it (though the version I run on a daily basis is about a year old now).
- # [01:16] <derf> So you can't make it worse.
- # [01:16] <mayhemer> bsmith: so, you suggest to assume the socket is always alive
- # [01:16] <bsmith> mayhemer: that would be my initial assumption, yes.
- # [01:16] <mayhemer> bsmith: I think it is the simplest solution for that bug, hopefully there is not another bug hidden under this
- # [01:17] <mayhemer> bsmith: how are we going to test this, on all platforms indeed...?
- # [01:17] <derf> And by "broken" I mean doesn't deal with connections that fail.
- # [01:17] <bsmith> mayhemer: with our comprehensive test suite, of course ;)
- # [01:17] <derf> I run gnubiff just to have something to check my mail that actually works.
- # [01:17] <derf> In terms of notifying me that new messages have arrived.
- # [01:17] <mayhemer> bsmith: hmm, be serious ;)
- # [01:18] <njn> khuey: we have NS_NOTREACHED, but it doesn't terminate execution
- # [01:18] <bsmith> mayhemer: I suggest we at lesat add assertions to the PSM NSS I/O layer to assert that it is being used on the main thread
- # [01:18] <njn> do we have something that does? NS_ABORT_IF_FALSE(false, ...) ?
- # [01:18] <bsmith> s/main thread/socket transport thread/
- # [01:18] <mayhemer> bsmith: btw, do you have time to work on this? I know you also need to fix the NSS issue
- # [01:19] <bsmith> mayhemer: which NSS issue?
- # [01:19] <bsmith> The SSL_RestartHandshakeAfterAuthCretificate issue?
- # [01:19] * Joins: bwinton_away (bwinton@D7D3C4B5.88764A66.72A31D6.IP)
- # [01:19] <khuey> njn: NS_ABORT?
- # [01:19] <bsmith> I think that isn't as urgent as Kai makes it seem
- # [01:19] <bsmith> I will email release-drivers and get confirmation that we can land the updated NSS later.
- # [01:19] <bsmith> on aurora
- # [01:19] * Joins: jhopkins (jhopkins@moz-41E1D586.tb.shawcable.net)
- # [01:19] <njn> khuey: I guess... it's |NS_DebugBreak(NS_DEBUG_ABORT, nsnull, nsnull, __FILE__, __LINE__)|, I hope GCC knows that never returns...
- # [01:19] <khuey> njn: or NS_RUNTIMEABORT if you want to die in opt builds too
- # [01:20] <khuey> njn: it doesn't :-)
- # [01:20] <bsmith> I think it will be no problem because the updated NSS will be just like the NSS in mozilla-central, except with a different version number
- # [01:20] <njn> khuey: ugh
- # [01:20] <bsmith> plus maybe very important bug fixes, if we find bugs in the implementation of SSL_RestartHandshakeAfterAuthCertificate
- # [01:20] <njn> khuey: I have these -Wuninitialized warnings caused by assigning within switch statements that don't have default cases
- # [01:20] <khuey> njn: lame
- # [01:20] <njn> GCC is technically correct
- # [01:21] <bsmith> plus a more comprehensive test suite
- # [01:21] <mayhemer> bsmith: assertions adding is nice, but be aware not to break any contracts
- # [01:21] <bsmith> mayhemer: I didn't think I was breaking any contracts already
- # [01:21] * Quits: jhopkins (jhopkins@moz-41E1D586.tb.shawcable.net) (Ping timeout)
- # [01:21] <mayhemer> bsmith: I've read email from kai about NSS version sync
- # [01:22] <bsmith> mayhemer: also, we need to make sure that Thunderbird's test suite is using IMAPS (not just IMAP), SMTPS (not just SMTP), and LDAPS (not just LDAP)
- # [01:22] <bsmith> but, actually, how does thunderbird test their IMAP/SMTP/LDAP support. What is used for the server.
- # [01:22] <bsmith> ?
- # [01:22] <NeilAway> bsmith: iirc they have a server written in JS
- # [01:23] <NeilAway> at least, for imap/smtp, I don't know of any ldap tests
- # [01:23] * rail is now known as rail_away
- # [01:24] <mayhemer> bsmith: comm-central\mailnews\test\fakeserver\
- # [01:24] <bsmith> mayhemer: NeilAway: Thanks
- # [01:24] <bsmith> mayhemer: I am in Beijing right now
- # [01:24] <bsmith> mayhemer: and I fly back today
- # [01:24] <mayhemer> bsmith: LDAP seems untested
- # [01:24] <mayhemer> bsmith: I don't think there is SSL support, at the first sight
- # [01:24] <bsmith> mayhemer: so, basically, no internet for 16 hours
- # [01:25] * Quits: coop|away (Chris@moz-201D1F65.dsl.bell.ca) (Quit: coop|away)
- # [01:25] <mayhemer> bsmith: so, looks like someone else should do this
- # [01:25] <mayhemer> bsmith: I quit for today soon (1:20 AM here, hungry, tired)
- # [01:26] <bsmith> mayhemer: things we should try: (a) remove IsAlive call from IMAP, (b) hack PSM's PSMRecv to pass PEEK calls to underlying TCP socket
- # [01:26] <bsmith> mayhemer: I will comment in the bugs
- # [01:26] <bsmith> mayhemer: But, we still have the issue of the LDAP code calling PR_Close() on the main thread
- # [01:26] <mayhemer> bsmith: why (b) if we have (a) ?
- # [01:26] <bsmith> mayhemer: only (b) if we can't do (a)
- # [01:26] <mayhemer> bsmith: I have spent some time on LDAPS and I also seems to have some test env for it
- # [01:26] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [01:27] * Quits: grubshka (grubshka@moz-498626E5.w86-216.abo.wanadoo.fr) (Ping timeout)
- # [01:27] <mayhemer> bsmith: aha, I still think we should first try (b) to be safer from hidden bugs (we lack test for IMAPS)
- # [01:27] <mayhemer> bsmith: I can test manually with some server, but how the hell shuold I drop the connections...
- # [01:28] <bsmith> mayhemer: come to China. They will drop your connections for you
- # [01:28] <mayhemer> :D:D
- # [01:28] <bsmith> mayhemer: seriaously though...
- # [01:28] <mayhemer> I know..
- # [01:29] <mayhemer> and how have you been in China btw?
- # [01:29] <bsmith> I would rather come back during warmer weather. But, it was good working with Mozilla China guys
- # [01:29] <bsmith> It seems like *every* site here is using plugins and/or ActiveX
- # [01:30] <bsmith> so, that pretty much sucks
- # [01:31] <mayhemer> bsmith: I've header something liek that... bit crazy in this era
- # [01:31] * Quits: mcsmurf (mcsmurf@moz-BE426369.dip.t-dialin.net) (Quit: )
- # [01:31] <bsmith> There are many sites that are IE6-only
- # [01:31] <bsmith> not just IE-only
- # [01:31] <mayhemer> bsmith: so, let's make some deal, I can work on both those bugs, but I still have SPDY review to be done
- # [01:31] <bsmith> but IE6-only
- # [01:31] <mayhemer> bsmith: breath of history :)
- # [01:33] <bsmith> mayhemer: if you want to try (b), maybe ifdef it to mailnews only
- # [01:34] <bsmith> mayhemer: or, at least, "not Firefox"
- # [01:34] <mayhemer> bsmith: my idea was fd = fd->lower only if (SSL && !socketTread)
- # [01:34] <bsmith> mayhemer: OK
- # [01:35] <bsmith> but, i think you need fd->lower->lower
- # [01:35] <mayhemer> bsmith: no
- # [01:35] <bsmith> because you need to skip the PSM layer and the SSL layer
- # [01:35] <mayhemer> bsmith: I need PR_GetIndentitiesLayer
- # [01:35] <bsmith> oh, because the SSL layer is thread-safe, you probably don't need to skip it
- # [01:35] <bsmith> oh, wrong again.
- # [01:36] <bsmith> you do
- # [01:36] <bsmith> to avoid the deadlock. dumb me
- # [01:36] <mayhemer> bsmith: yep, I need the tcp socket, I actually want to do what is already tested as working, IMAP is doing it all times with non-secure connections
- # [01:37] <bsmith> mayhemer: OK, that should take care of IMAP for now
- # [01:37] <bsmith> mayhemer: but, what about LDAPS
- # [01:37] <mayhemer> bsmith: we are dep on an LDAP library that we hook to
- # [01:38] <bsmith> AFAICT, the problem is that the socket should be closed by the socket transport thread before it is Released()
- # [01:38] <mayhemer> bsmith: it does the reading and the code is even worse reading then NSS code...
- # [01:38] <bsmith> But, instead, they rely on the destructor to close it
- # [01:38] <mayhemer> bsmith: the code is not that simple
- # [01:39] <mayhemer> bsmith: however, unbind could be potentially posted to socket thread
- # [01:39] <mayhemer> bsmith: also, bind could be as well
- # [01:39] <bsmith> mayhemer: are you saying that not just the close is done off the socket transport thread?
- # [01:40] <mayhemer> bsmith: problem is that LDAP seems to be interacting with the UI in very sync fashin, so not easy to separete
- # [01:40] <mayhemer> bsmith: I think all operations are made on the main thread
- # [01:40] <mayhemer> bsmith: including the socket creation
- # [01:40] <bsmith> :(
- # [01:40] <mayhemer> yeah..
- # [01:40] <bsmith> How could it ever work
- # [01:40] <bsmith> ?
- # [01:40] <mayhemer> ssl thread?
- # [01:41] <mayhemer> and proxies...
- # [01:41] <bsmith> But what about the non-SSL case?
- # [01:41] <mayhemer> then are not using socket transport and the service
- # [01:41] <bsmith> :(
- # [01:41] <mayhemer> the LDAP lib is polling it self
- # [01:41] <mayhemer> we just provide a virtual API impl for sockets
- # [01:42] <mayhemer> mayhemer: but I have to check on that ones more...
- # [01:42] <bsmith> mayhemer: then, we have to change SyncRunnableBase
- # [01:42] * Joins: DGMurdockIII (dgmurdocki@moz-E933B63.hsd1.il.comcast.net)
- # [01:43] <bsmith> so that it just calls the Run() method on the same thread, if it is being called on the main thread
- # [01:43] * Joins: nattokirai (nattokirai@moz-348F61F0.mozilla.or.jp)
- # [01:43] <bsmith> But, I don't know all the consequences of that...
- # [01:44] <bsmith> I remember writing so many comments: "this is OK because we're on the socket transport thread"
- # [01:44] <bsmith> "don't worry, we're on the socket transport thrad"
- # [01:44] <bsmith> "nothing can go wrong, because we're on the socket transport thread"
- # [01:45] <mayhemer> bsmith: break: here - https://bug704984.bugzilla.mozilla.org/attachment.cgi?id=581743 - it seems that the second socket on thread 4 is not even a socket used for LDAP
- # [01:45] <mayhemer> bsmith: ah! it might be, because you are posting the cert verifycation event to the socket thread
- # [01:46] <bsmith> mayhemer: probably, we would have to change the code to post the events to the thread that the I/O is happening on, instead of always the socket transport thread
- # [01:47] <mayhemer> bsmith: yep, but then also change the SyncRunable
- # [01:47] <bsmith> mayhemer: but, this kind of work is totally backwards, writing more code to support Inetwork /O on the main thread
- # [01:47] * Quits: bjacob (bjacob@moz-ADCA75DC.cpe.net.cable.rogers.com) (Quit: Konversation terminated!)
- # [01:48] <mayhemer> bsmith: I'll take a look on how much LDAP iteracts with the UI
- # [01:48] <mayhemer> bsmith: after all, it's just a download of a list
- # [01:49] * Quits: DGMurdockIII (dgmurdocki@moz-E933B63.hsd1.il.comcast.net) (Quit: get satisfied! :: core-networks.de ««« (Gamers.IRC) »»» gamersirc.net ::)
- # [01:49] <bsmith> Is the thunderbird merge schedule the same as Firefox's?
- # [01:49] <bsmith> Do they merge Tuesday like us?
- # [01:50] <bsmith> I guess they are us.
- # [01:50] * Joins: northWind (northWind@moz-67A4DECF.airyork.yorku.ca)
- # [01:50] <khuey> we're all on the thunderbird team
- # [01:51] <bsmith> khuey: yes, I know. I wish I could erase that
- # [01:51] <khuey> that was supposed to be a joke
- # [01:51] <khuey> did I miss a memo? :-P
- # [01:53] <bsmith> The answer appears to be yes, they merge the same day as us
- # [01:53] <bsmith> at lesat, they did for August 16th
- # [01:54] * Quits: jduell (jduell@moz-A446278F.hfc.comcastbusiness.net) (Ping timeout)
- # [01:54] <bsmith> mayhemer: the other option would be to revert comm-central's PSM to pre-XPCOM-proxy removal
- # [01:55] <bsmith> mayhemer: especially since they don't need SPDY
- # [01:55] <mayhemer> bsmith: that occured to me too
- # [01:56] <mayhemer> actually, we can change client.py that comm-central will checkout on a given revision
- # [01:56] <mayhemer> one way
- # [01:56] <khuey> that would freeze the rest of m-c for them too
- # [01:56] <khuey> they might not like that
- # [01:56] <mayhemer> khuey: true
- # [01:56] <philor> :)
- # [01:57] <philor> solves all those "do we need to do a chemspill release for this change?" questions pretty neatly, though
- # [01:58] <bsmith> mayhemer: we might be able to extend the SSL thread removal and XPCOM proxiy removal patchee to handle this LDAP code
- # [01:58] <bsmith> but, not in one day or two days
- # [01:58] <bsmith> Would they take such a fix on their Earlybird channel?
- # [02:00] <bsmith> mayhemer: the main question is, do we want to support this non-STS networking long-term
- # [02:01] <bsmith> mayhemer: Also, remember that I initialize the SSL cert verification service in the STS thread too :) So many things to go wrong
- # [02:01] <mayhemer> bsmith: that is not a question for today's date
- # [02:01] <mayhemer> bsmith: yeah, we had to fix that too...
- # [02:01] <bsmith> mayhemer: well, it is a significant amount of work either way
- # [02:02] <bsmith> so, we should choose to either move the LDAP code to the STS thread or make PSM work off the STS thread just for LDAP
- # [02:02] <mayhemer> bsmith: we can go with LDAP being broken in Aurora theoretically
- # [02:02] <bsmith> my vote would be to fix the LDAP code so that it runs on the STS thread. Don't know how though
- # [02:02] <mayhemer> bsmith: I think it will not be possible, but I'll check
- # [02:03] <mayhemer> bsmith: building comm-central now
- # [02:04] <mayhemer> bsmith: actually, I'm not sure how the LDAP code ensure NSS is up, vtw ;)
- # [02:04] <mayhemer> btw
- # [02:05] <mayhemer> bsmith: I think we should also seriously consider telling the world "LDAP is broken in TB Aurorar"
- # [02:05] * nthomas|away is now known as nthomsa
- # [02:05] * Quits: mike5w3c (MikeS@moz-58F37FDB.pool.e-mobile.ne.jp) (Ping timeout)
- # [02:05] * nthomsa is now known as nthomas
- # [02:06] <mayhemer> and fix it first on m-c/c-c and then merge to Aurora as an important stability fix
- # [02:07] <bsmith> mayhemer: Look here: http://mxr.mozilla.org/comm-central/source/ldap/xpcom/src/nsLDAPConnection.cpp#366
- # [02:07] <bsmith> it looks like the code is already doing most operations on a background thread
- # [02:07] <bsmith> I wonder if it could just be changed to do them on the STS thread
- # [02:07] <mayhemer> nice
- # [02:08] <mayhemer> bsmith: I'll investigate
- # [02:08] <bsmith> mayhemer: we would still need to move the Close() out of the destructor though
- # [02:09] <mayhemer> bsmith: proxyrelease?
- # [02:09] <bsmith> mayhemer: perhaps, I have never used it
- # [02:09] <bsmith> mayhemer: I think we should try this approach first
- # [02:09] <bsmith> to change the LDAP code. It looks doable to me
- # [02:10] <mayhemer> bsmith: I belive there will be other bugs, I'll try that right now
- # [02:10] <Asa> how does one pronounce Ms2ger?
- # [02:10] <bsmith> mayhemer: OK. I gotta go to the office now. ttyl
- # [02:10] <mayhemer> bsmith: I have no way to reproduce instead of instrumenting the code...
- # [02:10] <mayhemer> bsmith: see you
- # [02:12] <heycam> Asa, of the various pronunciations I've heard, "miz too ger" is the one find the easiest
- # [02:12] <Asa> heycam: thanks!
- # [02:13] * Quits: sfink (chatzilla@moz-4EAF93BB.dsl.pltn13.sbcglobal.net) (Ping timeout)
- # [02:14] * bwinton_away is now known as bwinton
- # [02:14] <bsmith> mayhemer: the LDAP code should be closing the connection in its "profile-change-net-teardown" observer
- # [02:14] <bsmith> mayhemer: I bet that is not hapening
- # [02:14] <bsmith> happening
- # [02:14] <bsmith> just FYI
- # [02:15] <mayhemer> bsmith: sight
- # [02:15] * bwinton is now known as bwinton_away
- # [02:16] * Joins: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com)
- # [02:18] * Quits: bsmith (bsmith@A2413EEC.695C1090.1F72B910.IP) (Ping timeout)
- # [02:18] * Ziggy_Maes is now known as Ziggy|AWAY
- # [02:21] <mattwoodrow> khuey: ping
- # [02:24] <smaug> ejpbruel: just curious, what is a fake frame?
- # [02:24] <ejpbruel> smaug: oh, thats gonna be fun :)
- # [02:25] * Joins: kanru (user@moz-BA086EAD.dynamic.hinet.net)
- # [02:25] <ejpbruel> smaug: so, what the jetpack guys want to be able to do is create hidden windows on the fly
- # [02:25] <smaug> ok, so it is about frame elements and such
- # [02:25] <ejpbruel> smaug: yeah
- # [02:26] <smaug> not about nsIFrame or any other "frame" we have :)
- # [02:26] <ejpbruel> smaug: so for a lot of things they do in the add-on sdk they require a hidden window. firefox has only one such window, so there are many possibilites for conflict there.
- # [02:26] <ejpbruel> smaug: no, i dont think so
- # [02:27] <ejpbruel> smaug: basically, what we want is an object that behaves as a hidden window, with all the API's normally available on it, without necessarily having it implemented as an native OS window
- # [02:27] * Joins: jdm (jdm@moz-9AEDE212.cpe.net.cable.rogers.com)
- # [02:28] <ejpbruel> smaug: i might actually get back to you on this this week :)
- # [02:29] <smaug> k
- # [02:29] <smaug> good night
- # [02:30] * Joins: jhopkins (jhopkins@moz-41E1D586.tb.shawcable.net)
- # [02:30] * Quits: Waldo (waldo@moz-534B4EF1.hsd1.mi.comcast.net) (Quit: ChatZilla 0.9.87-4.1450hg.fc15 [XULRunner 8.0/20111108090029])
- # [02:31] <ejpbruel> night
- # [02:31] * Parts: ejpbruel (ejpbruel@933E3A2E.83A511AD.F413BF4D.IP)
- # [02:31] * Quits: smaug (chatzilla@moz-57D0D6F1.elisa-mobile.fi) (Ping timeout)
- # [02:31] * Quits: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP) (Quit: jfkthame)
- # [02:32] * Quits: jhopkins (jhopkins@moz-41E1D586.tb.shawcable.net) (Ping timeout)
- # [02:37] * Joins: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP)
- # [02:39] * Quits: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP) (Quit: jfkthame)
- # [02:40] * Joins: jhopkins (jhopkins@moz-41E1D586.tb.shawcable.net)
- # [02:40] * Quits: kaze (kaze@moz-6386E8B2.w90-9.abo.wanadoo.fr) (Ping timeout)
- # [02:45] * Joins: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com)
- # [02:46] * Joins: janv (varga@moz-C1261AFF.flarion.as5628.telecom.sk)
- # [02:47] * Joins: kaie2 (kaie@moz-F1678BAE.dip.t-dialin.net)
- # [02:48] * Quits: kaie (kaie@moz-53593440.dip.t-dialin.net) (Ping timeout)
- # [02:48] * kaie2 is now known as kaie
- # [02:49] * Joins: m_kato (m_kato@moz-348F61F0.mozilla.or.jp)
- # [02:52] * Quits: janv (varga@moz-C1261AFF.flarion.as5628.telecom.sk) (Quit: This computer has gone to sleep)
- # [02:53] <mbrubeck> njn's push to inbound has another weird cluster of failures just like the one from the last night's m-c nightly build
- # [02:53] <mbrubeck> (and why did it get two sets of Linux debug tests, anyway?)
- # [02:56] <nthomas> two sets of tests if normal on the rev that is used for a nightly
- # [02:56] <nthomas> s/if/are/
- # [02:56] <nthomas> got to build it to be a nightly, and nightlies trigger tests the same as dep builds
- # [02:57] <khuey> mattwoodrow: pong
- # [02:57] <mattwoodrow> khuey: Do you have any idea why operator delete() from within skia code (on android) would be going through libstdc++ instead of mozalloc? Or know who I could ask that might know. I'm assuming its a build/link problem
- # [02:57] * Joins: ewong|afk (chatzilla@moz-83240463.belfordhk.com)
- # [02:58] <khuey> mattwoodrow: it's almost certainly a linking problem
- # [02:58] <khuey> mattwoodrow: glandium is probably the best person to ask
- # [02:58] <khuey> it's 2 am for him right now though iirc
- # [02:58] * ewong|afk is now known as ewong
- # [02:58] <mattwoodrow> alright, I'll try him tomorrow then
- # [02:58] <mattwoodrow> thanks khuey
- # [02:59] <philor> nthomas: gotta be ye olde race condition, though, for a debug build in the middle of the afternoon
- # [03:01] * nthomas bumbles around trying to make sense of philor
- # [03:01] <philor> nthomas: https://tbpl.mozilla.org/?tree=Mozilla-Inbound&onlyunstarred=1&rev=7b4007e3cafa has the doubled-tests in question, debug tests, so not a nightly
- # [03:02] <Unfocused> understanding philor is hard, lets go shopping
- # [03:02] <nthomas> ah, I get it
- # [03:02] <nthomas> Unfocused++
- # [03:02] <jbuck> firebot: uuid
- # [03:02] <firebot> 390c059a-0a26-4a44-96b6-3f8817bf92e9 (/msg firebot cid for CID form)
- # [03:02] <jbuck> firebot: uuid
- # [03:02] <firebot> 2274055b-8b3a-4a5a-8d72-5d5aea07021a (/msg firebot cid for CID form)
- # [03:03] * Joins: mike5w3c (MikeS@moz-50169F6C.tokyo.ocn.ne.jp)
- # [03:03] <northWind> lol that's useful
- # [03:03] * Quits: jhopkins (jhopkins@moz-41E1D586.tb.shawcable.net) (Input/output error)
- # [03:04] * Joins: akeybl (akeybl@moz-52D39FF6.hsd1.ca.comcast.net)
- # [03:04] * Quits: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com) (Ping timeout)
- # [03:05] * philor hopes nobody's going to need anything from the testpool for a little while
- # [03:07] * Unfocused looks anxiously over to his Try run
- # [03:08] <philor> except for Windows, I'm not using all of them
- # [03:17] * Joins: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com)
- # [03:18] * Quits: wg9s (bill@moz-7A06A043.hsd1.ma.comcast.net) (Quit: ChatZilla 0.9.87-4.1450hg.fc16 [XULRunner 8.0/20111108090055])
- # [03:20] * KaIRC is now known as KaiRo
- # [03:23] * Joins: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com)
- # [03:26] * Quits: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com) (Ping timeout)
- # [03:28] * Quits: northWind (northWind@moz-67A4DECF.airyork.yorku.ca) (Ping timeout)
- # [03:29] * khuey is now known as khuey|away
- # [03:31] * Quits: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net) (Ping timeout)
- # [03:32] <philor> hmm, are we going to ship a parsing error in ua.css for Fx10?
- # [03:32] * Joins: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net)
- # [03:33] <Unfocused> oh yea.. i keep forgetting to see if there's a bug on that
- # [03:34] * Unfocused needs to be less crazy-busy
- # [03:34] <philor> bug 706192
- # [03:35] <mayhemer> how can I create a try build of thunderbird?
- # [03:36] <philor> https://wiki.mozilla.org/Thunderbird/Infrastructure/TryServer
- # [03:36] * Quits: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com) (Input/output error)
- # [03:36] <mayhemer> c:\Mozilla\src\comm-central\.hg\patches\bugs\704984-daily-crash-ldaps.patch ?
- # [03:36] <mayhemer> hg push -f ssh://hg.mozilla.org/try-comm-central/
- # [03:36] <mayhemer> sorry for the first line..
- # [03:36] * Quits: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net) (Ping timeout)
- # [03:37] <philor> should work
- # [03:37] * Quits: chewey (chewey@moz-F0009F13.dip0.t-ipconnect.de) (NickServ (GHOST command used by chewey_))
- # [03:38] * Joins: chewey (chewey@moz-F80275DC.dip0.t-ipconnect.de)
- # [03:38] * Joins: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net)
- # [03:38] <Unfocused> *cough* use try chooser *cough*
- # [03:38] <mayhemer> thanks
- # [03:39] <philor> yeah, I like the way that page tells you to use try chooser with a link to the docs for Fx's try chooser
- # [03:40] <Unfocused> heh
- # [03:40] <philor> I'm sure it saves lots of resources to say you want your Tb build with -t none
- # [03:40] * Joins: anky (anky@37C658C.E0DAC67D.A3D1B221.IP)
- # [03:41] * Quits: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net) (Ping timeout)
- # [03:41] * Quits: IanN (chatzilla@moz-3F5A461C.cable.virginmedia.com) (Quit: ChatZilla 0.9.87 [Firefox 3.6.24/20111108102743])
- # [03:42] <philor> ah, at last, finally got some orange on try
- # [03:42] * Joins: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net)
- # [03:43] * heycam is now known as heycam|away
- # [03:43] <Unfocused> you have an odd concept of success
- # [03:43] * Quits: rshetty (quassel@3BEA28DB.503916DD.C842849F.IP) (Ping timeout)
- # [03:45] <philor> yup, but if, as they were, your experiment and your control are both green for the first 25 runs, it's hard to be persuasive that your experiment made a difference
- # [03:45] <Unfocused> ah, yes, fair enough :)
- # [03:46] <Unfocused> unless it's a refactor...
- # [03:46] * Quits: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net) (Ping timeout)
- # [03:47] <philor> yeah, but most of my experiments are for "hypothesis: if I disable test_foo, I'll be less likely to want to bite people"
- # [03:48] * Joins: northWind (northWind@moz-67A4DECF.airyork.yorku.ca)
- # [03:48] * Joins: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net)
- # [03:48] <Unfocused> heh
- # [03:49] * Quits: MichaelR (MichaelR@moz-B782A5DE.bstnma.east.verizon.net) (Ping timeout)
- # [03:53] * Joins: glob (glob@moz-DF237567.glob.com.au)
- # [03:53] * Quits: northWind (northWind@moz-67A4DECF.airyork.yorku.ca) (Quit: )
- # [03:54] * Joins: dvander`home (dvander@moz-EC56C22F.hsd1.ca.comcast.net)
- # [03:54] * Joins: rshetty (quassel@14F7D0B8.8F230C46.C842849F.IP)
- # [03:57] * Joins: josh (josh@moz-2EE66546.nyc.res.rr.com)
- # [03:57] <firebot> Check-in: http://hg.mozilla.org/mozilla-central/rev/d75ebb37080e - Phil Ringnalda - Bug 695345 - Bring back ISO8601DateUtils.jsm, a=akeybl
- # [03:58] * philor searches for an appropriate word
- # [03:59] <Unfocused> "facepalm" ?
- # [04:03] * Quits: mike5w3c (MikeS@moz-50169F6C.tokyo.ocn.ne.jp) (Quit: mike5w3c)
- # [04:03] * philor is now known as philor|away
- # [04:04] * Quits: rjohnson19 (rjohnson19@moz-9148485F.hsd1.ma.comcast.net) (Quit: Leaving)
- # [04:05] <reuben> http://dl.dropbox.com/u/10968786/Mozilla/test/test.html
- # [04:05] <reuben> does it render 'fi' for any of you?
- # [04:06] <Unfocused> nope
- # [04:06] <Unfocused> i was wondering why you were testing a blank page, til i looked at the source
- # [04:06] <reuben> weirdest bug – found it while doing a presentation @ work
- # [04:07] * reuben files
- # [04:08] <Unfocused> i bet it has something to do with ligatures
- # [04:11] <glob> hrm, it wasn't working for me, but now it is
- # [04:11] * Quits: KaiRo (robert@moz-CA29FA62.adsl.highway.telekom.at) (Input/output error)
- # [04:16] * Quits: akeybl (akeybl@moz-52D39FF6.hsd1.ca.comcast.net) (Quit: Leaving...)
- # [04:18] * Quits: mayhemer (Miranda@B3D46202.F87A741B.F23860FD.IP) (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
- # [04:20] * Joins: Ventron (michael@moz-ED780ED3.dyn.iinet.net.au)
- # [04:21] * Joins: merinui (merinui@moz-61C7235E.osk2.eonet.ne.jp)
- # [04:24] * heycam|away is now known as heycam
- # [04:27] * Joins: darktrojan (geoff@moz-BC95E278.dsl.telstraclear.net)
- # [04:28] * philor|away is now known as philor
- # [04:31] * Quits: anky (anky@37C658C.E0DAC67D.A3D1B221.IP) (Client exited)
- # [04:32] * Quits: ewong (chatzilla@moz-83240463.belfordhk.com) (Quit: BrB)
- # [04:35] * Joins: ewong (chatzilla@moz-B5A86794.com)
- # [04:35] * Joins: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw)
- # [04:40] * rail_away is now known as rail
- # [04:41] * Quits: Hendikins (wolfox@moz-9A361C95.static.internode.on.net) (Ping timeout)
- # [04:42] * Joins: Hendikins (wolfox@moz-9A361C95.static.internode.on.net)
- # [04:46] * Quits: raccettura (raccettura@moz-660B8F4B.hsd1.nj.comcast.net) (Quit: raccettura)
- # [04:46] * philor is now known as philor|away
- # [04:48] * Quits: karl (karl@538BABFE.A073F3E.97BBD552.IP) (Ping timeout)
- # [04:50] * Joins: Mook (mook@moz-B1410066.dsl.teksavvy.com)
- # [04:56] * Joins: tabb0t (tabb0t@51A1651E.7D2FD62.422D491.IP)
- # [04:57] * philor|away is now known as philor
- # [04:58] * Quits: NeilAway (neil@moz-32AA0D01.in-addr.btopenworld.com) (Ping timeout)
- # [05:02] * Joins: irving (irving@moz-92D12271.dsl.bell.ca)
- # [05:03] * Quits: irving (irving@moz-92D12271.dsl.bell.ca) (Input/output error)
- # [05:06] * Joins: squib (squib@moz-F5CA0CFB.dhcp.mdsn.wi.charter.com)
- # [05:08] * philor is now known as philor|away
- # [05:08] * Joins: cjones (cjones@BF856838.CE318AFF.4B6E8EFF.IP)
- # [05:09] * philor|away is now known as philor
- # [05:10] <nthomas> gavin: is it possible that bug 696436 has broken passing links from other applications ? My chatzilla and t'bird aren't opening links in aurora since I updated from the 2011-12-15 nightly
- # [05:12] <Unfocused> nthomas: from what i understand, that wasn't meant to change the name for existing installations, therefore it wouldn't affect stuff like that
- # [05:12] * Quits: cpearce (chatzilla@538BABFE.A073F3E.97BBD552.IP) (Ping timeout)
- # [05:13] * Quits: tabb0t (tabb0t@51A1651E.7D2FD62.422D491.IP) (Quit: leaving)
- # [05:16] * philor is now known as philor|away
- # [05:17] * Quits: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw) (Quit: regen)
- # [05:21] * nthomas is now known as nthomas|away
- # [05:22] * Joins: alex (alex@86599527.E7FBFC77.AB369DBA.IP)
- # [05:23] * Joins: mike5w3c (MikeS@moz-81CEC26.pool.e-mobile.ne.jp)
- # [05:23] * Quits: mike5w3c (MikeS@moz-81CEC26.pool.e-mobile.ne.jp) (Quit: mike5w3c)
- # [05:26] * Quits: Mook (mook@moz-B1410066.dsl.teksavvy.com) (NickServ (GHOST command used by Mook_))
- # [05:26] * Joins: Mook (mook@moz-B1410066.dsl.teksavvy.com)
- # [05:28] * Quits: rwaldron (rwaldron@moz-71B3012E.c3-0.abr-ubr1.sbo-abr.ma.cable.rcn.com) (Quit: Leaving...)
- # [05:35] * Joins: gal (gal@moz-5FD3DD7.dsl.sndg02.sbcglobal.net)
- # [05:36] * Joins: kumar (kmcmillan@moz-F2D05B8.c3-0.stk-ubr1.chi-stk.il.cable.rcn.com)
- # [05:36] * Joins: karl (karl@moz-E8CBA3A5.jetstream.xtra.co.nz)
- # [05:36] * rail is now known as rail_away
- # [05:36] * Quits: sewardj (sewardj@moz-C79D927B.cable.virginmedia.com) (Ping timeout)
- # [05:37] * Joins: akeybl (akeybl@moz-52D39FF6.hsd1.ca.comcast.net)
- # [05:38] * Joins: sewardj (sewardj@moz-C79D927B.cable.virginmedia.com)
- # [05:43] <gavin> nthomas|away: indeed, that would surprise me
- # [05:48] * Quits: bwinton_away (bwinton@D7D3C4B5.88764A66.72A31D6.IP) (Ping timeout)
- # [05:48] * Quits: cjones (cjones@BF856838.CE318AFF.4B6E8EFF.IP) (Ping timeout)
- # [05:50] * Joins: cjones (cjones@moz-B6A93570.dsl.sndg02.sbcglobal.net)
- # [05:51] * Joins: cpearce (chatzilla@moz-55A2BC7D.xdsl.xnet.co.nz)
- # [05:51] * Quits: kumar (kmcmillan@moz-F2D05B8.c3-0.stk-ubr1.chi-stk.il.cable.rcn.com) (Quit: kumar)
- # [05:55] * Quits: gal (gal@moz-5FD3DD7.dsl.sndg02.sbcglobal.net) (Quit: gal)
- # [05:55] * Quits: victorporof (victorporo@D0477DB.DA9DCE7D.79933D60.IP) (Quit: victorporof)
- # [05:59] * Joins: gal (gal@moz-5FD3DD7.dsl.sndg02.sbcglobal.net)
- # [06:00] * Quits: zandr (zandr@moz-891BD824.milewski.org) (Max SendQ exceeded)
- # [06:00] * Quits: gmoro (guilherme@3DCBFB93.F520FC74.D159334F.IP) (Ping timeout)
- # [06:03] * Joins: zandr (zandr@moz-891BD824.milewski.org)
- # [06:06] * Quits: Polynomial-C (Poly-C@moz-F63A073A.dip.t-dialin.net) (Ping timeout)
- # [06:06] * Joins: Polynomial-C (Poly-C@moz-7F02F440.dip.t-dialin.net)
- # [06:09] * Joins: biesi (cbiesinger@2E747C8D.C1499D0.396E4C6D.IP)
- # [06:09] * philor|away is now known as philor
- # [06:11] * Quits: akeybl (akeybl@moz-52D39FF6.hsd1.ca.comcast.net) (Quit: Leaving...)
- # [06:11] * Quits: darktrojan (geoff@moz-BC95E278.dsl.telstraclear.net) (Ping timeout)
- # [06:16] * Joins: mdr (chatzilla@moz-DECFDD00.phlapa.fios.verizon.net)
- # [06:17] * Quits: Polynomial-C (Poly-C@moz-7F02F440.dip.t-dialin.net) (Ping timeout)
- # [06:19] * Quits: tonymec (tonymec@moz-479B0542.adsl-dyn.isp.belgacom.be) (Ping timeout)
- # [06:19] * Quits: tonymec|away (tonymec@moz-479B0542.adsl-dyn.isp.belgacom.be) (Ping timeout)
- # [06:20] * Quits: roc (chatzilla@538BABFE.A073F3E.97BBD552.IP) (Ping timeout)
- # [06:23] * Joins: tonymec__ (tonymec@BD0053DF.4AE5B600.277517C1.IP)
- # [06:23] * tonymec__ is now known as tonymec|away
- # [06:27] * Parts: lsumar (lsumar@538BABFE.A073F3E.97BBD552.IP) (Leaving)
- # [06:37] * Joins: janv (varga@moz-C1261AFF.flarion.as5628.telecom.sk)
- # [06:39] * Quits: merinui (merinui@moz-61C7235E.osk2.eonet.ne.jp) (Quit: Leaving...)
- # [06:40] * Quits: lmandel (lmandel@moz-113D7D7C.cable.teksavvy.com) (Quit: lmandel)
- # [06:41] * Joins: AndChat| (AndChat@538BABFE.A073F3E.97BBD552.IP)
- # [06:41] * Quits: eflores (AndChat@538BABFE.A073F3E.97BBD552.IP) (Connection reset by peer)
- # [06:42] * Quits: AndChat| (AndChat@538BABFE.A073F3E.97BBD552.IP) (Ping timeout)
- # [06:44] <njn> anyone here know about the style system?
- # [06:45] * Quits: biesi (cbiesinger@2E747C8D.C1499D0.396E4C6D.IP) (Ping timeout)
- # [06:46] * Joins: sfink (chatzilla@moz-4EAF93BB.dsl.pltn13.sbcglobal.net)
- # [06:47] * Quits: mconley (mconley@A00174C9.68469654.6816E6B7.IP) (Input/output error)
- # [06:47] * Quits: Mook (mook@moz-B1410066.dsl.teksavvy.com) (Ping timeout)
- # [06:48] * Joins: Mook (mook@68E3C9C2.16C74E88.6F478678.IP)
- # [06:49] * Joins: bwinton_away (bwinton@D7D3C4B5.88764A66.72A31D6.IP)
- # [06:54] <tonymec|away> njn: glandium maybe? anyway, just ask your question then wait (maybe hours), someone may wake up and see it
- # [06:57] * Quits: kennyluck (kennyluck@moz-661C37AF.dynamic.hinet.net) (Ping timeout)
- # [07:00] * Joins: tonymec (tonymec@BD0053DF.4AE5B600.277517C1.IP)
- # [07:01] * philor is now known as philor|away
- # [07:02] * Joins: biesi (cbiesinger@2E747C8D.C1499D0.396E4C6D.IP)
- # [07:02] * Quits: Mook (mook@68E3C9C2.16C74E88.6F478678.IP) (NickServ (GHOST command used by Mook_))
- # [07:02] * Joins: Mook (mook@moz-B1410066.dsl.teksavvy.com)
- # [07:07] * Quits: bwinton_away (bwinton@D7D3C4B5.88764A66.72A31D6.IP) (Ping timeout)
- # [07:09] * Joins: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw)
- # [07:10] * Joins: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP)
- # [07:14] * Joins: darktrojan (geoff@moz-BC95E278.dsl.telstraclear.net)
- # [07:14] * Joins: dRdR (dRdR@moz-F979C50C.cpe.net.cable.rogers.com)
- # [07:15] * Quits: njn (chatzilla@32788AE8.60066DFE.8660782D.IP) (Quit: ChatZilla 0.9.87 [Firefox 11.0a1/20111212152829])
- # [07:20] * philor|away is now known as philor
- # [07:20] * sancus is now known as sancus_
- # [07:20] * sancus_ is now known as sancus
- # [07:21] * Quits: Hendikins (wolfox@moz-9A361C95.static.internode.on.net) (Ping timeout)
- # [07:22] * Joins: Hendikins (wolfox@moz-9A361C95.static.internode.on.net)
- # [07:23] * Quits: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP) (Quit: jfkthame)
- # [07:29] * Joins: grubshka (grubshka@moz-498626E5.w86-216.abo.wanadoo.fr)
- # [07:31] * glob is now known as glob|away
- # [07:34] * philor is now known as philor|away
- # [07:37] * Quits: dRdR (dRdR@moz-F979C50C.cpe.net.cable.rogers.com) (Client exited)
- # [07:37] * Quits: biesi (cbiesinger@2E747C8D.C1499D0.396E4C6D.IP) (Ping timeout)
- # [07:37] * Quits: kanru (user@moz-BA086EAD.dynamic.hinet.net) (Ping timeout)
- # [07:37] * jwir3 is now known as jwir3|away
- # [07:39] * Joins: kanru (user@moz-BA086EAD.dynamic.hinet.net)
- # [07:42] * mattwoodrow is now known as mattwoodrow|away
- # [07:42] * Joins: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP)
- # [07:49] * Joins: cpeterson (cpeterson@moz-18338A6E.hsd1.ca.comcast.net)
- # [07:49] * Quits: Mook (mook@moz-B1410066.dsl.teksavvy.com) (Ping timeout)
- # [07:50] <nthomas|away> gavin: restarted to a new profile, and back to my normal profile, and external links worked fine in both so something must have gotten confused
- # [07:50] * Quits: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP) (Ping timeout)
- # [07:50] * Joins: Mook (mook@moz-B1410066.dsl.teksavvy.com)
- # [07:51] <nthomas|away> quite possibly the meat in front of the hardware
- # [07:51] * Joins: bsmith (bsmith@8FB758AE.17E825E1.75B241AF.IP)
- # [07:51] * Quits: bsmith (bsmith@8FB758AE.17E825E1.75B241AF.IP) (Connection reset by peer)
- # [07:52] * Quits: cpeterson (cpeterson@moz-18338A6E.hsd1.ca.comcast.net) (Input/output error)
- # [07:58] <jesup> njn: bz, and probably roc. (and some others in MV)
- # [07:59] * mattwoodrow|away is now known as mattwoodrow
- # [08:06] * Joins: bsmith (bsmith@8FB758AE.17E825E1.75B241AF.IP)
- # [08:07] * glob|away is now known as glob
- # [08:09] * Quits: alex (alex@86599527.E7FBFC77.AB369DBA.IP) (Ping timeout)
- # [08:11] * Joins: wolfiR (wolfiR@moz-145F9C08.dip0.t-ipconnect.de)
- # [08:11] * Joins: sicking (chatzilla@moz-1FF05400.hsd1.ca.comcast.net)
- # [08:17] * Quits: zuzelvp (zuzelvp@2112147D.C3507A2D.9A8C35B4.IP) (Ping timeout)
- # [08:17] * Quits: karl (karl@moz-E8CBA3A5.jetstream.xtra.co.nz) (Ping timeout)
- # [08:18] * Quits: cjones (cjones@moz-B6A93570.dsl.sndg02.sbcglobal.net) (Ping timeout)
- # [08:18] * Quits: diogogmt (kvirc@moz-4D628198.cpe.net.cable.rogers.com) (Quit: KVIrc 4.1.1 Equilibrium http://www.kvirc.net/)
- # [08:19] * Quits: Mossop (mossop@moz-E325C2EC.dsl.dynamic.sonic.net) (Ping timeout)
- # [08:20] * Joins: jduell (jduell@moz-2D9EDA98.lightspeed.sntcca.sbcglobal.net)
- # [08:21] * Joins: Mossop (mossop@moz-E325C2EC.dsl.dynamic.sonic.net)
- # [08:21] * Joins: cjones (cjones@moz-B6A93570.dsl.sndg02.sbcglobal.net)
- # [08:22] * Quits: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw) (Quit: regen)
- # [08:23] * Joins: imphil (philipp@moz-655EF802.customer.m-online.net)
- # [08:23] * Joins: dRdR (dRdR@moz-F979C50C.cpe.net.cable.rogers.com)
- # [08:24] * Joins: kennyluck (kennyluck@moz-661C37AF.dynamic.hinet.net)
- # [08:25] * Joins: alex (alex@moz-BD8D0A09.lightspeed.sntcca.sbcglobal.net)
- # [08:27] * Joins: TheOne (TheOne@moz-D58488C3.dfki.uni-kl.de)
- # [08:28] * Quits: Mossop (mossop@moz-E325C2EC.dsl.dynamic.sonic.net) (Ping timeout)
- # [08:28] * Quits: gal (gal@moz-5FD3DD7.dsl.sndg02.sbcglobal.net) (Quit: gal)
- # [08:28] * Joins: RemusPop (remuspop@601F3B17.33662590.A5830293.IP)
- # [08:32] * Quits: ewong (chatzilla@moz-B5A86794.com) (Ping timeout)
- # [08:32] * Joins: ewong_ (chatzilla@moz-83240463.belfordhk.com)
- # [08:32] * ewong_ is now known as ewong
- # [08:34] * Joins: gal (gal@moz-5FD3DD7.dsl.sndg02.sbcglobal.net)
- # [08:37] * Joins: zuzelvp (zuzelvp@2112147D.C3507A2D.9A8C35B4.IP)
- # [08:39] * Quits: Mook (mook@moz-B1410066.dsl.teksavvy.com) (NickServ (GHOST command used by Mook_))
- # [08:39] * Joins: Mook (mook@moz-B1410066.dsl.teksavvy.com)
- # [08:42] * Joins: cpeterson (cpeterson@moz-18338A6E.hsd1.ca.comcast.net)
- # [08:42] <glob> anyone on a mac who can sanity check a bug for me?
- # [08:42] <glob> STR: http://glob.pastebin.mozilla.org/1412240
- # [08:43] <cpeterson> glob: I can take a quick look..
- # [08:43] <glob> cpeterson, thanks
- # [08:44] <glob> note you have to use gestures to go back, using the back button won't break it for me
- # [08:44] <cpeterson> What is the back gesture?
- # [08:44] <glob> 2 finger swipe left
- # [08:44] * Quits: imphil (philipp@moz-655EF802.customer.m-online.net) (Ping timeout)
- # [08:45] <cpeterson> hmm. Firefox froze. (I'm using a recent Nightly build.)
- # [08:46] <cpeterson> Was that the bug? I didn't read to the end of the STR first. <:)
- # [08:46] <glob> i got an emtpy modal dialog which i couldn't dismiss
- # [08:46] <glob> session restore also decided to no restore that window
- # [08:46] <glob> *not
- # [08:47] * glob files a bug
- # [08:47] <glob> thanks cpeterson
- # [08:47] <cpeterson> HTH. :)
- # [08:48] * Quits: ericb2 (X@moz-9C4C3DED.fbx.proxad.net) (Quit: . . . ........)
- # [08:48] <cpeterson> btw, swiping back on Aurora, I see half of an OK/Cancel dialog that I can't dismiss. But on Nightly, the browser freezes.
- # [08:52] * Joins: glazou (glazou@moz-204094DD.disruptive-innovations.fr)
- # [08:53] * Joins: past (past@moz-7A93B333.dsl.dyn.forthnet.gr)
- # [08:54] * Joins: Ms2ger (Ms2ger@CC7F0FF.207206FA.37724B0D.IP)
- # [08:55] <glazou> bonjour
- # [08:55] * Quits: Mook (mook@moz-B1410066.dsl.teksavvy.com) (Ping timeout)
- # [08:55] <GPHemsley> Is there a particular channel for discussing MPL 2.0?
- # [08:55] * Joins: whimboo (whimboo@moz-216F9AFA.superkabel.de)
- # [08:57] * Joins: Mook (mook@moz-B1410066.dsl.teksavvy.com)
- # [08:58] * Quits: Cork (none@moz-7DD8BE5D.ilait.se) (Ping timeout)
- # [08:59] * Joins: Fallen|mac (kewisch@moz-5481D33.wlan.uni-hamburg.de)
- # [08:59] <cpeterson> GPHemsley: sorry, not that I know of.
- # [09:00] * Joins: jhorak (jhorak@moz-107AD163.redhat.com)
- # [09:03] * Quits: nattokirai (nattokirai@moz-348F61F0.mozilla.or.jp) (Quit: nattokirai)
- # [09:03] * Quits: Ms2ger (Ms2ger@CC7F0FF.207206FA.37724B0D.IP) (Ping timeout)
- # [09:04] * Joins: Ms2ger (Ms2ger@CC7F0FF.207206FA.37724B0D.IP)
- # [09:04] * Quits: zuzelvp (zuzelvp@2112147D.C3507A2D.9A8C35B4.IP) (Ping timeout)
- # [09:04] * Joins: roc (chatzilla@C0ACF8B.5E1E9EEA.613E47D1.IP)
- # [09:05] * Quits: merike|away (quassel@moz-A6468214.cable.starman.ee) (Ping timeout)
- # [09:05] * Joins: Fallen|mac1 (kewisch@moz-5481D33.wlan.uni-hamburg.de)
- # [09:06] * Quits: Fallen|mac (kewisch@moz-5481D33.wlan.uni-hamburg.de) (Ping timeout)
- # [09:08] * Quits: Mook (mook@moz-B1410066.dsl.teksavvy.com) (Quit: zzz)
- # [09:08] * Quits: hub (hub@moz-E2FCA694.figuiere.net) (Ping timeout)
- # [09:11] * Parts: whimboo (whimboo@moz-216F9AFA.superkabel.de)
- # [09:12] * Joins: gandalf (zbraniecki@moz-D977477A.neoplus.adsl.tpnet.pl)
- # [09:13] * Quits: alex (alex@moz-BD8D0A09.lightspeed.sntcca.sbcglobal.net) (Client exited)
- # [09:15] * heycam is now known as heycam|away
- # [09:18] * Quits: logbot (logbot@moz-785868D2.glob.com.au) (Client exited)
- # [09:18] * Joins: logbot (logbot@moz-785868D2.glob.com.au)
- # [09:19] * Quits: bsmith (bsmith@8FB758AE.17E825E1.75B241AF.IP) (Ping timeout)
- # [09:20] * Joins: graydot (jeba@63F3AA56.9CFBD76D.30E7AE68.IP)
- # [09:21] * Quits: logbot (logbot@moz-785868D2.glob.com.au) (Client exited)
- # [09:21] * Joins: logbot (logbot@moz-785868D2.glob.com.au)
- # [09:22] * Joins: merike|away (quassel@moz-A6468214.cable.starman.ee)
- # [09:22] * Joins: Cork (none@moz-7DD8BE5D.ilait.se)
- # [09:29] <glazou> this week starts well, with a wifi office router totally dead
- # [09:29] <glazou> lovely
- # [09:30] * Joins: protz (protz@moz-E29D2A15.inria.fr)
- # [09:33] * mattwoodrow is now known as mattwoodrow|away
- # [09:33] <ewong> glad to see Murphy working hard
- # [09:33] <ewong> ;P
- # [09:34] * Joins: pnemsak (Miranda@moz-3D67D819.rainside.sk)
- # [09:34] * Joins: NeilAway (neil@moz-32AA0D01.in-addr.btopenworld.com)
- # [09:34] * Joins: yuan (ywang@moz-6A0DF30E.hsd1.ca.comcast.net)
- # [09:34] * Joins: waschtl (waschtl@moz-A4ECE553.hsi4.kabel-badenwuerttemberg.de)
- # [09:34] * Quits: yuan (ywang@moz-6A0DF30E.hsd1.ca.comcast.net) (Quit: yuan)
- # [09:34] * Quits: birtles (chatzilla@moz-348F61F0.mozilla.or.jp) (Quit: ChatZilla 0.9.87-rdmsoft [XULRunner 1.9.0.17/2009122204])
- # [09:38] * Joins: robhawkes (robhawkes@moz-33A339B7.dsl.cnl.uk.net)
- # [09:38] <glazou> oh cool, my printer seems to be dead too
- # [09:38] <glazou> what a day
- # [09:39] <nigelb> How can a Monday get better.
- # [09:39] <ewong> Oooh..
- # [09:39] * Quits: rshetty (quassel@14F7D0B8.8F230C46.C842849F.IP) (Ping timeout)
- # [09:39] <ewong> that's really asking for Murphy to work harder... ;P
- # [09:40] <glazou> I suspected something was wrong... no wifi this morning and the router did that strange "crrrcrrrrcrrr" sound :-)
- # [09:41] <ewong> routers shouldn't be making any sound (duh)... as there's no moving parts...
- # [09:41] <glazou> sure
- # [09:41] * Quits: dRdR (dRdR@moz-F979C50C.cpe.net.cable.rogers.com) (Client exited)
- # [09:41] <ewong> ergo, electrical systems' fried.
- # [09:41] <glazou> usually, a sound is the last step before a big flame coming from the silicium :-)
- # [09:41] <ewong> hmm suddenly I yearn for KFC
- # [09:42] * Joins: bwinton_away (bwinton@D7D3C4B5.88764A66.72A31D6.IP)
- # [09:42] <ewong> oh..
- # [09:43] <KWierso> kentucky fried packets?
- # [09:43] * Quits: micahg (micahg@moz-C7CC8D27.c3-0.arm-ubr1.chi-arm.il.cable.rcn.com) (Ping timeout)
- # [09:43] <ewong> KWierso: doesn't matter.. if it takes finger-linking good.. sure!
- # [09:43] <ewong> lol... finger licking...
- # [09:43] * KWierso is not sure how one would take finger-linking in a bad way
- # [09:44] <ewong> but unfortunately... (wait for it...) finger is always blocked at the router... *ba da bum*
- # [09:44] <KWierso> I feel bad for having read that pun
- # [09:44] * Quits: allisterb (allisterb@2CE5E668.5725EB66.F43E9E20.IP) (Ping timeout)
- # [09:44] <ewong> :)
- # [09:45] <ewong> thankfully, this is developers.. otherwise, if it were at foxy's ... the pun would last forever..
- # [09:46] * Quits: tonymec (tonymec@BD0053DF.4AE5B600.277517C1.IP) (Connection reset by peer)
- # [09:46] * Quits: tonymec|away (tonymec@BD0053DF.4AE5B600.277517C1.IP) (Connection reset by peer)
- # [09:46] * Quits: gal (gal@moz-5FD3DD7.dsl.sndg02.sbcglobal.net) (Quit: gal)
- # [09:47] * Joins: micahg (micahg@moz-C7CC8D27.c3-0.arm-ubr1.chi-arm.il.cable.rcn.com)
- # [09:47] * Joins: smaug (chatzilla@moz-1D1B327E.elisa-mobile.fi)
- # [09:48] * Quits: cpeterson (cpeterson@moz-18338A6E.hsd1.ca.comcast.net) (Input/output error)
- # [09:48] <bkero> nothing good can come from yearning for kfc
- # [09:49] <mcpherrin> KFC popcorn chicken can come from yearning KFC. I'd qualify that as good.
- # [09:50] * Joins: Standard8 (Standard8@B7F1AE36.48015583.54C3481B.IP)
- # [09:52] <glazou> is chromebug abandoned?
- # [09:55] * Quits: merike|away (quassel@moz-A6468214.cable.starman.ee) (Ping timeout)
- # [09:55] * Joins: merike|away (quassel@moz-A6468214.cable.starman.ee)
- # [09:55] * Quits: Standard8 (Standard8@B7F1AE36.48015583.54C3481B.IP) (Ping timeout)
- # [09:57] * Quits: sewardj (sewardj@moz-C79D927B.cable.virginmedia.com) (Ping timeout)
- # [09:57] * Joins: rshetty (quassel@407D23FE.DA1B42A8.D30E9BEF.IP)
- # [09:58] * Joins: MarcoZ (marco.zehe@moz-5F68A19.dip.t-dialin.net)
- # [09:58] * Joins: Standard8 (Standard8@B7F1AE36.48015583.54C3481B.IP)
- # [09:59] * Joins: allisterb (allisterb@2CE5E668.5725EB66.F43E9E20.IP)
- # [10:02] * glob is now known as glob|away
- # [10:02] * Joins: gcp (gpascutto@moz-D0E475EA.access.telenet.be)
- # [10:02] <glazou> what are we supposed to debug chrome xul/js with these days on trunk?
- # [10:02] <glazou> venkman is dead
- # [10:03] <glazou> chromebug seems to be dead since 1.8.0a2
- # [10:04] <darktrojan> alert
- # [10:04] <glazou> omg...
- # [10:05] <mrbkap> Callek: pong
- # [10:05] * Quits: JPeterson (JPeterson@moz-B2998FD7.cust.tele2.se) (Ping timeout)
- # [10:05] * Quits: macmaN (chezburger@moz-DA23949E.dyn.estpak.ee) (Ping timeout)
- # [10:05] * Quits: @Hixie (ianh@moz-E9FB2A6A.no) (Ping timeout)
- # [10:05] * Quits: pucko (andreas@2385E740.58E63864.630EAAA0.IP) (Ping timeout)
- # [10:05] <Unfocused> glazou: no, honza is still working on it every now and then
- # [10:06] * Joins: JPeterson (JPeterson@moz-B2998FD7.cust.tele2.se)
- # [10:06] <Unfocused> you probably want to get the source though
- # [10:06] <glazou> Unfocused: at this time, a xulrunner-based app based on m-c has NO debugging environement at all
- # [10:06] * Joins: pucko (andreas@moz-66F4E006.customer.telia.com)
- # [10:06] * Joins: Hixie (ianh@moz-E9FB2A6A.no)
- # [10:06] * ChanServ sets mode: +o Hixie
- # [10:06] <glazou> Unfocused: probably yes
- # [10:06] <glazou> and build instructions ? :-)
- # [10:07] <Honza> glazou, Unfocused: actually, nobody is working on Chromebug write know
- # [10:07] <Unfocused> yea :\ the dev-tools on m-c are catching up though
- # [10:07] <glazou> ah
- # [10:07] <Unfocused> orly? ok
- # [10:07] <glazou> Honza: so how are we suppose to debug chrome JS ?
- # [10:07] * Joins: macmaN (chezburger@moz-DA23949E.dyn.estpak.ee)
- # [10:07] <ewong> venkman is dead????
- # [10:07] <Honza> the last working version is Firebug 1.7 + Chromebug 1.7
- # [10:07] <glazou> sigh
- # [10:07] <glazou> Honza: and I don't suppose they work with nightlies at all
- # [10:08] <Honza> Somebody did a good progress on this issue report (Firebug issue list)
- # [10:08] <Honza> http://code.google.com/p/fbug/issues/detail?id=4517
- # [10:08] <ewong> oh right.. didn't someone (Waldo..? or at least someone with a W...) said he was going to work on a next-gen venkman?
- # [10:08] <Honza> So, perhaps with some further help, Chromebug could work again
- # [10:08] <Honza> (but Firebug team is small so, we don't have enough resource to also work on Chromebug)
- # [10:09] <glazou> it's just unbelievable Mozilla is not giving some serious love to a xul/js debugger
- # [10:09] <glazou> Honza: not your fault
- # [10:09] <ewong> someone has to give serious love to the xul/js debugger...
- # [10:10] <Honza> ewong yeah, that would be great, I am missing Chromebug a lot too!
- # [10:10] <ewong> where's the chromebug code?
- # [10:10] * Joins: cornelius (Mibbit@D428F312.8C1A2EFD.5642DB81.IP)
- # [10:10] <Honza> It's part of Firebug SVN, here: http://code.google.com/p/fbug/source/browse/#svn%2Fchromebug%2Fbranches%2Fchromebug1.8
- # [10:11] * Quits: rshetty (quassel@407D23FE.DA1B42A8.D30E9BEF.IP) (Ping timeout)
- # [10:11] <Honza> brb
- # [10:11] <ewong> oh.. svn.. googlecode.... *sigh*
- # [10:12] <ewong> umm what's the diff between Chromebug, firebug and Venkman?
- # [10:14] <ewong> nvm...
- # [10:14] <ewong> does Venkman have a channel?
- # [10:15] * mattwoodrow|away is now known as mattwoodrow
- # [10:16] * Joins: erione (erione@C4936045.62FF2CA1.C752B3FA.IP)
- # [10:17] <Honza> ewong Firebug is for debugging web pages, Chromebug for chrome JS, not sure about Venkman, for both? (but also not sure if there is anyone maintaining Vekman at the moment)
- # [10:18] <glazou> venkman's for both
- # [10:18] <ewong> Honza thanks for the clarification..
- # [10:18] <glandium> venkman's been pretty broken for a while
- # [10:18] <glazou> no, nobody's maintaining venkman ATM
- # [10:18] <cornelius> http://code.google.com/p/fbug/
- # [10:18] <glazou> and that's _very_ sad
- # [10:18] * Joins: gerv (gerv@moz-8E68CF56.in-addr.arpa)
- # [10:18] <ewong> glazou: yeah.. *very* sad indeed :(
- # [10:18] <glazou> the mozilla ecosystem needs a good debugging environment
- # [10:19] <glandium> venkman was a working one, not really a good one
- # [10:19] * Joins: beaufour (beaufour@moz-7BF7903.cpe.xe-6-0-0-1104.boanqu2.customer.tele.dk)
- # [10:19] <Honza> Totally agree, it's a pain to debug Firebug at the moment
- # [10:20] <glazou> glandium: "not a good one" is still waaaaaaay better than "none"
- # [10:20] <nigelb> Isn't it possible to debug chrome with firebug?
- # [10:20] <glazou> nope
- # [10:20] <glandium> hopefully, now that there is the new jsd api, things should move in the right direction
- # [10:21] * Quits: fzzzy (donovan@moz-2B41AF9B.lightspeed.mtvwca.sbcglobal.net) (Quit: fzzzy)
- # [10:21] * Joins: Goldorak (chatzilla@9CD43D39.E6A99E96.187A1082.IP)
- # [10:22] <glazou> glandium: have a url for that new api ?
- # [10:24] <glandium> https://developer.mozilla.org/en/SpiderMonkey/JS_Debugger_API_Guide
- # [10:24] <glazou> yhanks
- # [10:24] <glazou> thanks even
- # [10:24] * Quits: grubshka (grubshka@moz-498626E5.w86-216.abo.wanadoo.fr) (Ping timeout)
- # [10:24] * Joins: Polynomial-C (Poly-C@moz-7F02F440.dip.t-dialin.net)
- # [10:31] <ewong> is there work on a venkman2?
- # [10:31] <ewong> using the new JSD api?
- # [10:33] * Quits: squib (squib@moz-F5CA0CFB.dhcp.mdsn.wi.charter.com) (Quit: Leaving)
- # [10:38] * Quits: cornelius (Mibbit@D428F312.8C1A2EFD.5642DB81.IP) (Quit: http://www.mibbit.com ajax IRC Client)
- # [10:39] * Joins: msucan (msucan-@FA9E8863.56E67207.699550A1.IP)
- # [10:40] * Joins: imphil (philipp@moz-655EF802.customer.m-online.net)
- # [10:43] * Joins: mib_82oz7v (Mibbit@D428F312.8C1A2EFD.5642DB81.IP)
- # [10:46] * Quits: mib_82oz7v (Mibbit@D428F312.8C1A2EFD.5642DB81.IP) (Quit: http://www.mibbit.com ajax IRC Client)
- # [10:47] * Joins: mw22 (chatzilla@moz-FB753258.adsl.wanadoo.nl)
- # [10:47] * Quits: @dveditz (dveditz@moz-34991AF4.dhcp.cruzio.com) (Quit: dveditz)
- # [10:52] * Quits: erione (erione@C4936045.62FF2CA1.C752B3FA.IP) (Connection reset by peer)
- # [10:53] * Joins: kaze (kaze@moz-A2A3C9E6.vlan402.asr1.cdg3.gblx.net)
- # [10:53] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [10:55] * Joins: sewardj (sewardj@moz-4388D0F3.themeltingpotedinburgh.org.uk)
- # [10:56] <NeilAway> Ms2ger: Asa was asking how to pronounce your nick earlier
- # [10:56] * Quits: gandalf (zbraniecki@moz-D977477A.neoplus.adsl.tpnet.pl) (Ping timeout)
- # [10:56] * Joins: gmoro (guilherme@3DCBFB93.F520FC74.D159334F.IP)
- # [10:56] * Joins: daim (David_Mart@779E3E00.1773D26C.C0FF2207.IP)
- # [10:56] <Ms2ger> Any way he wishes
- # [10:56] * mattwoodrow is now known as mattwoodrow|away
- # [10:57] * Joins: gandalf (zbraniecki@moz-144E18A2.neoplus.adsl.tpnet.pl)
- # [10:59] * Joins: mgoodwin (mgoodwin@moz-4BB9E5E9.cable.virginmedia.com)
- # [11:00] <nigelb> Ms2ger: I'm curious how to pronounce it as well. I've been pronouncing it has Messgenger in my head all the while :P
- # [11:00] <nigelb> s/has/as
- # [11:00] * nthomas|away is now known as nthomas
- # [11:00] <Ms2ger> That works :)
- # [11:01] <mcpherrin> I've always read it as MizzTwoger
- # [11:01] <smaug> m-s-2-ger
- # [11:03] * glazou is now known as glazou_meeting
- # [11:03] * Quits: cpearce (chatzilla@moz-55A2BC7D.xdsl.xnet.co.nz) (Ping timeout)
- # [11:07] * CwiiisAway is now known as Cwiiis
- # [11:07] * ewong is now known as ewong|afk
- # [11:08] <smaug> waiting for bugzilla.mozilla.org...
- # [11:10] * Joins: quibbit (Mibbit@D428F312.8C1A2EFD.5642DB81.IP)
- # [11:11] * Quits: quibbit (Mibbit@D428F312.8C1A2EFD.5642DB81.IP) (Quit: http://www.mibbit.com ajax IRC Client)
- # [11:12] * Quits: cjones (cjones@moz-B6A93570.dsl.sndg02.sbcglobal.net) (Quit: Leaving)
- # [11:13] * Quits: Cork[home] (none@moz-9CA8A1B8.bredband.skanova.com) (Ping timeout)
- # [11:15] * Quits: darktrojan (geoff@moz-BC95E278.dsl.telstraclear.net) (Ping timeout)
- # [11:15] * Joins: florian (florian@moz-87C33FDA.kimsufi.com)
- # [11:17] * Joins: grubshka (grubshka@moz-5519B649.w86-216.abo.wanadoo.fr)
- # [11:19] * Joins: maikmerten (merten@moz-E254386D.cs.uni-dortmund.de)
- # [11:20] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [11:20] * Quits: jduell (jduell@moz-2D9EDA98.lightspeed.sntcca.sbcglobal.net) (Ping timeout)
- # [11:21] * Joins: darktrojan (geoff@moz-BC95E278.dsl.telstraclear.net)
- # [11:22] * Quits: mgoodwin (mgoodwin@moz-4BB9E5E9.cable.virginmedia.com) (Input/output error)
- # [11:22] * Joins: AutomatedTester (AutomatedT@moz-F0CBC6B6.as13285.net)
- # [11:23] * Quits: merike|away (quassel@moz-A6468214.cable.starman.ee) (Ping timeout)
- # [11:24] * Joins: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP)
- # [11:24] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [11:26] <darktrojan> oh windows, I hate you so
- # [11:26] <darktrojan> just pass my test already damnit
- # [11:30] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [11:31] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [11:31] * Quits: smaug (chatzilla@moz-1D1B327E.elisa-mobile.fi) (Ping timeout)
- # [11:32] * Joins: smaug (chatzilla@moz-1D1B327E.elisa-mobile.fi)
- # [11:33] * Joins: merike|away (quassel@moz-A6468214.cable.starman.ee)
- # [11:37] * Quits: Wes_ (chatzilla@moz-BEF0C255.page.ca) (Ping timeout)
- # [11:44] * Quits: darktrojan (geoff@moz-BC95E278.dsl.telstraclear.net) (Quit: darktrojan)
- # [11:47] * Joins: bjarne (bjarne@moz-46A85847.nextgentel.com)
- # [11:49] * Quits: slowpoke (slowpoke@moz-AE67AB45.gigabit.perfect-privacy.com) (Ping timeout)
- # [11:49] * Quits: Hendikins (wolfox@moz-9A361C95.static.internode.on.net) (Ping timeout)
- # [11:50] * Joins: Hendikins (wolfox@moz-9A361C95.static.internode.on.net)
- # [11:50] * Joins: slowpoke (slowpoke@moz-AE67AB45.gigabit.perfect-privacy.com)
- # [11:50] * Quits: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net) (Ping timeout)
- # [11:51] <janv> Ms2ger: -I$(topsrcdir)/js/src is needed there
- # [11:57] * Joins: mib_u0ro2p (Mibbit@D428F312.8C1A2EFD.5642DB81.IP)
- # [11:58] * Quits: mib_u0ro2p (Mibbit@D428F312.8C1A2EFD.5642DB81.IP) (Quit: http://www.mibbit.com ajax IRC Client)
- # [11:59] * Quits: AutomatedTester (AutomatedT@moz-F0CBC6B6.as13285.net) (Quit: AutomatedTester)
- # [12:00] * Joins: Pike (Pike@moz-75994370.dip.t-dialin.net)
- # [12:01] <Ms2ger> janv, then make it not be needed
- # [12:02] <smaug> including js/src looks evil
- # [12:02] <janv> #include "jsnum.h"
- # [12:03] <janv> JSDOUBLE_IS_NaN
- # [12:03] <janv> JSDOUBLE_IS_NaN is not exported
- # [12:03] <janv> and we need it in indexeddb
- # [12:03] <janv> Jonas will file a followup
- # [12:03] * Joins: romeo (romeo@moz-A9325888.k459.webspeed.dk)
- # [12:10] * Joins: anky (anky@D337C2AB.83619B47.A3D1B221.IP)
- # [12:11] * Quits: kanru (user@moz-BA086EAD.dynamic.hinet.net) (Ping timeout)
- # [12:12] <Ms2ger> No, a followup isn't enough, it shouldn't be added in the first place
- # [12:15] <Ms2ger> num != num is enough to check NaNness, no?
- # [12:15] <smaug> or does num == NaN work?
- # [12:16] <smaug> oh, NaN == NaN is false
- # [12:16] * nthomas is now known as nthomas|away
- # [12:16] <smaug> in JS
- # [12:17] <Pike> which is how NaN is defined
- # [12:17] <Standard8> how about isNaN() ?
- # [12:17] <Pike> isnan to the rescue
- # [12:18] <Ms2ger> How about in C++?
- # [12:18] <Standard8> heh
- # [12:18] * Joins: mak (chatzilla@moz-6699258.retail.telecomitalia.it)
- # [12:22] * Quits: peterv (peterv@moz-715F6D16.access.telenet.be) (Ping timeout)
- # [12:22] * Joins: peterv (peterv@moz-715F6D16.access.telenet.be)
- # [12:24] <smaug> "I am planning to shift from gecko version 1.9 to gecko version 2" o_O
- # [12:24] * Quits: merike|away (quassel@moz-A6468214.cable.starman.ee) (Ping timeout)
- # [12:24] <glob|away> smaug, heh
- # [12:25] * Joins: tonymec__ (tonymec@561D0D9A.23AE5289.277517C1.IP)
- # [12:26] * tonymec__ is now known as tonymec|away
- # [12:26] * Quits: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP) (Quit: jfkthame)
- # [12:31] * Joins: jfkthame (jfkthame@78C7B4BE.A1C12133.9542EC20.IP)
- # [12:31] * Ziggy|AWAY is now known as Ziggy_Maes
- # [12:31] * Quits: Hendikins (wolfox@moz-9A361C95.static.internode.on.net) (Ping timeout)
- # [12:32] * Joins: Hendikins (wolfox@moz-9A361C95.static.internode.on.net)
- # [12:37] * Quits: decoder (quassel@moz-216446B9.own-hero.net) (Ping timeout)
- # [12:38] * Joins: decoder (quassel@moz-216446B9.own-hero.net)
- # [12:38] * Joins: darktrojan (geoff@moz-BC95E278.dsl.telstraclear.net)
- # [12:40] * Joins: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw)
- # [12:46] * Quits: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw) (Quit: regen)
- # [12:48] * Joins: gabor (gabor@moz-3B57BCD1.catv.pool.telekom.hu)
- # [12:48] * Quits: anky (anky@D337C2AB.83619B47.A3D1B221.IP) (Client exited)
- # [12:49] * Joins: mike5w3c (MikeS@moz-7E49D139.pool.e-mobile.ne.jp)
- # [12:50] * Joins: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw)
- # [12:53] <darktrojan> NeilAway, ping?
- # [12:54] * Joins: KaiRo (robert@moz-CA29FA62.adsl.highway.telekom.at)
- # [12:55] <hsivonen> is gcli in m-c generated from some upstream repo? Where's the upstream?
- # [12:55] * Quits: m_kato (m_kato@moz-348F61F0.mozilla.or.jp) (Quit: Leaving...)
- # [12:58] * Quits: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw) (Quit: regen)
- # [13:01] * Joins: merike|away (quassel@moz-A6468214.cable.starman.ee)
- # [13:01] <gabor> Ms2ger: do you know how could I create a JSObject from an IIDBFactory? so what I want is wrap a native object, so I can set a property with it on a JSObject (mozIndexedDB...)
- # [13:01] * Joins: graememcc (chatzilla@moz-6D728BDA.range86-150.btcentralplus.com)
- # [13:02] <sicking> Ms2ger: i don't think a != a works in C++ reliably
- # [13:03] <sicking> Ms2ger: at least it didn't used to since the JS engine explicitly checks for it when implementing the != operator
- # [13:04] <gabor> sicking: maybe you could help me with my question a few lines above, if you have a sec?
- # [13:05] <sicking> Ms2ger: so we can add a pile of jsnum.h to somewhere, the question is where
- # [13:06] <sicking> Ms2ger: probably everything from line 68 to line 126 of jsnum.h
- # [13:06] * Joins: dao (dao@moz-961B3183.dip.tu-dresden.de)
- # [13:06] <sicking> Ms2ger: with 'moz' instead of 'js' as prefix?
- # [13:07] * Joins: fuogo (Mibbit@D428F312.8C1A2EFD.5642DB81.IP)
- # [13:07] * mak is now known as mak|afk
- # [13:08] * Quits: dao (dao@moz-961B3183.dip.tu-dresden.de) (Quit: Leaving.)
- # [13:08] * Quits: fuogo (Mibbit@D428F312.8C1A2EFD.5642DB81.IP) (Quit: http://www.mibbit.com ajax IRC Client)
- # [13:08] * Joins: peregrino (peregrino@moz-E051D345.telecom.net.ar)
- # [13:09] * Quits: darktrojan (geoff@moz-BC95E278.dsl.telstraclear.net) (Quit: darktrojan)
- # [13:10] * Joins: dao (dao@moz-90D353A4.dip.tu-dresden.de)
- # [13:11] * Quits: Fallen|mac1 (kewisch@moz-5481D33.wlan.uni-hamburg.de) (Quit: Leaving.)
- # [13:13] * Joins: evilpie (chatzilla@moz-671907A1.pools.arcor-ip.net)
- # [13:15] * Quits: sicking (chatzilla@moz-1FF05400.hsd1.ca.comcast.net) (Input/output error)
- # [13:16] * Joins: sicking (chatzilla@moz-1FF05400.hsd1.ca.comcast.net)
- # [13:16] * Joins: mib_1x3q33 (Mibbit@D428F312.8C1A2EFD.5642DB81.IP)
- # [13:16] * Joins: anky (anky@D337C2AB.83619B47.A3D1B221.IP)
- # [13:22] <NeilAway> darktrojan: pongish
- # [13:23] <hsivonen> do all telemetry additions require privacy review or are there rules about what requires privacy review?
- # [13:23] * Quits: merike|away (quassel@moz-A6468214.cable.starman.ee) (Ping timeout)
- # [13:23] * Quits: NeilAway (neil@moz-32AA0D01.in-addr.btopenworld.com) (Connection reset by peer)
- # [13:24] * Quits: graememcc (chatzilla@moz-6D728BDA.range86-150.btcentralplus.com) (Ping timeout)
- # [13:25] * Joins: NeilAway (neil@moz-32AA0D01.in-addr.btopenworld.com)
- # [13:25] * khuey|away is now known as khuey
- # [13:26] <NeilAway> well that worked well
- # [13:27] * Joins: graememcc (chatzilla@moz-E0C3C808.range86-148.btcentralplus.com)
- # [13:27] * glob|away is now known as glob
- # [13:28] <hsivonen> who can see the collected telemetry data and where?
- # [13:30] * Joins: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net)
- # [13:31] * Quits: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net) (Client exited)
- # [13:31] <smaug> https://metrics.mozilla.com/pentaho/content/pentaho-cdf-dd/Render?solution=metrics&path=telemetry/&file=TelemetryHistogram.wcdf
- # [13:31] <hsivonen> smaug: thanks
- # [13:32] <smaug> er, is that the old page
- # [13:32] <smaug> it has link to v2
- # [13:32] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [13:33] <smaug> both versions are insanely slow to load
- # [13:33] <gabor> khuey: do you know how could I create a JSObject from an IIDBFactory object? so what I want is wrap a native object, so I can set a property with it on a JSObject (mozIndexedDB on the global...)
- # [13:34] <hsivonen> does telemetry support recording tuples of enumerated values and querying by fixing one element of the tuple?
- # [13:35] <khuey> gabor: nsContentUtils::WrapNative?
- # [13:36] <hsivonen> that is, can a telemetry event record the character encoding and the firefox locale and can encoding use frequencies then be queried by locale?
- # [13:37] <smaug> hsivonen: ask #metrics
- # [13:37] <hsivonen> smaug: ok
- # [13:37] <gabor> khuey++
- # [13:37] * Quits: jprmc (jprmc@moz-7F2FF3EB.cpe.net.cable.rogers.com) (Ping timeout)
- # [13:37] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [13:38] * Joins: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw)
- # [13:38] <dao> smaug: hrm, I see no link to v2
- # [13:39] <hsivonen> dao: top of the page
- # [13:39] <smaug> dao: top of that page
- # [13:39] <Callek> mrbkap: I think my ping was about an old review request that was from a new contrib, and brought up here.... but I can't recall for sure
- # [13:44] * NeilAway would unping darktrojan but hadn't noticed that he'd left again
- # [13:48] * Quits: regen (Miller@moz-93971929.adsl.dynamic.seed.net.tw) (Quit: regen)
- # [13:49] * Quits: bhearsum (bhearsum@moz-FBAE94.members.linode.com) (Quit: ZNC - http://znc.sourceforge.net)
- # [13:49] * Joins: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com)
- # [13:50] * Joins: bhearsum (bhearsum@moz-FBAE94.members.linode.com)
- # [13:56] * Quits: dao (dao@moz-90D353A4.dip.tu-dresden.de) (Client exited)
- # [13:56] * Quits: slowpoke (slowpoke@moz-AE67AB45.gigabit.perfect-privacy.com) (Ping timeout)
- # [13:56] * Joins: jacek (jacek@moz-5D707D3B.psi.wroc.pl)
- # [13:57] * ewong|sleep is now known as ewong
- # [13:57] * Joins: dao (dao@moz-961B3183.dip.tu-dresden.de)
- # [13:59] * Quits: sicking (chatzilla@moz-1FF05400.hsd1.ca.comcast.net) (Input/output error)
- # [14:00] * rail_away is now known as rail
- # [14:03] * Joins: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com)
- # [14:06] * Joins: mgoodwin (mgoodwin@moz-4BB9E5E9.cable.virginmedia.com)
- # [14:07] * Quits: mike5w3c (MikeS@moz-7E49D139.pool.e-mobile.ne.jp) (Ping timeout)
- # [14:07] * Joins: mike5w3c (MikeS@moz-5EC55E5A.pool.e-mobile.ne.jp)
- # [14:10] * Joins: bjacob (bjacob@moz-ADCA75DC.cpe.net.cable.rogers.com)
- # [14:10] * Quits: graydot (jeba@63F3AA56.9CFBD76D.30E7AE68.IP) (Quit: graydot)
- # [14:10] <Callek> khuey++
- # [14:10] <Callek> :-)
- # [14:10] * Joins: graydot (jeba@63F3AA56.9CFBD76D.30E7AE68.IP)
- # [14:10] * Quits: graydot (jeba@63F3AA56.9CFBD76D.30E7AE68.IP) (Quit: graydot)
- # [14:10] * Quits: jdm (jdm@moz-9AEDE212.cpe.net.cable.rogers.com) (Client exited)
- # [14:12] * Joins: slowpoke (slowpoke@moz-F02DAE63.gigabit.perfect-privacy.com)
- # [14:13] <khuey> Callek: hmm?
- # [14:14] * Quits: @Hixie (ianh@moz-E9FB2A6A.no) (Ping timeout)
- # [14:14] * Quits: bjarne (bjarne@moz-46A85847.nextgentel.com) (Ping timeout)
- # [14:14] * Joins: Hixie (ianh@moz-E9FB2A6A.no)
- # [14:14] * ChanServ sets mode: +o Hixie
- # [14:14] * Quits: Cork (none@moz-7DD8BE5D.ilait.se) (Ping timeout)
- # [14:14] * Quits: macmaN (chezburger@moz-DA23949E.dyn.estpak.ee) (Ping timeout)
- # [14:15] * Quits: JPeterson (JPeterson@moz-B2998FD7.cust.tele2.se) (Ping timeout)
- # [14:15] * Joins: JPeterson (JPeterson@moz-B2998FD7.cust.tele2.se)
- # [14:15] * Joins: bjarne (bjarne@moz-46A85847.nextgentel.com)
- # [14:15] * Joins: Cork (none@moz-7DD8BE5D.ilait.se)
- # [14:15] * Joins: macmaN (chezburger@moz-DA23949E.dyn.estpak.ee)
- # [14:15] * Joins: merike|away (quassel@moz-A6468214.cable.starman.ee)
- # [14:18] <Callek> khuey: blog post
- # [14:19] <khuey> ah
- # [14:19] <khuey> yes
- # [14:21] * rail is now known as rail-coffee
- # [14:24] * Quits: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com) (Input/output error)
- # [14:25] * Joins: regen (Miller@moz-82F245A.adsl.dynamic.seed.net.tw)
- # [14:26] * Joins: prip (prip@moz-C330E3B5.k559.webspeed.dk)
- # [14:26] * Joins: ejpbruel (ejpbruel@moz-5EE20326.adsl2.static.versatel.nl)
- # [14:26] <ejpbruel> khuey++
- # [14:26] <ejpbruel> khuey: for your post on pushing compilers to the limit :)
- # [14:26] * Joins: jprmc (jprmc@9EABD95B.5BCEC6DB.DA78B690.IP)
- # [14:27] <smaug> khuey gets many ++ today.
- # [14:28] <khuey> shame firebot's karma feature is dead
- # [14:28] <khuey> I could start giving dolske a run for his money
- # [14:29] * Joins: graydot (jeba@63F3AA56.9CFBD76D.30E7AE68.IP)
- # [14:29] * smaug is now known as smaugLunch
- # [14:29] * mak|afk is now known as mak
- # [14:30] <ewong> karma feature is dead?
- # [14:31] <ewong> btw, was the blog post cut on "experimental"?
- # [14:31] <romeo> only on planet
- # [14:31] * Quits: regen (Miller@moz-82F245A.adsl.dynamic.seed.net.tw) (Quit: regen)
- # [14:32] * Joins: andreasn_ (andreasn@moz-BB6A8755.bredband.comhem.se)
- # [14:32] <khuey> yeah tumblr does stupid things to html tags in the syndicated feed
- # [14:33] <NeilAway> nice blog title ;-)
- # [14:34] * Joins: regen (Miller@moz-82F245A.adsl.dynamic.seed.net.tw)
- # [14:35] <khuey> I stole it from someone
- # [14:35] <khuey> don't remember who though
- # [14:37] <mak> sigh, bugzilla is so slow
- # [14:37] <ejpbruel> mak++
- # [14:37] <ejpbruel> oh, thats right, no karma. damnit
- # [14:38] * ejpbruel wonders how much developer time is wasted each day on waiting for bugzilla
- # [14:38] <khuey> well given that the choices are between slow and not working
- # [14:38] <khuey> probably less than the alternative
- # [14:39] * Quits: regen (Miller@moz-82F245A.adsl.dynamic.seed.net.tw) (Quit: regen)
- # [14:39] * Quits: graydot (jeba@63F3AA56.9CFBD76D.30E7AE68.IP) (Quit: graydot)
- # [14:40] * Joins: dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net)
- # [14:40] * ChanServ sets mode: +o dbaron
- # [14:40] * Quits: dao (dao@moz-961B3183.dip.tu-dresden.de) (Ping timeout)
- # [14:41] * Quits: jprmc (jprmc@9EABD95B.5BCEC6DB.DA78B690.IP) (Quit: Leaving)
- # [14:41] * Joins: dbradley (dbradley@moz-8FACAA93.fuse.net)
- # [14:41] * rail-coffee is now known as rail
- # [14:42] <ejpbruel> khuey: and fast is out of the question because...
- # [14:43] * ejpbruel waaaaaits for it...
- # [14:43] * Joins: joey (chatzilla@moz-EFCB4CBF.princetowncable.com)
- # [14:43] <jesup> it's bugzilla
- # [14:44] <khuey> indeed
- # [14:48] * gabor is dreaming about a reliable desktop alternative that does proper caching an syncing with bugzilla
- # [14:48] <khuey> bugzilla.app :-P
- # [14:48] <mrbkap> Callek: I can't find any such request.
- # [14:48] * Joins: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com)
- # [14:49] <Callek> might have been stolen, but if not I still can't remember why I pinged you, and I lost scrollback
- # [14:49] <Callek> sorry
- # [14:50] <mrbkap> np
- # [14:51] <mrbkap> Callek: no need to say "sorry" btw. If I'm tardy on a review I want to know about it.
- # [14:52] <jfkthame> khuey: at this point, would you be willing to r+ a one-line patch to set MOZ_GRAPHITE=1 in configure.in? i'd like to do that today if possible
- # [14:52] <mrbkap> I think I owe gabor some sort of alcohol for all of his reviews that I haven't gotten to in a timely manner.
- # [14:53] * Quits: michal (michal@F8B4DDD2.FC749DA6.F23860FD.IP) (Ping timeout)
- # [14:54] <khuey> jfkthame: yes, rs=me
- # [14:55] <khuey> jfkthame: we can check the memory usage numbers before and after to make sure we're still safe
- # [14:55] <gabor> mrbkap: haha :D I'll get back to you on that one :)
- # [14:55] <khuey> jfkthame: with ted's new instrumentation
- # [14:55] <jfkthame> khuey: great, thanks - i'll push it to inbound shortly
- # [14:55] * Joins: michal (michal@moz-5C5A58A8.broadband14.iol.cz)
- # [14:55] <NeilAway> firebot--
- # [14:56] <NeilAway> for not doing karma any more
- # [14:56] <ted> it doesn't?
- # [14:56] <khuey> yeah
- # [14:56] <khuey> it's broke
- # [14:56] <khuey> firebot: karma
- # [14:56] <ted> bummer
- # [14:56] <firebot> khuey: karma is invented by http://www.cs.cmu.edu/People/dkindred/ - oz liked it and brought it to the infobots. It's a community rating system. or secretly sunyata or a web-based Oracle database health monitoring tool at
- # [14:56] <khuey> er
- # [14:56] <khuey> firebot: rank
- # [14:56] <firebot> khuey: Please use 'shutup' instantly to make me shutup as this is a long send!
- # [14:58] <ted> shutup
- # [14:58] <ted> firebot: shutup
- # [14:58] <firebot> ted: I wasn't talking to you.
- # [14:58] <ted> fuck you
- # [14:58] <glob> haha
- # [14:58] <Ms2ger> !summon sicking
- # [14:58] <khuey> Ms2ger: he just went to sleep
- # [14:58] <khuey> anybody know where flash is actually installed on windows?
- # [14:59] <Ms2ger> Isn't it 6AM in MV?
- # [14:59] <khuey> roughly, yes
- # [15:00] <gabor> khuey: users/deafult/appdataroaming?
- # [15:00] <gabor> or the user's appdata
- # [15:01] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
- # [15:01] * Joins: AaronMT (AaronMT@moz-DEDA283.cpe.net.cable.rogers.com)
- # [15:01] <khuey> gabor: nope
- # [15:01] <khuey> C:\Windows\SysWow64\ ...
- # [15:02] <khuey> not what I would have expected
- # [15:02] <gabor> pfff
- # [15:02] <Ms2ger> Man, this key encoding stuff is complicated
- # [15:02] <gabor> no, not at all
- # [15:02] * Joins: armenzg (armenzg@moz-7B7ED579.home1.cgocable.net)
- # [15:03] * Joins: Suresh (chatzilla@CED6E36A.84993DFD.EB06F97B.IP)
- # [15:03] * Joins: irving (irving@moz-92D12271.dsl.bell.ca)
- # [15:04] * Joins: bbondy (bbondy@moz-28CF6D1C.home.cgocable.net)
- # [15:04] * Joins: Cork[home] (none@moz-681F9DBE.bredband.skanova.com)
- # [15:06] * Quits: peregrino (peregrino@moz-E051D345.telecom.net.ar) (Ping timeout)
- # [15:08] * Joins: peregrino (peregrino@moz-E051D345.telecom.net.ar)
- # [15:09] * Parts: Ventron (michael@moz-ED780ED3.dyn.iinet.net.au)
- # [15:09] * rail is now known as rail_away
- # [15:09] * Joins: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net)
- # [15:10] * Quits: peregrino (peregrino@moz-E051D345.telecom.net.ar) (Ping timeout)
- # [15:11] * Joins: dao (dao@moz-6CFAB7E1.superkabel.de)
- # [15:14] * Joins: espindola (espindola@37AAE0A1.1815BF32.6F478678.IP)
- # [15:14] <evilpie> try is green !
- # [15:14] * Joins: ek (Mibbit@moz-A2B32020.adsl.dynamic.totbb.net)
- # [15:15] <Ms2ger> All of it?!
- # [15:15] <evilpie> duno
- # [15:15] <evilpie> but until now
- # [15:15] * Joins: mconley (mconley@F2D29657.F60B0462.67AC9B1.IP)
- # [15:15] * Quits: grubshka (grubshka@moz-5519B649.w86-216.abo.wanadoo.fr) (Input/output error)
- # [15:15] * glazou_meeting is now known as glazou
- # [15:16] <Ms2ger> "try: -b do -p all -u all -t all CLOSED TREE"
- # [15:16] <Ms2ger> Oh really?
- # [15:16] <khuey> you like that?
- # [15:16] * Quits: romeo (romeo@moz-A9325888.k459.webspeed.dk) (Quit: Leaving)
- # [15:16] * rail_away is now known as rail
- # [15:17] <Ms2ger> khuey, I wonder if there's a way to break people who try to add js/src to their local_includes
- # [15:17] * Joins: peregrino (peregrino@moz-B84004EE.telecom.net.ar)
- # [15:19] * Joins: rwaldron (rwaldron@moz-BDCCF091.hfc.comcastbusiness.net)
- # [15:19] * Quits: rwaldron (rwaldron@moz-BDCCF091.hfc.comcastbusiness.net) (Client exited)
- # [15:19] * Joins: ajuma (ajuma@F2D29657.F60B0462.67AC9B1.IP)
- # [15:19] * Joins: rwaldron (rwaldron@moz-BDCCF091.hfc.comcastbusiness.net)
- # [15:21] <ted> sweet, we have a Core:MFBT component now
- # [15:21] <evilpie> Ms2ger i would volunteer to add a python hook for this
- # [15:21] <ted> evilpie: i'm assuming he means "error during the build"
- # [15:21] <khuey> Ms2ger: stick some generic header in js/src with #error in it?
- # [15:21] <khuey> that *might* work
- # [15:22] <ted> heh
- # [15:23] <ted> is there some sort of JS_INTERNAL define?
- # [15:23] <ted> such that you coudl just #error in the headers if that's not defined?
- # [15:23] <Ms2ger> Not that I know of
- # [15:23] <ted> so add one :)
- # [15:23] <Ms2ger> Heh
- # [15:23] <ted> or do public JSAPI headers wind up including non-public internal headers?
- # [15:23] <Ms2ger> No
- # [15:24] <ted> and clearly whatever you're desiring here will have to be worked around in xpconnect
- # [15:24] * Joins: Fallen|mac (kewisch@moz-1445363D.adsl.alicedsl.de)
- # [15:24] <Ms2ger> I'm fixing xpconnect already
- # [15:24] <dria> 2
- # [15:24] * Joins: Fallen|mac1 (kewisch@moz-1445363D.adsl.alicedsl.de)
- # [15:24] <ted> hooray
- # [15:24] <Ms2ger> dria, 3
- # [15:24] <evilpie> i wonder why some stuff is in INSTALLED_HEADERS when we don't include it from somewhere else
- # [15:24] <khuey> maybe the js engine should just accept that it gets no privacy
- # [15:24] <khuey> :-P
- # [15:24] <ted> hah
- # [15:25] <Ms2ger> !summon dmandelin
- # [15:25] * Joins: coop (Chris@moz-A6FE435.build.sjc1.mozilla.com)
- # [15:26] * Quits: Fallen|mac (kewisch@moz-1445363D.adsl.alicedsl.de) (Ping timeout)
- # [15:27] * Quits: By-Tor (bytor@moz-46974D0B.dyn.optonline.net) (Quit: Leaving)
- # [15:28] <ted> Ms2ger: but seriously, that would be my suggestion
- # [15:28] <ted> js/src does -DJS_INTERNAL or something
- # [15:28] <ted> and private headers do #ifndef JS_INTERNAL #error
- # [15:29] <Ms2ger> I was hoping to get there without changing all internal headers, but that works :)
- # [15:29] * Joins: jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP)
- # [15:29] * Joins: sheppy (sheppy@moz-F39D62DA.dhcp.kgpt.tn.charter.com)
- # [15:30] * Joins: vingtetun (vingtetun@moz-A2A3C9E6.vlan402.asr1.cdg3.gblx.net)
- # [15:30] * Joins: pascalc (chatzilla@moz-B337C47A.rev.sfr.net)
- # [15:31] <evilpie> can't we do this in one header we only include internally?
- # [15:32] * Quits: graememcc (chatzilla@moz-E0C3C808.range86-148.btcentralplus.com) (Ping timeout)
- # [15:32] * Joins: mkaply (chatzilla@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
- # [15:32] * ChanServ sets mode: +o mkaply
- # [15:32] * Joins: graememcc (chatzilla@moz-91797375.range31-52.btcentralplus.com)
- # [15:34] * Quits: anky (anky@D337C2AB.83619B47.A3D1B221.IP) (Client exited)
- # [15:34] * Quits: ek (Mibbit@moz-A2B32020.adsl.dynamic.totbb.net) (Quit: http://www.mibbit.com ajax IRC Client)
- # [15:36] * Joins: beaufour_ (beaufour@moz-7BF7903.cpe.xe-6-0-0-1104.boanqu2.customer.tele.dk)
- # [15:36] * Quits: beaufour (beaufour@moz-7BF7903.cpe.xe-6-0-0-1104.boanqu2.customer.tele.dk) (Connection reset by peer)
- # [15:36] * beaufour_ is now known as beaufour
- # [15:36] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [15:37] <Ms2ger> ted, I guess #define FOO #error "" doesn't work?
- # [15:37] * Quits: kaze (kaze@moz-A2A3C9E6.vlan402.asr1.cdg3.gblx.net) (Ping timeout)
- # [15:37] * Quits: jacek (jacek@moz-5D707D3B.psi.wroc.pl) (Client exited)
- # [15:38] <ted> nope
- # [15:38] <ted> but you could #include "jsinternalonly.h" or something
- # [15:39] * Quits: andreasn_ (andreasn@moz-BB6A8755.bredband.comhem.se) (Quit: Ex-Chat)
- # [15:39] <Ms2ger> Hmm, something with static asserts, maybe
- # [15:40] * Joins: mcmanus (mcmanus@moz-FE9B5BFD.twcny.res.rr.com)
- # [15:42] * Quits: rwaldron (rwaldron@moz-BDCCF091.hfc.comcastbusiness.net) (Ping timeout)
- # [15:42] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [15:42] <jfkthame> khuey: just wondering, would it be worth triggering PGO builds on 1c542f9a2e10 (m-i) to get an earlier indication of whether it's ok?
- # [15:43] <khuey> jfkthame: wouldn't hurt
- # [15:43] * jfkthame doesn't know how to do stuff like that - could you take care of it?
- # [15:44] <khuey> sure
- # [15:44] <jfkthame> thanks
- # [15:45] <khuey> I did before and after
- # [15:45] <khuey> so we can see the change
- # [15:45] * bear-afk is now known as bear
- # [15:45] * bear is now known as bear-buildduty
- # [15:46] <jfkthame> great - i'm not really expecting it to have problems, but still, i'll feel better once we see actual builds happening successfully
- # [15:46] * Quits: @dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net) (Ping timeout)
- # [15:46] <khuey> yeah
- # [15:46] <khuey> I expect everything will be fine
- # [15:47] <jfkthame> optimism is a lovely thing :)
- # [15:47] * Ms2ger points at the /topic
- # [15:47] <jfkthame> :P
- # [15:47] * mcote|afk is now known as mcote
- # [15:48] * Joins: dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net)
- # [15:48] * ChanServ sets mode: +o dbaron
- # [15:51] * Quits: @mkaply (chatzilla@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Ping timeout)
- # [15:53] * Joins: dbaron_ (dbaron@moz-389E0BB7.phlapa.fios.verizon.net)
- # [15:53] * Joins: arno (arno@moz-DFB73ABF.fbx.proxad.net)
- # [15:53] * Quits: @dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net) (NickServ (GHOST command used by dbaron_))
- # [15:53] * dbaron_ is now known as dbaron
- # [15:53] * Quits: dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net) (Input/output error)
- # [15:54] * Joins: dbaron_ (dbaron@moz-389E0BB7.phlapa.fios.verizon.net)
- # [15:54] * Joins: ahal (ahal@F2D29657.F60B0462.67AC9B1.IP)
- # [15:54] * Joins: kaze (kaze@moz-A2A3C9E6.vlan402.asr1.cdg3.gblx.net)
- # [15:54] <mak> sigh "Bug Updates Temporarily Suspended" :(
- # [15:55] * dbaron_ is now known as dbaron
- # [15:55] <glob> yeah, sorry, i have quite a few fields to add
- # [15:55] * Quits: dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net) (Quit: 8403864 bytes have been tenured, next gc will be global.)
- # [15:55] <mak> I have quite a few bugs to file! :)
- # [15:55] * Joins: dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net)
- # [15:55] * ChanServ sets mode: +o dbaron
- # [15:57] * mak is now known as mak|afk
- # [15:57] <jbuck> is there a secret to using nsContentUtils::ReportToConsole? I'm adding a property to dom.properties, and calling it like so: http://pastebin.mozilla.org/1412477 but it's not finding the string within the bundle
- # [15:58] * armenzg is now known as armenzg_brb
- # [15:58] * Quits: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com) (Ping timeout)
- # [15:58] * Joins: gal (gal@moz-5FD3DD7.dsl.sndg02.sbcglobal.net)
- # [16:00] * Quits: bwinton_away (bwinton@D7D3C4B5.88764A66.72A31D6.IP) (Ping timeout)
- # [16:00] * Joins: ericb2 (X@moz-9C4C3DED.fbx.proxad.net)
- # [16:00] * Joins: bwinton_away (bwinton@D7D3C4B5.88764A66.72A31D6.IP)
- # [16:02] * Quits: ericb2 (X@moz-9C4C3DED.fbx.proxad.net) (Quit: . . . ........)
- # [16:02] <@dbaron> I wonder what field we're adding to bugzilla...
- # [16:02] * Joins: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com)
- # [16:03] <glob> dbaron, rapid release tracking fields (2 of 6)
- # [16:03] <glob> added now, try again
- # [16:06] <arno> Hi, is there an api to access the element at a given point from an extension/xr appli ? I known there is the utils.elementFromPoint, but after calling nsLayoutUtils::GetFrameForPoint it goes up to the parent frame, so I get the frame element instead of an element of the inner document?
- # [16:07] * sancus is now known as sancus_
- # [16:07] * sancus_ is now known as sancus
- # [16:08] * Joins: mkaply (chatzilla@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
- # [16:08] * ChanServ sets mode: +o mkaply
- # [16:09] * Joins: hipokrit (hipokrit@moz-502C3BF.rackspace.net)
- # [16:09] * Joins: akeybl (akeybl@moz-52D39FF6.hsd1.ca.comcast.net)
- # [16:10] * Joins: Fallen|mac (kewisch@moz-6F64C71A.adsl.alicedsl.de)
- # [16:12] * Quits: Fallen|mac1 (kewisch@moz-1445363D.adsl.alicedsl.de) (Ping timeout)
- # [16:12] * Quits: @mkaply (chatzilla@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Connection reset by peer)
- # [16:13] * Joins: mkaply (chatzilla@moz-92EDDD02.lightspeed.austtx.sbcglobal.net)
- # [16:13] * ChanServ sets mode: +o mkaply
- # [16:13] * glob adds field 3 of 6
- # [16:14] <glob> added
- # [16:15] * glob adds field 4 of 6
- # [16:15] * catlee-away is now known as catlee
- # [16:16] * Joins: peregrino_ (peregrino@moz-B84004EE.telecom.net.ar)
- # [16:16] <glob> added
- # [16:17] * Quits: peregrino (peregrino@moz-B84004EE.telecom.net.ar) (Ping timeout)
- # [16:17] * peregrino_ is now known as peregrino
- # [16:18] * Quits: ahal (ahal@F2D29657.F60B0462.67AC9B1.IP) (Client exited)
- # [16:18] * Joins: ahal (ahal@F2D29657.F60B0462.67AC9B1.IP)
- # [16:19] * khuey is a bit amused that the list of "hypercritical" infrastructure is nothing that he uses on a regular basis
- # [16:19] <glob> khuey, where be said list?
- # [16:20] * Joins: mayhemer (Miranda@B3D46202.F87A741B.F23860FD.IP)
- # [16:20] <khuey> glob: mrz's zimbra postmortem post on the intranet
- # [16:20] <glob> khuey, ah, thanks
- # [16:20] <jhford> does anyone else on recent (aurora+nightly) find that certain key combinations cause the menubar blue highlighting to stay long after the key combo?
- # [16:21] <glob> jhford, yes!
- # [16:21] * Quits: dao (dao@moz-6CFAB7E1.superkabel.de) (Ping timeout)
- # [16:21] * Joins: redfive (chatzilla@moz-75DD4703.dsl.pltn13.sbcglobal.net)
- # [16:22] <espindola> jrmuizel, ping
- # [16:22] * Joins: AutomatedTester (AutomatedT@moz-F0CBC6B6.as13285.net)
- # [16:22] <jrmuizel> espindola: pong
- # [16:23] <espindola> sorry about 711658
- # [16:23] <espindola> that is really surprising
- # [16:23] * Joins: jimm (jmathies@moz-7F164CA1.pn.at.cox.net)
- # [16:23] <espindola> are you in the office today, I should pass by and can help you debug it if you want
- # [16:23] <jrmuizel> espindola: yeah, I'll be in a bit
- # [16:23] <espindola> cool. See you there
- # [16:23] <jrmuizel> espindola: I know what's going on
- # [16:24] <espindola> ah, cool
- # [16:24] <espindola> what is it?
- # [16:24] * philor|away is now known as philor
- # [16:24] <jrmuizel> you just ended up checking the OS version earlier than we do
- # [16:24] <jrmuizel> this causes the quirks to come out wrong
- # [16:25] <espindola> jrmuizel, ah, you probably have to move the os check to NS_LogInit
- # [16:25] <espindola> (which does a lot more than init logging...)
- # [16:25] <jrmuizel> we just need to do our business before NS_LogInit
- # [16:25] <jrmuizel> I don't know why we didn't in the first place
- # [16:25] <espindola> or in NS_LogInit before the call to StackWalkInitCriticalAddress, no?
- # [16:25] * Joins: jlebar (jlebar@moz-3F3A6302.dyn.columbia.edu)
- # [16:26] * Quits: ahal (ahal@F2D29657.F60B0462.67AC9B1.IP) (Connection reset by peer)
- # [16:26] * Quits: RemusPop (remuspop@601F3B17.33662590.A5830293.IP) (Client exited)
- # [16:26] <mounir> khuey: hi
- # [16:26] * Joins: ahal (ahal@F2D29657.F60B0462.67AC9B1.IP)
- # [16:26] * jwir3|away is now known as jwir3
- # [16:26] <khuey> mounir: hello
- # [16:26] <mounir> khuey: are you less busy than last week?
- # [16:27] <khuey> yes
- # [16:27] <khuey> what's up?
- # [16:27] <mounir> khuey: I wanted to know if you could have a look at bug 707578
- # [16:27] <jrmuizel> espindola: I'm sort of hesitant to add more stuff to NS_LogInit that doesn't need to be there :)
- # [16:28] <mconnor> khuey: I'm pretty sure you use LDAP...
- # [16:28] <mounir> khuey: and bug 707579 comment 2
- # [16:28] <khuey> mconnor: sometimes
- # [16:29] <mconnor> khuey: Hg, if nothing else :)
- # [16:29] <khuey> mconnor: if bugzilla and mxr stay up I'm pretty happy though
- # [16:29] <khuey> even if everything else goes down
- # [16:29] * Joins: ehsan (ehsan@F2D29657.F60B0462.67AC9B1.IP)
- # [16:29] <khuey> mconnor: yeah, but that can back up for a bit before it becomes a problem
- # [16:29] <mconnor> khuey: I think if hg/LDAP was down for a day it'd be a pretty savage problem :)
- # [16:30] <khuey> yeah
- # [16:30] <khuey> mounir: r=me
- # [16:31] <mounir> khuey: and for the second bug? :)
- # [16:31] * Quits: bwinton_away (bwinton@D7D3C4B5.88764A66.72A31D6.IP) (Ping timeout)
- # [16:32] <khuey> mounir: can't you just use hte makefile rules to determine whether or not to invoke xptlink.pl?
- # [16:32] <Ms2ger> Ugh, .pl
- # [16:32] * Quits: Pike (Pike@moz-75994370.dip.t-dialin.net) (Ping timeout)
- # [16:32] <mounir> khuey: not really because xptlink.pl is looking at xpt files
- # [16:33] * Joins: cjones (cjones@moz-B6A93570.dsl.sndg02.sbcglobal.net)
- # [16:33] <mounir> I mean, it's looking at which components have to be merged with which files
- # [16:33] <khuey> mounir: hmm
- # [16:33] * Joins: Pike (Pike@moz-5E07E9D5.dip.t-dialin.net)
- # [16:33] <mounir> khuey: if it's doable in a Makefile that seems non-trivial to me
- # [16:33] <khuey> yeah
- # [16:33] <khuey> :-(
- # [16:33] * jmaher|afk is now known as jmaher
- # [16:34] <ted> mounir: you could probably do something like
- # [16:34] <ted> browser.xpt: $(wildcard whatever/*.xpt)
- # [16:35] <ted> <do the linking>
- # [16:35] <ted> but that's definitely tricky with our current setup
- # [16:35] <khuey> we really should rip out xpt-link.pl
- # [16:35] <ted> yes
- # [16:35] <khuey> i'm kinda tempted just to r+ the hack in the bug
- # [16:35] * armenzg_brb is now known as armenzg
- # [16:35] <mounir> khuey: please :)
- # [16:37] * Joins: jhopkins (jhopkins@moz-41E1D586.tb.shawcable.net)
- # [16:39] * ctalbert|afk is now known as ctalbert
- # [16:42] * Joins: ericb2 (X@moz-9C4C3DED.fbx.proxad.net)
- # [16:43] * Joins: dao-m (AndChat@moz-B8502916.web.vodafone.de)
- # [16:43] * Quits: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net) (Ping timeout)
- # [16:44] * Joins: Joeh (joe@5A3923AA.BC22908.C7CEC4ED.IP)
- # [16:44] * Quits: AutomatedTester (AutomatedT@moz-F0CBC6B6.as13285.net) (Quit: AutomatedTester)
- # [16:44] * Quits: ericb2 (X@moz-9C4C3DED.fbx.proxad.net) (Quit: . . . ........)
- # [16:47] * Quits: bjacob (bjacob@moz-ADCA75DC.cpe.net.cable.rogers.com) (Quit: Konversation terminated!)
- # [16:47] * Quits: redfive (chatzilla@moz-75DD4703.dsl.pltn13.sbcglobal.net) (Ping timeout)
- # [16:47] * coop is now known as coop|afk
- # [16:49] <khuey> nice
- # [16:49] <khuey> egencia doesn't work in nightlies
- # [16:50] * Quits: bjarne (bjarne@moz-46A85847.nextgentel.com) (Quit: Leaving.)
- # [16:50] * Joins: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net)
- # [16:51] * Joins: peregrino_ (peregrino@moz-75D579F4.telecom.net.ar)
- # [16:51] * Quits: peregrino (peregrino@moz-B84004EE.telecom.net.ar) (Ping timeout)
- # [16:51] * peregrino_ is now known as peregrino
- # [16:52] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [16:52] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [16:54] * armenzg is now known as armenzg_brb
- # [16:55] * Joins: bwinton (bwinton@F2D29657.F60B0462.67AC9B1.IP)
- # [16:56] * Joins: zuzelvp (zuzelvp@2112147D.C3507A2D.9A8C35B4.IP)
- # [16:57] * Joins: jwatt (roslea@jwatt.irc.users.mozilla.org)
- # [16:58] * smaugLunch is now known as smaug
- # [16:59] * Joins: ashughes (ashughes@6202C507.852FE72D.DA78B690.IP)
- # [16:59] * Joins: victorporof (victorporo@5D14C11A.70F4578F.79933D60.IP)
- # [17:02] * Quits: gerv (gerv@moz-8E68CF56.in-addr.arpa) (Connection reset by peer)
- # [17:02] * Joins: gerv (gerv@moz-8E68CF56.in-addr.arpa)
- # [17:03] * Joins: grubshka (grubshka@moz-5519B649.w86-216.abo.wanadoo.fr)
- # [17:03] * Quits: maikmerten (merten@moz-E254386D.cs.uni-dortmund.de) (Quit: Verlassend)
- # [17:05] * Joins: hub (hub@83874EA1.EB7C1AF9.6F478678.IP)
- # [17:05] * Joins: Wes_ (chatzilla@moz-BEF0C255.page.ca)
- # [17:07] * Joins: kumar (kmcmillan@moz-F2D05B8.c3-0.stk-ubr1.chi-stk.il.cable.rcn.com)
- # [17:07] * Quits: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com) (Input/output error)
- # [17:07] * Joins: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com)
- # [17:08] * Quits: jrmuizel (jrmuizel@moz-20EF8EAA.cpe.net.cable.rogers.com) (Input/output error)
- # [17:08] <khuey> bsmedberg: ping?
- # [17:08] * Joins: cpeterson (cpeterson@moz-18338A6E.hsd1.ca.comcast.net)
- # [17:13] * Joins: jorendorff (jorendorff@moz-91590D94.hsd1.tn.comcast.net)
- # [17:13] * ewong is now known as ewong|sleep
- # [17:15] * Joins: lmandel (lmandel@F2D29657.F60B0462.67AC9B1.IP)
- # [17:15] * bear-buildduty is now known as bear-buildduty-afk
- # [17:16] * mak|afk is now known as mak
- # [17:16] <bsmedberg> khuey: pong
- # [17:17] <khuey> bsmedberg: do 'cached actor is out of date' assertions mean anything?
- # [17:17] * vladan-afk is now known as vladan
- # [17:17] <bsmedberg> khuey: probably not, no, I thought I wrote a patch to remove the common case there
- # [17:17] <khuey> ok
- # [17:17] <bsmedberg> khuey: basically we're comparing the cached actor to a NULL thing during teardown
- # [17:17] <bsmedberg> which is harmless
- # [17:17] <khuey> perfect
- # [17:17] * khuey ignores these
- # [17:18] * jfkthame is now known as jfkthame_afk
- # [17:19] * Quits: cpeterson (cpeterson@moz-18338A6E.hsd1.ca.comcast.net) (Input/output error)
- # [17:20] * joduinn-home is now known as joduinn-coffee
- # [17:21] * Quits: pnemsak (Miranda@moz-3D67D819.rainside.sk) (Quit: pnemsak)
- # [17:24] * Quits: protz (protz@moz-E29D2A15.inria.fr) (Quit: Leaving)
- # [17:24] * Quits: gabor (gabor@moz-3B57BCD1.catv.pool.telekom.hu) (Ping timeout)
- # [17:25] * Joins: ericb2 (X@moz-9C4C3DED.fbx.proxad.net)
- # [17:26] * Joins: cpeterson (cpeterson@moz-18338A6E.hsd1.ca.comcast.net)
- # [17:26] * Quits: @mkaply (chatzilla@moz-92EDDD02.lightspeed.austtx.sbcglobal.net) (Client exited)
- # [17:26] * Joins: bnicholson (bnicholson@moz-F7C50328.rcmdva.fios.verizon.net)
- # [17:26] * Quits: @dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net) (Quit: 8403864 bytes have been tenured, next gc will be global.)
- # [17:29] * Quits: cpeterson (cpeterson@moz-18338A6E.hsd1.ca.comcast.net) (Input/output error)
- # [17:32] * Joins: kdcw (kdc@moz-F7413045.pk.shawcable.net)
- # [17:32] * armenzg_brb is now known as armenzg
- # [17:32] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [17:32] * Quits: cjones (cjones@moz-B6A93570.dsl.sndg02.sbcglobal.net) (Ping timeout)
- # [17:34] * Joins: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com)
- # [17:35] * Quits: wesj (wesj@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: Instantbird 1.2a1pre)
- # [17:35] * Joins: wesj (wesj@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [17:36] * Quits: joey (chatzilla@moz-EFCB4CBF.princetowncable.com) (Quit: ChatZilla 0.9.87 [Firefox 8.0.1/20111120135848])
- # [17:37] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [17:38] * philor is now known as philor|away
- # [17:38] * Joins: edmorley (edmorley@moz-28ED10AB.range86-144.btcentralplus.com)
- # [17:38] * Joins: mw22_away (chatzilla@moz-FB753258.adsl.wanadoo.nl)
- # [17:39] * lsblakk|afk is now known as lsblakk
- # [17:42] * mkelly is now known as mkelly|wfh
- # [17:43] * Joins: gabor (gabor@moz-3B57BCD1.catv.pool.telekom.hu)
- # [17:44] * Joins: jdm (jdm@moz-9AEDE212.cpe.net.cable.rogers.com)
- # [17:44] <Bas> What do I need to do again to make the test-plugin work?
- # [17:44] * Joins: biesi (cbiesinger@2E747C8D.C1499D0.396E4C6D.IP)
- # [17:45] * Quits: espindola (espindola@37AAE0A1.1815BF32.6F478678.IP) (Quit: Ex-Chat)
- # [17:47] * Joins: andreasn (andreasn@moz-BB6A8755.bredband.comhem.se)
- # [17:50] <bsmedberg> Bas: explain the question?
- # [17:50] <Bas> bsmedberg: Hrm, well, I was noticing plugin-sanity.html wasn't showing what I was expecting. But it looks like it may be my fault.
- # [17:50] <Bas> I'm doing a clearn build now.
- # [17:50] * Quits: andreasn (andreasn@moz-BB6A8755.bredband.comhem.se) (Quit: Ex-Chat)
- # [17:50] * Joins: Waldo (waldo@moz-534B4EF1.hsd1.mi.comcast.net)
- # [17:50] * Joins: andreasn (andreasn@moz-BB6A8755.bredband.comhem.se)
- # [17:52] * gregglind_away is now known as gregglind
- # [17:53] * bear-buildduty-afk is now known as bear-buildduty
- # [17:53] * Quits: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net) (Ping timeout)
- # [17:54] <nemo> Hm. Does CORS not work from file:// to http:// ?
- # [17:55] <khuey> probably not
- # [17:55] <nemo> durn
- # [17:56] * nemo sets up a 2nd test server
- # [17:56] <nemo> I figured * meant * :)
- # [17:56] * ddahl is now known as ddahl|away
- # [17:56] * Quits: Fallen|mac (kewisch@moz-6F64C71A.adsl.alicedsl.de) (Quit: Leaving.)
- # [17:56] <nemo> khuey: btw, I've been reading and rereading the CORS spec, and as far as I can tell, I can whitelist * and I can whitelist foo.bar.com and baz.bar.com but I can't whitelist *.bar.com,bar.com :-/
- # [17:57] * Quits: graememcc (chatzilla@moz-91797375.range31-52.btcentralplus.com) (Ping timeout)
- # [17:57] * Quits: gal (gal@moz-5FD3DD7.dsl.sndg02.sbcglobal.net) (Quit: gal)
- # [17:57] <khuey> nemo: 302 sicking ;-)
- # [17:58] * Joins: graememcc (chatzilla@moz-4BCD914C.range86-148.btcentralplus.com)
- # [17:58] <jdm> biesi: ping
- # [17:58] <smaug> nemo: complain to Anne :)
- # [17:58] <biesi> jdm, pong
- # [17:58] <jdm> biesi: do you have time to chat about file streams and deferred open?
- # [17:58] <biesi> jdm, sure
- # [17:59] <nemo> khuey: hmmm. actually. that might have been a joke. but. yeah. I guess that *could* be enforced by checking connecting IP/referer?
- # [17:59] * coop|afk is now known as coop
- # [17:59] <nemo> khuey: admittedly, that is less convenient if I just wanted to whitelist all the local domain machines for accessing intranet resources
- # [17:59] <khuey> nemo: my knowledge of CORS is extremely limited
- # [17:59] <khuey> nemo: but sicking helped right the spec
- # [17:59] <nemo> ah
- # [17:59] <khuey> hence hte redirect
- # [17:59] <nemo> khuey: heh. I guessed that was what you meant.
- # [17:59] <nemo> but, then I started thinking about how redirects could do the job :)
- # [17:59] <jdm> biesi: I've been working on bug 702949, and we found that the problem is that the nsIDOMFile we obtain uses deferred opening, so at the point when we add the file stream to the multiplex stream that makes up the form data, we have no idea that the file stream is not actually valid
- # [18:00] <nemo> khuey: I could probably write an apache rule that only sets * if the domain fits...
- # [18:00] <jdm> biesi: we only find that out when calling Available on the multiplex stream when starting the http transaction, which breaks everything
- # [18:00] <nemo> wouldn't even need a redirect
- # [18:01] <jdm> biesi: I'm wondering if we should try creating a safe file stream that will just act as if the file's empty if it's unavailable, or if we should look into swallowing errors, or something else
- # [18:01] * Quits: jhorak (jhorak@moz-107AD163.redhat.com) (Client exited)
- # [18:01] <nemo> smaug: I'm right though? that's what I'm gonna have to do? I just figured this is a case where following the spec instead of TIAS could avoid flaky behaviour
- # [18:01] * Quits: janv (varga@moz-C1261AFF.flarion.as5628.telecom.sk) (Ping timeout)
- # [18:02] <biesi> jdm, I don't think making HTTP swallow the errors is a good approach, but I like the "safe" file stream one
- # [18:02] <biesi> except for the name :)
- # [18:02] <jdm> biesi: ReliableFileStream?
- # [18:02] <jdm> InfallibleFileStream?
- # [18:02] * Quits: TheOne (TheOne@moz-D58488C3.dfki.uni-kl.de) (Quit: )
- # [18:04] * Quits: dao-m (AndChat@moz-B8502916.web.vodafone.de) (Quit: Bye)
- # [18:04] <khuey> SuperFileStream
- # [18:05] * Joins: northWind (northWind@moz-20E58859.cable.teksavvy.com)
- # [18:05] <biesi> jdm, Infallible
- # [18:05] * Joins: jhammink (textual@moz-7B0110AD.mv.mozilla.com)
- # [18:05] <biesi> jdm, because this isn't safer or more reliable, it just doesn't report errors
- # [18:05] <biesi> so it's actually less safe! ;)
- # [18:05] <jdm> heh
- # [18:05] <sid0> http://www.haskell.org/haskellwiki/Wadlers_Law
- # [18:07] * Joins: janv (varga@moz-C1261AFF.flarion.as5628.telecom.sk)
- # [18:08] * Joins: joey (chatzilla@moz-EFCB4CBF.princetowncable.com)
- # [18:08] * Joins: nitot (nitot@moz-A2A3C9E6.vlan402.asr1.cdg3.gblx.net)
- # [18:08] * Joins: erione (erione@67A3F00C.DE807506.C752B3FA.IP)
- # [18:09] * Joins: rshetty (quassel@F4613178.6B56F8F4.D30E9BEF.IP)
- # [18:10] <ted> Bas: testplugin should "just work" if you're running from dist/bin, FWIW
- # [18:10] <ted> or running "make reftest"
- # [18:10] * Joins: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net)
- # [18:10] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
- # [18:10] * Quits: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net) (Connection reset by peer)
- # [18:11] * Quits: prip (prip@moz-C330E3B5.k559.webspeed.dk) (Client exited)
- # [18:12] * Quits: glazou (glazou@moz-204094DD.disruptive-innovations.fr) (Quit: bbl)
- # [18:13] * glob is now known as glob|away
- # [18:13] <jesup> biesi: BlindFileStream. or ErrorlessFileStream. Or MaybeFileStream ;-)
- # [18:14] <taras> khuey: what kind of stuff?
- # [18:14] * Joins: jduell (jduell@moz-2D9EDA98.lightspeed.sntcca.sbcglobal.net)
- # [18:15] * Quits: Elen (El@moz-EB0E359F.hsd1.ut.comcast.net) (Ping timeout)
- # [18:15] <biesi> jesup, sure, those work too :)
- # [18:15] <khuey> taras: for DOM Files, we want to pre-stat before we hand them to the page
- # [18:15] <khuey> taras: because there's a synchronous js api to get the size
- # [18:15] <khuey> taras: but we can delay handing them back
- # [18:15] <taras> what's a dom file?
- # [18:15] <khuey> taras: but we don't really have another thread to do that on
- # [18:16] <khuey> https://developer.mozilla.org/en/Using_files_from_web_applications
- # [18:16] * Joins: jhammel (jhammel@moz-EBD5D56.hsd1.ca.comcast.net)
- # [18:16] <taras> ah
- # [18:16] * rhungAFK is now known as rhung
- # [18:16] * Joins: fzzzy (donovan@moz-2B41AF9B.lightspeed.mtvwca.sbcglobal.net)
- # [18:17] * Joins: smooney (smooney@moz-BBE3ABD.mv.mozilla.com)
- # [18:17] * mcote is now known as mcote|lunch
- # [18:17] <khuey> taras: it seems like overkill to spin up a thread to do one stat
- # [18:17] * Joins: chrisccoulson (chr1s@moz-692D94C8.cust-3601.ip.static.uno.uk.net)
- # [18:17] <khuey> taras: but there's no existing thread we can sanely reuse, afaik
- # [18:18] <taras> khuey: correct
- # [18:18] <taras> khuey: no good answer for that, sorry
- # [18:19] <jduell> biesi: ping
- # [18:19] <khuey> taras: we should figure something out for this then
- # [18:19] <khuey> taras: I suspect this is a fairly large problem
- # [18:19] <khuey> that other bits of code suffer from
- # [18:19] <taras> well, you need async open/stat/close(?)
- # [18:20] <taras> and readdir
- # [18:20] <khuey> perhaps
- # [18:20] <khuey> (depends on how it's implemented)
- # [18:20] * Joins: Elen (El@moz-EB0E359F.hsd1.ut.comcast.net)
- # [18:20] <biesi> necko does have file i/o threads
- # [18:20] <biesi> well, thradpool
- # [18:20] <biesi> jduell, pong
- # [18:21] <khuey> I don't think we can rely on that
- # [18:21] <khuey> the user could be offline, no?
- # [18:21] <biesi> this threadpool exists anyway
- # [18:21] <khuey> interesting
- # [18:21] * khuey was not aware of this
- # [18:21] * taras neither
- # [18:21] * Joins: DGMurdockIII (dgmurdocki@moz-E933B63.hsd1.il.comcast.net)
- # [18:21] <biesi> http://mxr.mozilla.org/mozilla-central/source/netwerk/base/src/nsStreamTransportService.cpp#474
- # [18:22] <biesi> hmm there's also http://mxr.mozilla.org/mozilla-central/source/netwerk/base/src/nsIOThreadPool.cpp
- # [18:22] * rail is now known as rail-lunch
- # [18:22] <biesi> I don't know what that one is used for!
- # [18:23] <jduell> biesi: quick question for you. So I'm trying to fix websockets' allocation for large incoming messages. Right now we use a buffer that we realloc and memcpy as it grows every 8K, which is bad
- # [18:23] <khuey> that's a bad sign
- # [18:23] * Joins: myk (myk@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [18:23] <jduell> biesi: so I'm thinking of just realloc-ing once for a large message, to the total size of the message (fragment, actually, but that's not important)
- # [18:23] <biesi> jduell, sounds reasonable...
- # [18:24] * adam-afk is now known as adam
- # [18:24] <jduell> biesi: my question: is it a DOS risk to allow some server to say "I'm going to send you a 1 GB message" and we allocate 1 GB for it, but then the server could not send anything, and we're stuck with a 1 GB msg?
- # [18:24] <taras> khuey: cc me on any bugs on this
- # [18:24] <jduell> AFAIK we don't time out WS msgs
- # [18:25] <khuey> isn't it a DOS risk just to allocate an arbitrary amount that the server tells you to?
- # [18:25] <khuey> at least, an arbitrary unbounded amount
- # [18:25] * Joins: sicking (chatzilla@moz-1FF05400.hsd1.ca.comcast.net)
- # [18:25] <khuey> taras: ok
- # [18:25] <biesi> jduell, can we enforce a size limit on websocket messages?
- # [18:26] <biesi> i.e. does the spec allow it/suggest it/require it
- # [18:26] <jduell> biesi: heh. If you can get such an idea past sicking and smaug :)
- # [18:26] <jduell> The spec allows us to be arbitrary: we can claim "buffer full" at any time, so yes, we could have a limit
- # [18:26] <jduell> But they really don't want us to have one.
- # [18:26] <jduell> I was wondering if you know offhand what we do for HTTP with very large msgs. Allocate chunks?
- # [18:27] <biesi> jduell, ok, well you can always do it the other way - if message > 50 MB, fall back to realloc strategy
- # [18:27] <smaug> if we add some limit, it must be large enough to support reasonable sized files
- # [18:27] <biesi> jduell, HTTP never keeps the entire thing in memory at once
- # [18:27] <biesi> jduell, it passes on the chunks as it receives them
- # [18:27] <st3fan> 50 MB is huge for websockets i think
- # [18:27] <smaug> we've had bugs because XHR doesn't, IIRC, support more than 2GB
- # [18:27] <st3fan> i would not be surprised if the average websocket message is < 50 KB
- # [18:28] <biesi> st3fan, averages are useless for the purposes of this discussion :-)
- # [18:28] <jduell> smaug: in the long term (i.e. not for landing today :) I'm all for very large msg support, especially for file-backed blobs
- # [18:28] <st3fan> biesi: well if we are trying to lower firefox's memory usage then allocating a small average initially is a very relevant i think?
- # [18:29] <biesi> st3fan, it sounded like we know what the server wants us to allocate?
- # [18:29] <st3fan> oh is a content-length enforced?
- # [18:29] <st3fan> that is good :)
- # [18:29] <jduell> in the short term we don't do blobs to a file--we copy them into memory for now. So we need to balance the DOS issue with the realloc-efficiency issue when doing huge msgs
- # [18:29] <st3fan> i'm using websockets but i have not really looked at the protocol details
- # [18:29] <jduell> and some kind of temporary limit might be smart.
- # [18:29] <Bas> bsmedberg: Yup, sorry, 't was my own bug.
- # [18:30] <smaug> hsivonen: are those patches needed for FF11 ?
- # [18:30] <bsmedberg> np
- # [18:30] <biesi> jduell, yeah I think we really don't want to allocate gigabytes initially
- # [18:30] * Quits: jwatt (roslea@jwatt.irc.users.mozilla.org) (Input/output error)
- # [18:30] <biesi> jduell, but we can allocate everything if < 50 MB or whatever and grow it as we get the data
- # [18:30] <jduell> biesi: smaug: well, can we come up with a limit we can live with for FF 11, and revisit when we've got smarter large file support?
- # [18:30] <smaug> but we need to be able to handle real world size files
- # [18:31] <biesi> my suggestion does let us handle arbitrary sizes!
- # [18:31] * Joins: jwatt (roslea@jwatt.irc.users.mozilla.org)
- # [18:31] <jduell> biesi: true, I can probably knock off something today that does occasional reallocs for very large messages and we can support pretty big files
- # [18:32] <smaug> jduell: what is the problem supporting huge messages?
- # [18:32] <biesi> jduell, I think that's the best strategy for nbow
- # [18:32] <smaug> if OOM happens, the connection should be closed
- # [18:32] <jduell> Even then it might be nice/sane to have something less than a 2 GB limit for now until we stress test a bit?
- # [18:32] <jduell> Or maybe not--I do have the logic to fail the cxn if malloc fails.
- # [18:32] <biesi> I'll leave that part of the discussion to you/smaug/sicking :-)
- # [18:33] <jduell> biesi: smaug: ok, I think I'll do this then
- # [18:33] * biesi on vacation, should do vacationy things
- # [18:33] <jduell> 1) allow arbitrary msg size
- # [18:33] * philor|away is now known as philor
- # [18:33] <smaug> biesi: like hack Necko
- # [18:33] <jduell> 2) for up to XYZ msg size (50 MB?) just allocate full size of fragment, so no reallocs
- # [18:34] <jduell> 3) to avoid DOS attacks, for >50MB alloc a 50 MB buffer, and realloc it (by += 50MB) as message arrives
- # [18:34] <jduell> 4) Use something smaller than 50 MB for mobile--a lot smaller. 1 MB?
- # [18:35] <jduell> biesi: smaug: thoughts/
- # [18:35] <jduell> ?
- # [18:35] <khuey> 50 MB seems pretty big even for desktop
- # [18:35] * Joins: rillian (giles@21B7B9F2.B87E9213.6E712CE2.IP)
- # [18:35] <smaug> jduell: sounds ok. perhaps no need for 50MB
- # [18:36] <jduell> We can use a smaller cutoff. Most WS msgs will be smaller.
- # [18:36] <smaug> perhaps the current 16MB is enough for that
- # [18:36] <Waldo> that sounds good to me
- # [18:36] <Waldo> I wouldn't go above O(10MB)
- # [18:36] <jduell> OK then. As voted upon by an inchoate mess of IRC chatters.
- # [18:36] <jduell> 16 MB it is for now
- # [18:36] <smaug> :)
- # [18:36] <Waldo> we rule
- # [18:36] <jduell> (mobile I'll do 1 MB?)
- # [18:37] <Waldo> also sensible
- # [18:37] <smaug> sounds ok
- # [18:37] <jduell> yay
- # [18:37] * jduell runs off to code
- # [18:37] * Joins: philipp64|laptop (chatzilla@moz-764C1742.hfc.comcastbusiness.net)
- # [18:38] * Quits: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net) (Ping timeout)
- # [18:38] * Quits: nitot (nitot@moz-A2A3C9E6.vlan402.asr1.cdg3.gblx.net) (Ping timeout)
- # [18:38] * Quits: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com) (Ping timeout)
- # [18:38] * Joins: nitot (nitot@moz-A2A3C9E6.vlan402.asr1.cdg3.gblx.net)
- # [18:38] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
- # [18:39] <jduell> sicking: you ok with ^^^^ ?
- # [18:39] * Joins: GPHemsley (GPHemsley@moz-7D8B319.dyn.optonline.net)
- # [18:40] * Quits: waschtl (waschtl@moz-A4ECE553.hsi4.kabel-badenwuerttemberg.de) (Client exited)
- # [18:41] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
- # [18:41] * Joins: Enn (enn@moz-DB6467E3.cpe.net.cable.rogers.com)
- # [18:41] <sicking> jduell: looking
- # [18:42] * Joins: Sander (chatzilla@moz-B871F4D3.direct-adsl.nl)
- # [18:44] * Joins: jrmuizel (jrmuizel@F2D29657.F60B0462.67AC9B1.IP)
- # [18:44] <sicking> jduell: so for a 30MB message, we'd alloc a 16MB buffer, then realloc to 32MB?
- # [18:44] * jcranmer|away is now known as jcranmer|driving_like_mad
- # [18:44] * Quits: joey (chatzilla@moz-EFCB4CBF.princetowncable.com) (Ping timeout)
- # [18:44] <sicking> jduell: and if either allocation fails, close the connection?
- # [18:45] <imphil> is it normal that the JP tests on try fail?
- # [18:45] * Joins: treitter (treitter@moz-AE62400F.hsd1.ca.comcast.net)
- # [18:45] * Joins: Mook_as (mook@moz-1FCC0032.activestate.com)
- # [18:46] * Joins: Jake (Jake@moz-CDAFE004.phlapafg.dynamic.covad.net)
- # [18:46] * Quits: daim (David_Mart@779E3E00.1773D26C.C0FF2207.IP) (Client exited)
- # [18:46] * armenzg is now known as armenzg_lunch
- # [18:46] * Quits: ashughes (ashughes@6202C507.852FE72D.DA78B690.IP) (Ping timeout)
- # [18:47] <khuey> imphil: yes, those can be ignored
- # [18:47] <imphil> khuey, thanks
- # [18:47] <treitter> does anyone see what I'm doing wrong wrt adding a ref to this object?: http://pastebin.mozilla.org/1412572
- # [18:47] <jduell> sicking: yeah, something like that (maybe alloc 30 MB total not 32)
- # [18:47] <treitter> (there may be other errors as well; please point them out if you see them)
- # [18:48] <sicking> jduell: either sounds good to me
- # [18:48] <treitter> this class is just to roll up a context I can use in a callback and I'd like it to have nsRefPtr/nsCOMPtr-style reference counting if possible
- # [18:49] * Joins: Ami_Ty (Amie@moz-BBE3ABD.mv.mozilla.com)
- # [18:49] * Quits: Suresh (chatzilla@CED6E36A.84993DFD.EB06F97B.IP) (Quit: Suresh)
- # [18:49] * Quits: beaufour (beaufour@moz-7BF7903.cpe.xe-6-0-0-1104.boanqu2.customer.tele.dk) (Quit: beaufour)
- # [18:50] <Ms2ger> treitter, you're manually addreffing a smart pointer?
- # [18:50] * sheppy is now known as sheppy-afk
- # [18:50] <jhammel> is awesomebar misbehaving for anyone in the last two nightlies?
- # [18:50] * Quits: peregrino (peregrino@moz-75D579F4.telecom.net.ar) (Ping timeout)
- # [18:50] <treitter> Ms2ger: yeah, I need to use it in a GLib callback, so I need to make sure it lives until that's called (at which point I'll unref it)
- # [18:50] <jhammel> it doesn't seem to be, um, awesome anymore
- # [18:51] <treitter> Ms2ger: I'm new to XPCOM/the Mozilla codebase, so if I'm doing something obviously wrong, advice is greatly appreciated :)
- # [18:51] <jduell> sicking: thanks
- # [18:51] <smaug> treitter: but NS_IF_ADDREF(searchViewPrepareData); isn't possible
- # [18:51] * Joins: redfive (chatzilla@moz-B94E1A81.ipnetworksinc.net)
- # [18:51] <smaug> searchViewPrepareData is nsRefPtr
- # [18:51] <treitter> smaug: is it only for classes derived from ISupports?
- # [18:51] <Ms2ger> No
- # [18:51] <Ms2ger> You can't addref smart pointers
- # [18:51] <smaug> you could do something like NS_IF_ADDREF(searchViewPrepareData.get())
- # [18:51] <treitter> because I did have this class following that scheme and got the same build error
- # [18:52] <smaug> or just forget()
- # [18:52] <hsivonen> smaug: none of my patches in your queue are critical for Firefox 11
- # [18:52] <mbrubeck> ttaubert: Wow, you managed to find bug 712032 just 12 minutes after I filed it. :P
- # [18:52] <smaug> hsivonen: ok, thanks
- # [18:53] * Quits: Boriss (FlyingToas@moz-62AAA429.hsd1.ca.comcast.net) (Quit: Boriss)
- # [18:53] <treitter> smaug: hmm, which class is .get() inherited from?
- # [18:53] <Ms2ger> treitter, nsRefPtr::get
- # [18:53] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [18:53] * Joins: peregrino (peregrino@moz-96A63A2D.telecom.net.ar)
- # [18:53] * Joins: Boriss (FlyingToas@moz-62AAA429.hsd1.ca.comcast.net)
- # [18:54] * Quits: gandalf (zbraniecki@moz-144E18A2.neoplus.adsl.tpnet.pl) (Quit: Computer has gone to sleep.)
- # [18:54] <treitter> shouldn't my class derive from nsRefPtr then? I didn't see that in any other examples of RefPtr classes
- # [18:54] <Ms2ger> No
- # [18:54] <Ms2ger> You're calling a function on the nsRefPtr object on the stack
- # [18:54] <Ms2ger> Not on the pointer behind it
- # [18:55] * glob|away is now known as glob
- # [18:55] * Quits: peregrino (peregrino@moz-96A63A2D.telecom.net.ar) (Quit: peregrino)
- # [18:55] <ttaubert> mbrubeck: erm :D
- # [18:56] <treitter> Ms2ger: ah, of course.
- # [18:56] * Joins: rwaldron (rwaldron@moz-BDCCF091.hfc.comcastbusiness.net)
- # [18:56] * mcote|lunch is now known as mcote
- # [18:56] * Quits: micahg (micahg@moz-C7CC8D27.c3-0.arm-ubr1.chi-arm.il.cable.rcn.com) (Ping timeout)
- # [18:57] * Quits: Boriss (FlyingToas@moz-62AAA429.hsd1.ca.comcast.net) (Quit: Boriss)
- # [18:58] <ejpbruel> not sure who's the right person to ask, so just throwing this question out here
- # [18:58] <treitter> Ms2ger: so all the smart-pointer parts of RefPtr/COMPtr are just operating on the stack-allocated object, only affecting the heap-allocation portion if the ref-count reaches zero?
- # [18:58] * Joins: joey (chatzilla@moz-EFCB4CBF.princetowncable.com)
- # [18:58] <ejpbruel> currently, firefox has a single hidden window. is there a (technical or otherwise) reason we cant support more than one?
- # [18:58] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [18:58] <Ms2ger> Yeah
- # [18:58] * Quits: jwatt (roslea@jwatt.irc.users.mozilla.org) (Input/output error)
- # [18:58] <khuey> that depends on what you mean by 'hidden window'
- # [18:59] * joey is now known as IRCMonkey29569
- # [18:59] <khuey> if you just want a window that's not visible, yeah, we should support that
- # [18:59] <jwir3> if I add a mochitest to layout/style/chrome, do I need to do anything other than add it to the Makefile under _CHROME_TESTS = ... in order to get it to show up in the mochitest browser screen?
- # [18:59] * sheppy-afk is now known as sheppy
- # [18:59] <jwir3> It's copying the files to the correct directory, but it's not showing up when I do runtests.py --chrome --test-path=..,.
- # [18:59] <treitter> Ms2ger: and is NS_IF_ADDREF(searchViewPrepareData.get()) reasonable for my use case, or is there some cleaner way to do this?
- # [19:00] * Joins: TheLink (TheLink@moz-753C1E34.pools.arcor-ip.net)
- # [19:00] <Ms2ger> I guess that's the easiest
- # [19:00] <smaug> treitter: so you need to keep some object alive after the method returns?
- # [19:00] <treitter> OK, works for me
- # [19:00] <treitter> smaug: right
- # [19:00] <smaug> you could also use forget()
- # [19:00] * catlee is now known as catlee-lunch
- # [19:00] * Quits: Honza (chatzilla@E07BF19C.5BB5597D.D0083327.IP) (Connection reset by peer)
- # [19:00] * Quits: smooney (smooney@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [19:00] * Quits: IRCMonkey29569 (chatzilla@moz-EFCB4CBF.princetowncable.com) (Quit: ChatZilla 0.9.87 [Firefox 10.0a2/20111216042030])
- # [19:00] * joduinn-coffee is now known as joduinn-commute
- # [19:01] * Joins: joey (chatzilla@moz-EFCB4CBF.princetowncable.com)
- # [19:01] * Joins: smooney (smooney@moz-BBE3ABD.mv.mozilla.com)
- # [19:01] * Joins: romeo (romeo@moz-A9325888.k459.webspeed.dk)
- # [19:01] <smaug> treitter: that way you wouldn't need to addref manually
- # [19:01] <treitter> smaug: the docs I found for nsRefPtr don't go explain that class - what, exactly, does forget() do?
- # [19:01] * Joins: Honza (chatzilla@E07BF19C.5BB5597D.D0083327.IP)
- # [19:02] * Quits: biesi (cbiesinger@2E747C8D.C1499D0.396E4C6D.IP) (Ping timeout)
- # [19:02] <ejpbruel> khuey: well, the reason i'd want to have this hidden window is so that i can load xul documents on it
- # [19:02] <smaug> treitter: it causes nsRefPtr to drop the pointer to the object
- # [19:02] <smaug> which means Release() isn't called
- # [19:02] * Quits: akeybl (akeybl@moz-52D39FF6.hsd1.ca.comcast.net) (Quit: Leaving...)
- # [19:02] * Joins: rs (rs@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [19:02] <ejpbruel> khuey: currently, we have only one such hidden window, and that leads to conflicts
- # [19:03] * Joins: Mossop (mossop@moz-BBE3ABD.mv.mozilla.com)
- # [19:03] * Quits: rillian (giles@21B7B9F2.B87E9213.6E712CE2.IP) (Ping timeout)
- # [19:03] <khuey> ejpbruel: right, this is why I said it depends on what you mean by "hidden window"
- # [19:03] <treitter> smaug: hmm, in that case, it seems like I'd just be better off manually handling the memory for this object and not wrap it in nsRefPtr
- # [19:03] <smaug> http://mxr.mozilla.org/mozilla-central/source/xpcom/base/nsAutoPtr.h#1006
- # [19:03] <khuey> ejpbruel: I would expect you can just create your own XUL window that's invisible
- # [19:03] <khuey> and load whatever you want in it
- # [19:03] <smaug> treitter: why?
- # [19:04] <treitter> I'm basically just using them to pass some data into the callback. I'm not storing them anywhere else, so managing the memory is pretty simple
- # [19:04] <ejpbruel> khuey: is it possible to do something like that from js?
- # [19:04] <ejpbruel> assuming you'd have chrome privileges
- # [19:04] * Joins: rillian (giles@21B7B9F2.B87E9213.6E712CE2.IP)
- # [19:04] * Joins: espindola (espindola@F2D29657.F60B0462.67AC9B1.IP)
- # [19:05] <khuey> ejpbruel: http://mxr.mozilla.org/mozilla-central/source/xpfe/appshell/public/nsIAppShellService.idl#54 ?
- # [19:05] * Joins: stevee (Miranda@moz-BEBDF855.cable.virginmedia.com)
- # [19:05] <khuey> that's where I'd start
- # [19:05] * Joins: mwu (mwu@moz-59435430.hsd1.nj.comcast.net)
- # [19:05] * Joins: dveditz (dveditz@moz-BBE3ABD.mv.mozilla.com)
- # [19:05] * ChanServ sets mode: +o dveditz
- # [19:05] * Joins: diogogmt (kvirc@moz-4D628198.cpe.net.cable.rogers.com)
- # [19:05] <smaug> treitter: just calling forget() is simple
- # [19:06] * davidb is now known as davidb|mtg
- # [19:06] * Quits: sewardj (sewardj@moz-4388D0F3.themeltingpotedinburgh.org.uk) (Ping timeout)
- # [19:06] <Ms2ger> And hacky
- # [19:07] <ejpbruel> khuey: thanks once again :)
- # [19:07] <smaug> hacky? how?
- # [19:07] <jdm> what does aNotify in AfterSetAttr mean?
- # [19:07] * Joins: gkw (gkw@moz-A7D8CA2A.hsd1.ca.comcast.net)
- # [19:07] <smaug> whether mutationobservers will be notified about the change
- # [19:08] * Quits: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com) (Quit: nhirata)
- # [19:09] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
- # [19:09] * Joins: yuan (ywang@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [19:10] <NeilAway> Waldo: why did you make NS_IF_ADDREF a template instead of NS_ADDREF?
- # [19:10] * Quits: fzzzy (donovan@moz-2B41AF9B.lightspeed.mtvwca.sbcglobal.net) (Quit: fzzzy)
- # [19:10] <Waldo> NeilAway: I don't think I ever touched either...
- # [19:10] <Ms2ger> You added ArrayLength to thel
- # [19:10] <Ms2ger> *them
- # [19:11] <Waldo> hrm, really?
- # [19:11] <jdm> smaug: do you know where tests about crossorigin images live?
- # [19:11] <Waldo> probably just my LASER FOCUS or something
- # [19:11] <treitter> smaug: if I used forget(), when would I actually cause the heap-allocated memory to be freed?
- # [19:11] <treitter> (which I want to do at the end of the callback)
- # [19:11] * Quits: hub (hub@83874EA1.EB7C1AF9.6F478678.IP) (Ping timeout)
- # [19:11] * rail-lunch is now known as rail
- # [19:11] * Joins: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [19:11] <Ms2ger> treitter, when you release it manually at the end
- # [19:12] * Joins: squib (squib-@moz-3F6F2A9C.ep.wisc.edu)
- # [19:12] <treitter> Ms2ger: right. I'm just wondering if wrapping it in nsRefPtr really buys me anything in this case
- # [19:12] * mak is now known as mak|afk
- # [19:12] <treitter> vs. just the single new/delete that I'd do otherwise
- # [19:12] <Ms2ger> Depends if you have early returns, I geuss
- # [19:13] * Joins: micahg (micahg@moz-C7CC8D27.c3-0.arm-ubr1.chi-arm.il.cable.rcn.com)
- # [19:13] <NeilAway> Waldo: oh, I think I see now, you wwanted to make NS_ADDREF a template, but you didn't
- # [19:13] * Quits: Jake (Jake@moz-CDAFE004.phlapafg.dynamic.covad.net) (Quit: Leaving.)
- # [19:13] * Joins: cpeterson (cpeterson@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [19:13] <NeilAway> Waldo: meanwhile it turns out that NS_IF_ADDREF has been a template for 11 years and 10 months, and I only just noticed
- # [19:14] <Ms2ger> Don't tell sheppy
- # [19:14] <smaug> treitter: it gives you better coding style
- # [19:14] * sheppy ignores Ms2ger.
- # [19:14] <smaug> when using manual addreffing releasing, it is easy to forget release if there is early return
- # [19:15] <Waldo> code's read more often than it's written
- # [19:15] <Waldo> using the smart pointer means readers can read more quickly
- # [19:15] <Waldo> quite aside from how the work splits up when writing it
- # [19:16] <treitter> smaug: yeah, I suppose in a general sense, that's true
- # [19:16] <smaug> yeah, smart pointers make code more readable, and less error prone
- # [19:16] <treitter> it shouldn't happen in this case, but I'm trying to get a sense of what's conventional as I learn the codebase
- # [19:17] <treitter> the difficulty is that I'm also calling GLib/GObject-based code, so figuring out the right balance between the two can be tricky
- # [19:17] <treitter> smaug, Ms2ger: thanks for the advice!
- # [19:17] <treitter> I'll probably have more questions later :)
- # [19:17] <Ms2ger> Np
- # [19:19] * Joins: ashughes (ashughes@6202C507.852FE72D.DA78B690.IP)
- # [19:19] * Joins: protz (jonathan@moz-7F6750F6.xulforum.org)
- # [19:20] * Quits: jdm (jdm@moz-9AEDE212.cpe.net.cable.rogers.com) (Client exited)
- # [19:21] * Quits: irving (irving@moz-92D12271.dsl.bell.ca) (Ping timeout)
- # [19:21] * aki is now known as aki|commute
- # [19:21] * Quits: kaze (kaze@moz-A2A3C9E6.vlan402.asr1.cdg3.gblx.net) (Ping timeout)
- # [19:21] * Joins: pcwalton (pcwalton@moz-7B0110AD.mv.mozilla.com)
- # [19:22] * Quits: vingtetun (vingtetun@moz-A2A3C9E6.vlan402.asr1.cdg3.gblx.net) (Ping timeout)
- # [19:22] * adam is now known as adam-afk
- # [19:23] * Joins: tanvi (tanvi@moz-BBE3ABD.mv.mozilla.com)
- # [19:23] * armenzg_lunch is now known as armenzg
- # [19:25] * Joins: dbaron (dbaron@moz-389E0BB7.phlapa.fios.verizon.net)
- # [19:25] * ChanServ sets mode: +o dbaron
- # [19:26] <smaug> I wish there was some flag in patches where requester could indicate priority. like, would they really want to get the patch reviewed before next Aurora
- # [19:26] * Joins: hub (hub@21B7B9F2.B87E9213.6E712CE2.IP)
- # [19:26] <edmorley> khuey++
- # [19:27] * Joins: juanb (jbecerra@moz-BBE3ABD.mv.mozilla.com)
- # [19:27] <khuey> edmorley: hmm?
- # [19:28] * Joins: terrence|away (terrence@moz-BBE3ABD.mv.mozilla.com)
- # [19:28] * terrence|away is now known as terrence
- # [19:28] <Ms2ger> edmorley++
- # [19:28] <Ms2ger> Just because
- # [19:28] <edmorley> lol
- # [19:29] <edmorley> khuey: blog post
- # [19:29] <lmandel> gavin: New contributor has responded to your comments on bug 701182. Do you have time to review the patch today?
- # [19:29] * Joins: timA (tabraldes@moz-535753DA.hsd1.or.comcast.net)
- # [19:29] <khuey> edmorley: ah
- # [19:29] <edmorley> khuey, Ms2ger: how were your weekends? :-)
- # [19:30] <gavin> lmandel: yes
- # [19:30] <Ms2ger> Alright
- # [19:30] <Ms2ger> Enjoyed the time away from the tree with your girlfriend? :)
- # [19:30] * khuey had to move
- # [19:30] <khuey> that was lame
- # [19:30] <khuey> the rest of it was nice
- # [19:31] <Ms2ger> khuey, did you pass, btw?
- # [19:31] * Joins: Boriss (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [19:31] <lmandel> gavin: thx.
- # [19:31] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
- # [19:32] <edmorley> Ms2ger: yeah was good thanks, expected more of a backlog of bugmail today, but didn't exceed 100, so double bonus :-)
- # [19:32] <edmorley> now just to catch up with checkins
- # [19:32] <Ms2ger> Oh, you won't like me that much, then :)
- # [19:32] * Joins: stefanh (stefanh@moz-3EED0162.customers.ownit.se)
- # [19:32] <khuey> Ms2ger: I did
- # [19:32] <Ms2ger> Aww
- # [19:32] <edmorley> \o/
- # [19:33] <Ms2ger> khuey, congratulations, now you have two worthless pieces of paper
- # [19:34] * Joins: karl (karl@82F54078.8998A38C.C8A09C26.IP)
- # [19:34] <philor> where's that bz when I want him to back out?
- # [19:34] <khuey> Ms2ger: actually I don't get the paper for another couple months
- # [19:34] * Joins: damons (gnubeard@moz-BBE3ABD.mv.mozilla.com)
- # [19:34] <Ms2ger> philor, haven't seen him around lately
- # [19:35] <Ms2ger> The security bug?
- # [19:35] * lurking guesses that the uni had to import some more sheep from Austrailia
- # [19:35] <mwu> khuey: did you see planet ate your <canvas> ?
- # [19:35] <khuey> mwu: yeah
- # [19:36] <mwu> k
- # [19:36] * catlee-lunch is now known as catlee
- # [19:36] <khuey> tumblr+planet doesn't work well with html tags
- # [19:36] <mwu> looks like it
- # [19:36] * Joins: jgilbert (jgilbert@moz-BBE3ABD.mv.mozilla.com)
- # [19:36] * Quits: pascalc (chatzilla@moz-B337C47A.rev.sfr.net) (Ping timeout)
- # [19:37] * Joins: pascalc (chatzilla@moz-B337C47A.rev.sfr.net)
- # [19:38] * Joins: aja (chatzilla@A8F7EE0.72086529.7880DB15.IP)
- # [19:38] <philor> and Waldo needs a quick escort to the door, too
- # [19:38] <Waldo> ?
- # [19:38] * Joins: jdm (jdm@moz-9AEDE212.cpe.net.cable.rogers.com)
- # [19:38] <Waldo> oh, hm
- # [19:38] <philor> wonder how that process of hiring full-time sheriffs in sufficient number for 24/7 coverage is going
- # [19:38] * zpao|detached is now known as zpao
- # [19:39] <Ms2ger> philor, your job offer is underway :)
- # [19:39] <Waldo> sure, go ahead, back out
- # [19:39] * Waldo looks more closely at this
- # [19:39] * Quits: mw22_away (chatzilla@moz-FB753258.adsl.wanadoo.nl) (Ping timeout)
- # [19:39] <Waldo> oh, sigh, maybe
- # [19:40] <Waldo> I might know the fix
- # [19:40] * Quits: rillian (giles@21B7B9F2.B87E9213.6E712CE2.IP) (Ping timeout)
- # [19:40] <Waldo> but I'm not confident enough in it to try it
- # [19:40] * AaronMT is now known as AaronMT|afk
- # [19:40] <edmorley> philor: I'm just catching up / updating trees, can back stuff out once that's done if that helps?
- # [19:40] <Waldo> wish this had failed to build locally :-(
- # [19:41] <Waldo> as regards my patches, only the MOZ_FINAL one needs to go
- # [19:41] <Waldo> the build failure is clearly and obviously not from the other patch
- # [19:41] * Joins: rillian (giles@21B7B9F2.B87E9213.6E712CE2.IP)
- # [19:42] <mbrubeck> philor: Are you already working on the backout(s)?
- # [19:42] <mbrubeck> Because I can always tear myself away from this super-exciting Java code to back someone out. :)
- # [19:42] <Waldo> heh
- # [19:43] <jduell> smaug: sicking: what are your thoughts on us landing the WS v13 and unprefixing patches today, even if I don't have the large msg issue resolved? I expect we could get a patch for that aurora+'d
- # [19:43] <jduell> blizzard: ^^
- # [19:43] <Ms2ger> Sounds awful
- # [19:43] * davidb|mtg is now known as davidb
- # [19:44] <sicking> jduell: i'm not sure that you'd get approval for such an aurora patch
- # [19:44] <blizzard> jduell: yeah :(
- # [19:44] <sicking> jduell: can't you just crank the limit up for now?
- # [19:44] <blizzard> jduell: what's the large message problem?
- # [19:45] * Joins: fzzzy (donovan@moz-BBE3ABD.mv.mozilla.com)
- # [19:45] <jduell> sicking: ok, lemme just try to crank out a patch. Hopefully patrick will be around to +r
- # [19:45] <jduell> blizzard: I'd tell you but then I'd have less time to write the patch (we currently limit inbound WS msgs to 16 MB, and increasing has a few issues)
- # [19:45] <Ms2ger> I'd suggest taking the next train
- # [19:45] <blizzard> that's...large
- # [19:46] <jduell> blizzard: sicking: but if it comes down to it, are we happier with exposing a {prefixed, v8} websockets with 16MB limit, or a unprefixed, v13 one with same limit?
- # [19:46] <Ms2ger> Only six weeks, guys
- # [19:46] * mbrubeck starts backing out
- # [19:46] <mbrubeck> since I want to land something on inbound anyway
- # [19:47] <smaug> jduell: just increasing that 16MB for FF11 sounds ok to me
- # [19:47] <sicking> jduell: same here
- # [19:47] <blizzard> jduell: are those things related?
- # [19:47] <blizzard> jduell: unprefixing doesn't mean bug-free
- # [19:47] <smaug> I'd like to see unprefixed soon
- # [19:48] * jimm is now known as jimm-lunch
- # [19:48] <jduell> smaug: sicking: I can write a one-liner patch that increases from 16-32, but I'm not sure how gracefully it will handle things when malloc fails. It would use moz_xrealloc, which is "infallible": how does that actually handle memory failure?
- # [19:48] <Ms2ger> I'd like to see it prefixed early in a cycle, not on the last day
- # [19:48] <jduell> or from 16-> your choice of big size.
- # [19:48] <khuey> jduell: it crashes
- # [19:48] <smaug> since I think web devs think that Chrome has "stable" websocket impl, but gecko doesn't have. (IRCCloud for example doesn't support MozWebSocket)
- # [19:49] <jduell> Yeah, then I need at least a slightly more complex patch. And it'd be nice to avoi the current polynomial realloc/memcpy logic that's in there.
- # [19:49] * Quits: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [19:49] * Quits: florian (florian@moz-87C33FDA.kimsufi.com) (Quit: Instantbird 1.2a1pre)
- # [19:50] <Waldo> "infallible" is the BS name for "you will die if it fails" like "retention policy" is the BS name for "deletion policy"
- # [19:50] <jduell> blizzard: smaug sicking : but my above question remains. We speak v13 of the protocol (and the patches for that are trivial), so we could land that. And we could unprefix it we want.
- # [19:50] * Quits: pascalc (chatzilla@moz-B337C47A.rev.sfr.net) (Ping timeout)
- # [19:50] <mbrubeck> Maybe Erlang would have had more marketing success if they called it "infallible programming" instead of "crash-only programming"
- # [19:51] * mfinkle waits for mbrubeck before landing on m-i
- # [19:51] <jduell> all that is orthoganal in some sense (maybe not a PR sense, though) to the large msg issue
- # [19:51] <blizzard> jduell: it's just a limit
- # [19:51] <mbrubeck> mfinkle: and.... done!
- # [19:51] <blizzard> jduell: do other browsers have limits?
- # [19:51] <blizzard> jduell: since it's message based it still has to buffer
- # [19:51] <blizzard> jduell: I imagine they do
- # [19:51] <blizzard> jduell: I'd rather have the same limit, assuming there is one
- # [19:52] <jduell> blizzard: the spec allows us to barf if our "buffer" is full, and hixie has told me that can be whenever we like
- # [19:52] <smaug> if Chrome has, it is larger than what we have for incoming messages
- # [19:52] <smaug> IIRC
- # [19:52] <blizzard> can you guys ask if they have one?
- # [19:52] <blizzard> I'd like to syncronize
- # [19:52] <blizzard> on what the limit is
- # [19:52] <jduell> I'm not positive, but I suspect chrome is supporting much larger msgs than 16MB
- # [19:52] * Joins: mcsmurf (mcsmurf@moz-5BDA8BFA.dip.t-dialin.net)
- # [19:52] <blizzard> (delivering a 16MB message is lol)
- # [19:52] * Joins: By-Tor (bytor@moz-46974D0B.dyn.optonline.net)
- # [19:53] <smaug> the message could be a file
- # [19:53] <mbrubeck> Can someone with security privs tell bug 624621 that it has been backed out on inbound?
- # [19:53] <jduell> I know smaug/sicking want us to have just a "as big as malloc will give us" limit
- # [19:53] <sicking> jduell: i wouldn't worry too much about the infallibility, putting the limit at 128MB or so seems ok
- # [19:53] <smaug> in which case 16MB isn't that much
- # [19:53] * aki|commute is now known as aki
- # [19:53] <Waldo> mbrubeck: aww, you backed out both, not the one that clearly was offending :-\
- # [19:53] <blizzard> smaug: can it?
- # [19:53] <jduell> sicking: right, but mobile might well drop dead with such a limit
- # [19:53] <mbrubeck> Waldo: Sorry, feel free to re-land
- # [19:53] * Joins: akeybl (akeybl@moz-BBE3ABD.mv.mozilla.com)
- # [19:53] <blizzard> smaug: I thought we were limited to text or binary data, not references
- # [19:53] * Waldo readds both locally, he guesses, and fixes just the one
- # [19:53] * Quits: stefanh (stefanh@moz-3EED0162.customers.ownit.se) (Ping timeout)
- # [19:53] <blizzard> smaug: (note: I'm a fan of a file-based api there for big stuff)
- # [19:53] <mwu> that's kinda scary considering os's like to give you as much as you want until you actually try to use it
- # [19:54] <sicking> jduell: can you simply make a patch that changes infallible malloc to fallible
- # [19:54] <smaug> blizzard: you can send blobs
- # [19:54] <blizzard> mwu: you can fix that
- # [19:54] <sicking> jduell: and get that reviewed today?
- # [19:54] <philor> mbrubeck: sorry, yeah, I was talking rather than backing out because I'm at work without a tree
- # [19:54] * Quits: redfive (chatzilla@moz-B94E1A81.ipnetworksinc.net) (Ping timeout)
- # [19:54] <jduell> sicking: yes, probably. Maybe that's the fix. I'll write a dumb one first, and then see if I can avoid n-squared memcpys
- # [19:55] <sicking> jduell: sounds great!
- # [19:55] <blizzard> mwu: just write to it, catch segfaults :)
- # [19:56] * adam-afk is now known as adam
- # [19:56] <mwu> blizzard: yeah... I mean, I've done that before sure, but I don't want everyone else to :p
- # [19:56] * AaronMT|afk is now known as AaronMT
- # [19:56] * Joins: kaze (kaze@moz-79F78BFC.w90-46.abo.wanadoo.fr)
- # [19:57] <mwu> though I suspect we need that regardless to catch problems in mmap'd files
- # [19:57] <blizzard> mwu: :)
- # [19:57] <smaug> jduell: sicking: blizzard: I don't see any limits in webkit. Just: fail("Ran out of memory while receiving WebSocket data.");
- # [19:58] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
- # [19:58] <smaug> hmm, maxPayloadLength = 0x7FFFFFFFFFFFFFFF
- # [19:59] * Quits: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [20:00] * Joins: peregrino (peregrino@moz-96A63A2D.telecom.net.ar)
- # [20:00] * Quits: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [20:01] * Joins: jimb (user@moz-F4EC06CC.hsd1.or.comcast.net)
- # [20:01] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [20:01] * Waldo pushes his bustage fixes to try
- # [20:01] * Joins: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP)
- # [20:02] <Jesse> Waldo: i don't like the "infallible malloc" term either. do you have an alternative?
- # [20:02] * Joins: stefanh (stefanh@moz-3EED0162.customers.ownit.se)
- # [20:02] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [20:02] <Waldo> dunno
- # [20:02] * Joins: sewardj (sewardj@moz-C79D927B.cable.virginmedia.com)
- # [20:02] <Jesse> i do like the concept
- # [20:03] * Quits: smooney (smooney@moz-BBE3ABD.mv.mozilla.com) (Quit: smooney)
- # [20:03] <mbrubeck> Jesse: "crash-only malloc" to steal the Erlang terminology...
- # [20:03] * Quits: kaze (kaze@moz-79F78BFC.w90-46.abo.wanadoo.fr) (Ping timeout)
- # [20:03] <mwu> failfast malloc
- # [20:03] <mwu> death malloc
- # [20:03] <mccr8> malloc or die?
- # [20:04] <Jesse> can't call it death malloc, that's the name of my band
- # [20:04] <mbrubeck> Jesse: Since you're here, could you tell bug 624621 that it's been backed out on inbound (if someone else hasn't gotten to it yet)
- # [20:04] <mbrubeck> ?
- # [20:04] * Quits: hub (hub@21B7B9F2.B87E9213.6E712CE2.IP) (Ping timeout)
- # [20:04] <catlee> malloc of antioch
- # [20:04] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
- # [20:05] <Jesse> mbrubeck: commented
- # [20:05] <mbrubeck> thanks
- # [20:05] * Quits: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com) (Quit: mccr8)
- # [20:05] * Joins: sworkman-phone (sworkman@moz-BBE3ABD.mv.mozilla.com)
- # [20:05] <jwir3> is it possible to make a reftest wait before taking a snapshot? Like if I want to do something quick to the DOM before the snapshot is taken?
- # [20:05] * Quits: sworkman-phone (sworkman@moz-BBE3ABD.mv.mozilla.com) (Quit: sworkman-phone)
- # [20:06] * Quits: Jesse (jruderman@moz-537BCF9.hsd1.ca.comcast.net) (Quit: Jesse)
- # [20:06] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [20:07] <smaug> jwir3: yes
- # [20:07] * Quits: mike5w3c (MikeS@moz-5EC55E5A.pool.e-mobile.ne.jp) (Ping timeout)
- # [20:07] * joduinn-commute is now known as joduinn-mtg
- # [20:07] * Joins: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com)
- # [20:07] * smaug tries to remember what attribute or class name is needed for that
- # [20:07] * Joins: mike5w3c (MikeS@moz-82086631.pool.e-mobile.ne.jp)
- # [20:07] * Joins: smooney (smooney@moz-BBE3ABD.mv.mozilla.com)
- # [20:07] <dholbert> jwir3, reftest-wait
- # [20:08] <jwir3> dholbert, smaug: Awesome. thanks!
- # [20:08] <dholbert> jwir3, <html class="reftest-wait">
- # [20:08] <jwir3> perfect
- # [20:08] <dholbert> and then when you're ready for the snapshot, document.documentElement.removeAttribute("class")
- # [20:08] <jwir3> sweet
- # [20:08] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
- # [20:09] * Quits: Ami_Ty (Amie@moz-BBE3ABD.mv.mozilla.com) (Connection reset by peer)
- # [20:09] * Joins: Ami_Ty_ (Amie@moz-BBE3ABD.mv.mozilla.com)
- # [20:09] <lmandel> jprmc, jrmuizel: https://metrics.mozilla.com/pentaho/content/pentaho-cdf-dd/Render?solution=metrics2&path=%2Ftelemetry&file=telemetryHistogram.wcdf
- # [20:10] * Joins: fs (Elchi3@B9C9103E.56629902.2EC4CA51.IP)
- # [20:11] * Quits: andreasn (andreasn@moz-BB6A8755.bredband.comhem.se) (Ping timeout)
- # [20:11] * Joins: MarcoZ_ (marco.zehe@moz-5F68A19.dip.t-dialin.net)
- # [20:12] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
- # [20:12] * Quits: MarcoZ (marco.zehe@moz-5F68A19.dip.t-dialin.net) (Ping timeout)
- # [20:12] * MarcoZ_ is now known as MarcoZ
- # [20:12] <espindola> joe: for 708051, your preference is then to s/ipc/gfxipc/?
- # [20:13] * Joins: hub (hub@21B7B9F2.B87E9213.6E712CE2.IP)
- # [20:14] * Joins: caillon (caillon@moz-186D75E2.hsd1.ca.comcast.net)
- # [20:14] * Joins: lurking_work (chatzilla@moz-107FCDBA.hfc.comcastbusiness.net)
- # [20:16] * Quits: hub (hub@21B7B9F2.B87E9213.6E712CE2.IP) (Ping timeout)
- # [20:16] <jdm> ehsan: remember how I mentioned that I pushed a PSM bug at a volunteer? he posted a patch yesterday :D
- # [20:16] <Ms2ger> jdm, cruel
- # [20:16] <ehsan> jdm: awesome!!!
- # [20:17] <ehsan> what was his second patch?
- # [20:17] <ehsan> psm again?
- # [20:17] <jdm> hmm?
- # [20:17] <jdm> he did some e10s/memshrink patches too
- # [20:18] <ehsan> oh ok
- # [20:18] * Ms2ger wonders where bholley is
- # [20:18] <Waldo> "sm" stands for sadomasochism, right?
- # [20:18] <ehsan> I thought that he had only worked on psm
- # [20:18] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
- # [20:18] <mcpherrin> I just ran into a bug where the mouse cusor disappears on my mac when Firefox has focus. What should I file that under?
- # [20:19] <gavin> "strange occurences"
- # [20:19] * Joins: Mardak (Mardak@moz-BBE3ABD.mv.mozilla.com)
- # [20:19] * Joins: jet (junglecode@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [20:19] <smaug> mcpherrin: search for existing bugs
- # [20:20] <dholbert> possibly widget: cocoa
- # [20:20] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Quit: sworkman)
- # [20:20] <lurking_work> mcpherrin: tried safe-mode ?
- # [20:20] * Quits: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [20:20] * Quits: jdm (jdm@moz-9AEDE212.cpe.net.cable.rogers.com) (Client exited)
- # [20:20] <mcpherrin> lurking_work: Well, I'm trying to repro first, haven't quite managed yet :P
- # [20:21] * lurking_work hates intermittants
- # [20:21] <bsmedberg> mcpherrin: are you using synergy?
- # [20:21] * bsmedberg sees that a lot when using synergy, assumed it was a synergy bug
- # [20:22] * sheppy is now known as sheppy-afk
- # [20:22] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
- # [20:22] * Quits: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com) (Quit: mccr8)
- # [20:22] * Quits: jet (junglecode@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Quit: jet)
- # [20:22] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
- # [20:22] * Quits: gkw (gkw@moz-A7D8CA2A.hsd1.ca.comcast.net) (Quit: Instantbird 1.1)
- # [20:22] <mcpherrin> I have a feeling it has to do with flash, somehow
- # [20:22] <mcpherrin> Since it happened while watching air mozilla
- # [20:22] * mak|afk is now known as mak
- # [20:23] * jwir3 is now known as jwir3|lunch
- # [20:23] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
- # [20:23] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Input/output error)
- # [20:23] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
- # [20:24] * Joins: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com)
- # [20:24] * Quits: vladan (vladan@F2D29657.F60B0462.67AC9B1.IP) (Quit: Leaving.)
- # [20:24] * Quits: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [20:24] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
- # [20:24] * Joins: jet (junglecode@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [20:25] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
- # [20:25] <Waldo> hm, and I forgot the status meeting again :-\
- # [20:25] <Waldo> smrt
- # [20:26] <khuey> mmm me too
- # [20:27] * bc is now known as bc|afk
- # [20:27] <Ms2ger> Which meeting?
- # [20:27] * Joins: AutomatedTester (AutomatedT@moz-352F716D.ppp.as43234.net)
- # [20:27] * mattwoodrow|away is now known as mattwoodrow
- # [20:28] * jimm-lunch is now known as jimm
- # [20:28] * Quits: roc (chatzilla@C0ACF8B.5E1E9EEA.613E47D1.IP) (Ping timeout)
- # [20:29] * Joins: redfive (chatzilla@moz-B94E1A81.ipnetworksinc.net)
- # [20:29] <mbrubeck> philor: do we have a bug yet for "Reboot Device interrupted" from twisted.internet.error.ConnectionLost? https://tbpl.mozilla.org/php/getParsedLog.php?id=8037058&tree=Mozilla-Inbound
- # [20:29] <Ms2ger> i;r
- # [20:30] <Waldo> Ms2ger: probably not a bad reaction
- # [20:30] <philor> mbrubeck: "clumps" because at first they were just clumped
- # [20:30] <mbrubeck> That's what I've been doing, but it's starting to seem like a common failure mode
- # [20:30] <Waldo> just kind of like to keep minimal tabs on everything
- # [20:30] <Ms2ger> I keep heaps of tabs already
- # [20:30] <mbrubeck> philor: oh yeah, bug 711725
- # [20:32] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
- # [20:33] * Quits: fs (Elchi3@B9C9103E.56629902.2EC4CA51.IP) (Quit: Leaving)
- # [20:33] * Quits: jfkthame_afk (jfkthame@78C7B4BE.A1C12133.9542EC20.IP) (Ping timeout)
- # [20:34] * ctalbert is now known as ctalbert|brb
- # [20:34] * Joins: jfkthame (jfkthame@CE8EB10E.90783722.9542EC20.IP)
- # [20:34] * Joins: pnemsak (Miranda@moz-7409BD53.orange.sk)
- # [20:34] * Joins: vladan (vladan@F2D29657.F60B0462.67AC9B1.IP)
- # [20:34] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
- # [20:35] * Joins: hub (hub@21B7B9F2.B87E9213.6E712CE2.IP)
- # [20:35] * Quits: MarcoZ (marco.zehe@moz-5F68A19.dip.t-dialin.net) (Quit: See you tomorrow!)
- # [20:36] * Quits: ericb2 (X@moz-9C4C3DED.fbx.proxad.net) (Quit: . . . ........)
- # [20:39] * armenzg is now known as armenzg_brb
- # [20:39] * Quits: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Input/output error)
- # [20:39] * Joins: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [20:40] * Quits: caillon (caillon@moz-186D75E2.hsd1.ca.comcast.net) (Ping timeout)
- # [20:40] * Joins: caillon (caillon@moz-186D75E2.hsd1.ca.comcast.net)
- # [20:40] * ctalbert|brb is now known as ctalbert
- # [20:41] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
- # [20:42] * Joins: anant_ (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [20:42] * Quits: anant_ (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Input/output error)
- # [20:42] * Joins: anant_ (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [20:42] * Quits: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Connection reset by peer)
- # [20:43] * anant_ is now known as anant
- # [20:44] <Pike> bear-buildduty: I'm a tad confused about my push to try, https://build.mozilla.org/buildapi/self-serve/try/rev/eb4c53d9e9bb shows a plethora of test builds, while https://tbpl.mozilla.org/?tree=Try&rev=eb4c53d9e9bb is rather short with just M. Is that my bad? (haven't touched try and real code in a while)
- # [20:44] * Quits: robhawkes (robhawkes@moz-33A339B7.dsl.cnl.uk.net) (Quit: Linkinus - http://linkinus.com)
- # [20:45] * Joins: wolfiR_ (wolfiR@moz-145F9C08.dip0.t-ipconnect.de)
- # [20:45] <bear-buildduty> pike - not your bad, just tbpl being behind in what tests are running/done
- # [20:45] * Quits: janv (varga@moz-C1261AFF.flarion.as5628.telecom.sk) (Ping timeout)
- # [20:45] <bear-buildduty> it can take 15-30 minutes to catch up (because of the data it reads to know that)
- # [20:46] <philor> Pike: oh, and you can't have those mobile-desktop builds anymore
- # [20:46] <Pike> philor: yeah, lsblakk told me, and now removed them from trychooser
- # [20:47] <philor> and the things on native other than mochitests are hidden, because they don't run
- # [20:47] * Joins: janv (varga@moz-C1261AFF.flarion.as5628.telecom.sk)
- # [20:47] <philor> or rather, they run, and always fail, no matter what
- # [20:47] <Ms2ger> Sounds like mochitests
- # [20:47] <Pike> philor: awesome, thanks
- # [20:48] <philor> so https://tbpl.mozilla.org/?tree=Try&rev=eb4c53d9e9bb&noignore=1 and have fun interpreting whether or not you changed any of the failures :)
- # [20:48] <Pike> nah, if folks say "failure is known" I'm good
- # [20:48] <Pike> also, if the thing comes up at all, it should be fine, given that I just shuffle l10n files around
- # [20:49] * glob is now known as glob|away
- # [20:50] * jaws|away is now known as jaws
- # [20:52] * Joins: rniwa (rniwa@5CA6DC39.C60FE7DC.4065847B.IP)
- # [20:56] * Quits: Asa (u3626@moz-160C58C6.com) (Max SendQ exceeded)
- # [20:56] * Quits: jaws (u2871@moz-160C58C6.com) (Max SendQ exceeded)
- # [20:56] * Quits: felipe (u2734@moz-160C58C6.com) (Max SendQ exceeded)
- # [20:56] * Joins: Asa (u3626@moz-160C58C6.com)
- # [20:56] * Joins: felipe (u2734@moz-160C58C6.com)
- # [20:57] * mcote is now known as mcote|bbiab
- # [20:57] * Joins: jaws (u2871@moz-160C58C6.com)
- # [20:57] * Joins: cpearce (chatzilla@moz-55A2BC7D.xdsl.xnet.co.nz)
- # [21:00] * Joins: kaze (kaze@moz-79F78BFC.w90-46.abo.wanadoo.fr)
- # [21:00] * Quits: cpearce (chatzilla@moz-55A2BC7D.xdsl.xnet.co.nz) (Ping timeout)
- # [21:00] * Quits: smooney (smooney@moz-BBE3ABD.mv.mozilla.com) (Quit: smooney)
- # [21:01] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [21:01] * Joins: andreasn (andreasn@moz-3CDFCC6A.a336.corp.bahnhof.se)
- # [21:02] * armenzg_brb is now known as armenzg
- # [21:02] * Joins: gkw (gkw@moz-BBE3ABD.mv.mozilla.com)
- # [21:02] * Quits: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [21:02] * Joins: smooney (smooney@moz-BBE3ABD.mv.mozilla.com)
- # [21:03] * Joins: Theo (chatzilla@moz-499C8051.w90-48.abo.wanadoo.fr)
- # [21:03] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
- # [21:03] * Quits: gkw (gkw@moz-BBE3ABD.mv.mozilla.com) (Quit: Instantbird 1.1)
- # [21:03] * Joins: gkw (gkw@moz-BBE3ABD.mv.mozilla.com)
- # [21:04] * Quits: kaze (kaze@moz-79F78BFC.w90-46.abo.wanadoo.fr) (Ping timeout)
- # [21:06] * ddahl|away is now known as ddahl
- # [21:06] * catlee is now known as catlee-mtg
- # [21:06] * Quits: michal (michal@moz-5C5A58A8.broadband14.iol.cz) (Quit: Leaving.)
- # [21:08] * coop is now known as coop|mtg
- # [21:09] * zpao is now known as zpao|detached
- # [21:09] * bear-buildduty is now known as bear-buildduty-afk
- # [21:13] * nthomas|away is now known as nthomas
- # [21:13] <dholbert> dbaron, if I have multiple anonymous blocks in the same container, can I use the same styleContext (generated by ResolveAnonymousBoxStyle) for them all, or should I call that function to get a separate styleContext for each of 'em?
- # [21:14] <@dbaron> dholbert, you can use the same style context for each of them
- # [21:14] <dholbert> dbaron, cool, thanks
- # [21:14] <@dbaron> dholbert, the resolution function will share for you, actually
- # [21:14] <@dbaron> dholbert, as long as the parameters you'd have passed to ResolveAnonymousBoxStyle (the parent and the anon box) would be the same
- # [21:15] * Quits: jhammink (textual@moz-7B0110AD.mv.mozilla.com) (Ping timeout)
- # [21:15] <dholbert> dbaron, yup, they'd be the same. Ok, thanks
- # [21:15] * Joins: felipe|tab (AndChat@moz-BBE3ABD.mv.mozilla.com)
- # [21:15] * Quits: peregrino (peregrino@moz-96A63A2D.telecom.net.ar) (Ping timeout)
- # [21:16] * Quits: jgilbert (jgilbert@moz-BBE3ABD.mv.mozilla.com) (Connection reset by peer)
- # [21:16] * Joins: jgilbert (jgilbert@moz-BBE3ABD.mv.mozilla.com)
- # [21:17] <bsmedberg> jlebar: ShutdownObserver<SmartPtr> won't stringify differently because it's a template, will it?
- # [21:17] * kumar is now known as kumar|lunch
- # [21:17] * Joins: jhammink (textual@moz-BBE3ABD.mv.mozilla.com)
- # [21:17] <jlebar> bsmedberg, what do you mean by stringify?
- # [21:17] <bsmedberg> jlebar: the # operator for macros
- # [21:18] <jlebar> Oh, I sure hope not! But I'll check.
- # [21:18] * Joins: peregrino (peregrino@moz-96A63A2D.telecom.net.ar)
- # [21:18] <bsmedberg> jlebar: I mean, the basic error here is that the initial version just uses "ShutdownObserver"
- # [21:18] <bsmedberg> so it's recorded as the same class each time
- # [21:18] <jlebar> bsmedberg, Ah, I see. Sure.
- # [21:19] <bsmedberg> but I'm a little surprised that the other version would be stringifying as ShutdownObserver<Foo> instead of ShutdownObserver<SmartPtr>
- # [21:19] * jhammel is now known as jhammel|lunch
- # [21:19] * bsmedberg writes a testcase
- # [21:19] * Joins: Boriss_ (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [21:20] <jlebar> bsmedberg, Hm, yes, I bet it stringifies as "ShutdownObserver<SmartPtr>", because macros happen before templates.
- # [21:20] * khuey grumbles
- # [21:20] <khuey> it would be nice if the CRT told you why it was asserting
- # [21:20] * Quits: Boriss (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
- # [21:20] * Boriss_ is now known as Boriss
- # [21:20] <bsmedberg> jlebar: then I don't understand why your patch fixed anything
- # [21:20] * zpao|detached is now known as zpao
- # [21:20] * Quits: kdcw (kdc@moz-F7413045.pk.shawcable.net) (Quit: HydraIRC -> http://www.hydrairc.com <- Go on, try it!)
- # [21:21] <jlebar> bsmedberg, Was it a problem that the QI implementation stringified as "ShutdownObserver<SmartPtr>" while addref/release used "ShutdownObserver"?
- # [21:21] * Joins: michal (michal@moz-5C5A58A8.broadband14.iol.cz)
- # [21:21] * Quits: sicking (chatzilla@moz-1FF05400.hsd1.ca.comcast.net) (Ping timeout)
- # [21:21] <bsmedberg> jlebar: IIUC, the assertion you're seeing is that
- # [21:22] <bsmedberg> we first saw "ShutdownObserver" with sizeof(M)
- # [21:22] <bsmedberg> and then later saw it with sizeof(N)
- # [21:22] <jlebar> well, they should all have the same size.
- # [21:22] <bsmedberg> well, what was the asserted case?
- # [21:23] <jlebar> There were two conditions in the assertion, right?
- # [21:23] <bsmedberg> aInstanceSize == 0 || entry->GetClassSize() == aInstanceSize
- # [21:24] <mbrubeck> BenWa, mattwoodrow: burning on inbound...
- # [21:24] <BenWa> mbrubeck: ok, got it
- # [21:24] <jlebar> So it could be that entry->GetClassSize() is 0, but aInstanceSize is non-zero?
- # [21:24] <jlebar> bsmedberg, fwiw, I think ClearOnShutdown gets called just once in that mochitest before it first asserts.
- # [21:25] <bsmedberg> jlebar: it could be, yes. If that's the case then either somebody else has a class "ShutdownObserver"
- # [21:25] <bsmedberg> or we're passing inconsistent data
- # [21:25] <bsmedberg> but in fact there appear to be several other ShutdownObservers
- # [21:25] <bsmedberg> so I guess it's just a name collision
- # [21:25] <BenWa> mattwoodrow: I'll just back out everything I guess?
- # [21:26] <jlebar> bsmedberg, I see just one, in GfxInfoBase.
- # [21:26] * Joins: whimboo (whimboo@moz-216F9AFA.superkabel.de)
- # [21:26] <mattwoodrow> BenWa: I guess so, to be safe
- # [21:26] <jlebar> But in any case, yes.
- # [21:26] <bsmedberg> jlebar: you could probably use mozilla::ShutdownObserver to avoid the collision
- # [21:26] <khuey> bsmedberg: do you know how to convince MSVC to load the source for the CRT?
- # [21:26] <bsmedberg> in any case, I can r+ as-is
- # [21:26] <khuey> so that I can step through it?
- # [21:27] <jlebar> bsmedberg, okay. Glad we understand this now!
- # [21:27] <jlebar> bsmedberg, thanks for looking.
- # [21:27] <vlad> khuey: vs2010?
- # [21:27] <khuey> vlad: yep
- # [21:27] <bsmedberg> khuey: a prebuilt CRT or a custom-built one?
- # [21:27] <khuey> bsmedberg: the prebuilt debug crt
- # [21:27] <vlad> I think that if you sto inside the crt it'll prompt you for a file
- # [21:27] <bsmedberg> I don't know that I've ever done that.
- # [21:27] <mbrubeck> jmaher: could bug 710877 have broken reftests on Android?
- # [21:27] <vlad> just point it at the right file in the crt src dir
- # [21:27] <vlad> and it should find the rest automatically
- # [21:27] <bsmedberg> but vlad may be correct
- # [21:28] * Joins: jwq (jwq@moz-8238A1D6.irl.cri.nz)
- # [21:28] <khuey> vlad: it's just giving me the disasm
- # [21:28] <khuey> no prompts
- # [21:28] <vlad> mm, you may have disabled the missing-source prompting?
- # [21:28] <jmaher> mbrubeck: yes it is possible; although I recall testing it over the weekend
- # [21:28] * Quits: sewardj (sewardj@moz-C79D927B.cable.virginmedia.com) (Ping timeout)
- # [21:28] <vlad> khuey: but usually that's all it takes for me
- # [21:29] <vlad> khuey: I guess another thing to check is to look at the modules window
- # [21:29] * stefanh is now known as stefanh|away
- # [21:29] <mbrubeck> jmaher: Every reftest/crashtest/jsreftest since it landed has failed with repeated DeviceManager: error pulling file '/mnt/sdcard/tests/reftest/reftest.log': No such file or directory
- # [21:29] <vlad> and make sure it says that you have symbols for the crt
- # [21:29] <mbrubeck> like https://tbpl.mozilla.org/php/getParsedLog.php?id=8038033&tree=Mozilla-Inbound
- # [21:29] <jmaher> mbrubeck: most likely it broke it
- # [21:29] <mbrubeck> jmaher: Not 100% certain of the blame because it landed while Android was busted. Okay if I back it out to see if that fixes it?
- # [21:30] <edmorley> mattwoodrow, BenWa: inbound windows burning
- # [21:30] <jmaher> mbrubeck: yes, go ahead
- # [21:30] <mbrubeck> okay, thanks
- # [21:30] <BenWa> edmorley: Yup, looking the command to back out a range, the wiki isn't very useful
- # [21:30] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [21:30] <edmorley> BenWa: use https://wiki.mozilla.org/User:Mak77 :-)
- # [21:31] <BenWa> edmorley: looks a bit scary
- # [21:31] <edmorley> BenWa: I can do for now if you like?
- # [21:31] <BenWa> edmorley: If you don't mind
- # [21:31] <mbrubeck> edmorley: Can you back out 4f44ef4d21fb for Android test failures while you're at it?
- # [21:31] <edmorley> mbrubeck: sure
- # [21:32] <mbrubeck> thanks!
- # [21:32] * Quits: Unfocused (u2622@moz-160C58C6.com) (Max SendQ exceeded)
- # [21:33] * Joins: cpearce (chatzilla@538BABFE.A073F3E.97BBD552.IP)
- # [21:33] * Joins: Unfocused (u2622@moz-160C58C6.com)
- # [21:33] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [21:33] * sheppy-afk is now known as sheppy
- # [21:34] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [21:35] * Parts: whimboo (whimboo@moz-216F9AFA.superkabel.de)
- # [21:36] * bear-buildduty-afk is now known as bear-buildduty
- # [21:36] <edmorley> mbrubeck, BenWa: done :-)
- # [21:36] <BenWa> edmorley: thanks
- # [21:36] <edmorley> np
- # [21:37] * mkelly|wfh is now known as mkelly|gone
- # [21:37] * Quits: fzzzy (donovan@moz-BBE3ABD.mv.mozilla.com) (Quit: fzzzy)
- # [21:38] <imphil> if someone lands something, could you please take bug 693615 with it? (small and low-risk)
- # [21:39] * juanb is now known as juab|brb
- # [21:43] * Quits: rs (rs@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
- # [21:44] <khuey> vlad: so the "browse to find source" thing on the "no source available" windows is greyed out :-(
- # [21:44] * Quits: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
- # [21:44] <khuey> vlad: the modules thing says "symbols loaded (source information stripped)"
- # [21:45] <bsmedberg> khuey: they the debug CRT doesn't have source info! yummy
- # [21:45] <bsmedberg> you can build your own...
- # [21:46] * Quits: DGMurdockIII (dgmurdocki@moz-E933B63.hsd1.il.comcast.net) (Quit: get satisfied! :: core-networks.de ««« (Gamers.IRC) »»» gamersirc.net ::)
- # [21:46] <khuey> bsmedberg: gah
- # [21:46] <khuey> bsmedberg: is that even possible with 2010?
- # [21:46] <bsmedberg> oh, I don't know
- # [21:47] * Joins: peregrino_ (peregrino@moz-96A63A2D.telecom.net.ar)
- # [21:48] <khuey> bsmedberg: http://connect.microsoft.com/VisualStudio/feedback/details/374834/source-information-stripped-from-pdb-files-for-msvcr90-dll-and-msvcr90d-dll :-(
- # [21:48] * Quits: peregrino (peregrino@moz-96A63A2D.telecom.net.ar) (Ping timeout)
- # [21:48] * peregrino_ is now known as peregrino
- # [21:49] * Quits: smooney (smooney@moz-BBE3ABD.mv.mozilla.com) (Quit: smooney)
- # [21:49] <edmorley> mbrubeck: about to trigger pgo on inbound tip, given the current set is on the busted push, unless you've already done so?
- # [21:49] <mbrubeck> I haven't
- # [21:49] <mbrubeck> (I triggered the busted one, before I noticed the android bustage)
- # [21:49] <edmorley> ah
- # [21:49] <edmorley> done
- # [21:49] * Joins: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP)
- # [21:50] * khuey sighs
- # [21:50] <khuey> maybe I should just rebuild with 2005
- # [21:50] * Joins: rs (rs@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [21:51] * Joins: irving (irving@moz-92D12271.dsl.bell.ca)
- # [21:52] <mbrubeck> though I guess the PGO run on the busted changeset would actually be enough, since the only stuff on top of it is a backed-out landing and a test-only backout.
- # [21:53] * Quits: erione (erione@67A3F00C.DE807506.C752B3FA.IP) (Quit: Leaving)
- # [21:53] * kumar|lunch is now known as kumar
- # [21:53] * Joins: smooney (smooney@moz-BBE3ABD.mv.mozilla.com)
- # [21:54] * Joins: fzzzy (donovan@moz-2B41AF9B.lightspeed.mtvwca.sbcglobal.net)
- # [21:55] * jhammel|lunch is now known as jhammel
- # [22:00] * mcote|bbiab is now known as mcote
- # [22:01] * Joins: yuan_ (ywang@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [22:01] * jwir3|lunch is now known as jwir3
- # [22:01] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
- # [22:02] * Joins: jdm (jdm@moz-9AEDE212.cpe.net.cable.rogers.com)
- # [22:02] <vlad> khuey: can't build your own for vs2010.. but that's really weird
- # [22:02] <vlad> i swear i've done this
- # [22:02] * Quits: yuan (ywang@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
- # [22:02] * yuan_ is now known as yuan
- # [22:02] * Quits: wolfiR_ (wolfiR@moz-145F9C08.dip0.t-ipconnect.de) (Ping timeout)
- # [22:03] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
- # [22:03] <Ms2ger> By the way, if half the pushes on your project branch are merges, you don't need a project branch
- # [22:05] * zpao is now known as zpao|detached
- # [22:05] * Quits: rshetty (quassel@F4613178.6B56F8F4.D30E9BEF.IP) (Quit: No Ping reply in 180 seconds.)
- # [22:05] * Joins: rshetty (quassel@F4613178.6B56F8F4.D30E9BEF.IP)
- # [22:06] * Quits: Boriss (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Client exited)
- # [22:06] * Joins: Boriss (FlyingToas@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [22:07] <catlee-mtg> Ms2ger: who?
- # [22:07] <khuey> vlad: the internet gave me a tip, trying it now
- # [22:08] * Joins: mgoodwin_ (mgoodwin@moz-4BB9E5E9.cable.virginmedia.com)
- # [22:09] * Joins: tanvi1 (tanvi@moz-BBE3ABD.mv.mozilla.com)
- # [22:09] * Quits: tanvi (tanvi@moz-BBE3ABD.mv.mozilla.com) (Connection reset by peer)
- # [22:10] * Quits: mgoodwin (mgoodwin@moz-4BB9E5E9.cable.virginmedia.com) (Ping timeout)
- # [22:10] * Quits: past (past@moz-7A93B333.dsl.dyn.forthnet.gr) (Input/output error)
- # [22:10] * Quits: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [22:10] * Joins: benadida (ben@moz-BBE3ABD.mv.mozilla.com)
- # [22:12] * Quits: jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP) (Ping timeout)
- # [22:14] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
- # [22:14] * Joins: jprmc (jprmc@F2D29657.F60B0462.67AC9B1.IP)
- # [22:14] * irving is now known as irving|away
- # [22:16] * Joins: florian (florian@moz-87C33FDA.kimsufi.com)
- # [22:16] <edmorley> catlee-mtg: presume the s-c merge that's just happened into m-c, where of the 45 csets merged, there were 23 "Merge m-c to s-c."
- # [22:17] <gps> edmorley: yes, it was
- # [22:17] <gps> did I do something wrong with the merge?
- # [22:18] <edmorley> gps: nope :-) catlee-mtg was just referring to Ms2ger's comment above
- # [22:18] <gps> and the reason there were so many merge commits is because it had been a long time since s-c was merged to m-c
- # [22:18] <gps> and also increased merge frequency due to rapid development of mobile
- # [22:20] <rnewman> I do frequent merges; whenever there's a change in a branch, every subsequent merge-in is going to create a merge commit
- # [22:20] <rnewman> c'est la vie
- # [22:21] * stefanh|away is now known as stefanh
- # [22:21] <rnewman> alternative is to have rebasing :)
- # [22:21] <rnewman> but we usually have much higher volume.
- # [22:21] * philor likes nice up to date branches way better than ones full of old stale failures
- # [22:22] <rnewman> +1
- # [22:22] <gps> http://memegenerator.net/instance/12425198
- # [22:22] <mbrubeck> yes, m-i has exciting *new* failures!
- # [22:23] * Joins: rfkelly (rfkelly@moz-B619B37F.lns20.mel6.internode.on.net)
- # [22:23] * mbrubeck files this test_painting bug
- # [22:23] * philor waits for his browser to unhang from having opened a full log unexpectedly
- # [22:24] <philor> oh, crap, that was an M2, wasn't it? and M2 is where writer_starvation lives, isn't it?
- # [22:24] * zpao|detached is now known as zpao
- # [22:24] <philor> so I just opened a 50MB log, I'll bet
- # [22:25] <philor> getting to be a little less than my usual calm sweet self about that
- # [22:26] <khuey> philor: r=me to disable that test
- # [22:26] <gps> yikes - we just got 94 new compiler warnings on Windows: http://jenkins.gregoryszorc.com:9000/job/mozilla-central/223/
- # [22:26] <philor> khuey: thx, I'm only five hours from a tree, if my browser unhangs by then so I can go home
- # [22:27] <gps> "inconsistent dll linkage" huh?
- # [22:28] <mbrubeck> yup, it's 595368 (test_writer_starvation)
- # [22:28] * Quits: caillon (caillon@moz-186D75E2.hsd1.ca.comcast.net) (Ping timeout)
- # [22:29] <blizzard> jduell: I'm pretty excited that we might get to unprefix websockets
- # [22:29] * KaiRo is now known as KaiRo_away
- # [22:29] * Quits: kaie (kaie@moz-F1678BAE.dip.t-dialin.net) (Quit: Leaving)
- # [22:30] <jduell> blizzard: it would be swell
- # [22:31] <blizzard> squee!
- # [22:31] <jduell> blizzard: I want to make sure we don't make a rash engineering decision. I hate landing stuff on the last day--though it looks like safe patches, those are famous last words
- # [22:31] <smaug> jduell: do you need still reviews from me for websocket patches?
- # [22:32] <gps> ahh, it looks like the Graphite stuff created a bunch of new warnings
- # [22:32] <jduell> smaug: no, I don't think so. There's a patch I was tempted to knock off (close event ordering) but I think it'll have to wait.
- # [22:32] <blizzard> jduell: :)
- # [22:32] <blizzard> jduell: if they work they work
- # [22:32] <jhford> philor: fwiw, we are now logging the program that changes resolution so we can see whether we are running it twice, or incorrectly...
- # [22:33] <philor> ah, cool
- # [22:33] * Joins: sicking (chatzilla@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [22:33] <jduell> blizzard: and really the changes are actually mostly trivial ones. So....
- # [22:33] <jlebar> So for those people who develop in git: How do you do the equivalent of back up your patch queue?
- # [22:33] * jduell crosses fingers
- # [22:33] <jlebar> You can just push to a remote fork of m-c, but then you can't rebase your branches and whatnot.
- # [22:33] <jhammel> jlebar: use hg? :P
- # [22:33] * Joins: bholley (bholley@moz-BBE3ABD.mv.mozilla.com)
- # [22:34] <bholley> bsmedberg: ping
- # [22:34] <jlebar> ohai, bholley.
- # [22:34] <jlebar> <jlebar> So for those people who develop in git: How do you do the equivalent of back up your patch queue?
- # [22:34] <jlebar> <jlebar> You can just push to a remote fork of m-c, but then you can't rebase your branches and whatnot.
- # [22:34] <bholley> jlebar: push the branches to github
- # [22:34] <Ms2ger> Look, it's bholley
- # [22:34] <bholley> jlebar: you can absolutely rebase your branches
- # [22:34] <bholley> jlebar: just push -f
- # [22:34] <mwu> only a prob if you care about people pulling your branches
- # [22:35] <bholley> I make no such guarantees
- # [22:35] <bholley> I push to github for backups and so that people can do code review
- # [22:35] <jlebar> bholley, well, I care about pulling my own branches.
- # [22:35] <bholley> jlebar: pulling your own branches?
- # [22:35] <jlebar> bholley, for example, I pull my branches down onto other machines.
- # [22:35] <Ms2ger> Use hg
- # [22:36] <bholley> jlebar: shouldn't be an issue
- # [22:36] <jlebar> Ms2ger, Unfortunately, b2g uses git.
- # [22:36] <bholley> jlebar: you can still pull, and it will complain
- # [22:36] <bholley> jlebar: but you'll pull to a remote tracking branch
- # [22:36] <jlebar> Ms2ger, And as much of a pain as git is, a combined hg + git workflow is worse.
- # [22:36] * Joins: azakai|2_ (alon@moz-8D0CC798.lightspeed.sntcca.sbcglobal.net)
- # [22:36] <jlebar> bholley, Ah, I see. And then I just reset --hard.
- # [22:36] <bsmedberg> bholley: pong
- # [22:36] <bholley> jlebar: and then you can just rebase your local changes onto the new remote tracking branch
- # [22:37] <bholley> bsmedberg: do you know what the deal is with js/jetpack?
- # [22:37] * Quits: azakai|2 (alon@moz-90E3042.lightspeed.sntcca.sbcglobal.net) (Ping timeout)
- # [22:37] <bsmedberg> well, it's not built AFAIK
- # [22:37] <bholley> bsmedberg: can we kill it?
- # [22:37] <bsmedberg> it was buggy, and the jetpack guys don't seem to want it any more
- # [22:37] <bsmedberg> but you should ask them, I guess
- # [22:38] * Quits: pcwalton (pcwalton@moz-7B0110AD.mv.mozilla.com) (Quit: pcwalton)
- # [22:38] <bholley> bsmedberg: if they don't want it, we can get rid of it, right?
- # [22:39] <Callek> gavin: ping
- # [22:39] <gavin> pong
- # [22:39] <Callek> gavin: I think Bug 712115 might be your fault
- # [22:39] <gavin> yes, I saw the email
- # [22:39] * Quits: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [22:39] <Callek> ooo didn't realize you got an e-mail for that filing
- # [22:39] <gavin> I didn't
- # [22:40] <gps> jlebar: 1) rebasing on branches pulled by others (including yourself) is strictly frowned upon. 2) you can recover from it by using $ git rebase —onto
- # [22:40] * Quits: bwinton (bwinton@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
- # [22:40] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
- # [22:41] <jlebar> gps, So as soon as you want to back up your branches, git says "you're on your own?"
- # [22:41] <jlebar> gps, Forgive me, but this is lame.
- # [22:41] <gps> jlebar: well, just push your branches to somewhere else!
- # [22:41] * Joins: bugpage (blah@71CD2039.B4CD08EA.79496794.IP)
- # [22:41] <mbrubeck> jlebar: You could push your local branch to a differently-named remote branch
- # [22:42] <bholley> bsmedberg: jetpack guys don't want it for the moment, and can pull it out of the history if they need it again. sr=you on removal?
- # [22:42] <mbrubeck> "git push remotename mybranchname:someotherbranchname"
- # [22:42] <bugpage> anyone know why http://www.codedread.com/ looks bad in Firefox 11.0 Nightly but it looks ok on Firefox 9.0 Final? not sure whether it's a site/page bug or a Firefox 11 bug
- # [22:42] * Quits: davidb (davidb@F2D29657.F60B0462.67AC9B1.IP) (Quit: davidb)
- # [22:42] <gps> if you only care about patches, you can run $ git format-patch to export a bunch of patch files
- # [22:42] * mwu was assuming jlebar was talking about differently named branches
- # [22:43] <mbrubeck> b2g uses github, right?
- # [22:43] <jlebar> mbrubeck, I don't understand how that helps.
- # [22:43] <gps> but, then you have to deal with patches. just push your branches to somewhere more reliable, like GitHub
- # [22:43] <jlebar> mbrubeck, yes.
- # [22:43] <catlee-mtg> backing up .git/ doesn't do what you want?
- # [22:43] <mbrubeck> Why not just create your own fork to push/pull your WIP stuff to?
- # [22:43] * Joins: roc (chatzilla@538BABFE.A073F3E.97BBD552.IP)
- # [22:43] * Quits: ahal (ahal@F2D29657.F60B0462.67AC9B1.IP) (Client exited)
- # [22:43] <mbrubeck> (also, I don't understand Mozilla people's obsession with rebasing instead of merging)
- # [22:44] <jlebar> catlee-mtg, well, I'd like to back up by pushing to github, for example, rather than backing up .git via rsync.
- # [22:44] <bholley> mbrubeck: keeps the history clean
- # [22:44] <mbrubeck> I do use "git push -f" to sometimes overwrite github branches in my personal forks, when I've rewritten history that hasn't made it into a shared repo
- # [22:44] <jlebar> mbrubeck, Creating my own fork and push/pulling my WIP stuff is the idea. But the problem is, I'd like to share this fork between my Windows, Linux, and Mac development machcines.
- # [22:44] <mwu> mbrubeck: I think it's complicated by having to apply patches and work with bugzilla
- # [22:44] <bholley> mbrubeck: which you need with so many people
- # [22:45] <gps> rebasing is nice to clean up commit history, like when you need to get past the commit hooks
- # [22:45] <catlee-mtg> you can share rebased repos
- # [22:45] <mbrubeck> mwu: True, patch queues or rebasing help with the Bugzilla review workflow.
- # [22:45] <catlee-mtg> it's just not as easy as if you merge stuff
- # [22:45] * Joins: ahal (ahal@F2D29657.F60B0462.67AC9B1.IP)
- # [22:45] <mbrubeck> bholley: But we have merges all over the place -- look at the latest push to mozilla-central -- and it doesn't seem to cause any concrete problems.
- # [22:46] <bholley> mbrubeck: yes, but those are merges on an organization scale. If everybody included 10 merges in each patch (representing the 10 tries it took to get right), the history would be unmanageable
- # [22:46] <bholley> mbrubeck: the idea is that patches should be atomic
- # [22:46] <mbrubeck> (the latest push to m-c contained 24 merges and only 22 other commits)
- # [22:46] * jlebar thinks the m-c history is currently unmanageable. :)
- # [22:46] <bholley> and merges should be reserved for releng and sheriff tasks
- # [22:46] * Quits: mib_1x3q33 (Mibbit@D428F312.8C1A2EFD.5642DB81.IP) (Quit: http://www.mibbit.com ajax IRC Client)
- # [22:47] * Joins: darktrojan (geoff@moz-BC95E278.dsl.telstraclear.net)
- # [22:47] <mbrubeck> bholley: Yeah, I agree rewriting history is good to make patches into atomic, standalone changesets.
- # [22:47] <Pike> bear-buildduty: how do we create the mozconfig for mac nightlies these days?
- # [22:47] <bear-buildduty> they are in buildbot-config now
- # [22:48] <jlebar> I don't understand how merges make patches worse, though. Can't you still diff against master to generate a patch?
- # [22:48] <jlebar> Or does the problem occur only when you have a multi-patch queue?
- # [22:48] * Quits: msucan (msucan-@FA9E8863.56E67207.699550A1.IP) (Quit: .)
- # [22:48] <Pike> bear-buildduty: we're not using the one in the trees yet? say https://hg.mozilla.org/releases/mozilla-aurora/diff/bd759f82e517/browser/config/mozconfigs/macosx-universal/nightly ?
- # [22:48] <mbrubeck> jlebar: For backing up you "patch queue", does creating your own fork on github and "git push -f" meet your needs? That's what I do for my Rust patches.
- # [22:48] <jlebar> Which git sucks at, afaict, because it's (intentionally) hard to go back a commit, fix it, and then move all the forward commits.
- # [22:49] * lurking_work thinks something in broken in Linux builds with the latest merge from sc - 211 test fails - empty summary
- # [22:49] <mbrubeck> jlebar: I think bholley's case is mainly about using rebase not just to pull in upstream changes but also to rewrite your own patches as you change them.
- # [22:49] <jlebar> mbrubeck, Maybe. That's really what my question was -- does that work when I share that github repo between three development machines and want to be able to rebase my branches?
- # [22:49] * Parts: rs (rs@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [22:49] * Joins: rs (rs@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [22:50] <mbrubeck> jlebar: Yes, it should
- # [22:50] <jlebar> mbrubeck, Sure, making the git history match the N patches you're going to post.
- # [22:50] <gps> jlebar: if you are actively working across that many machines, I recommend not rebasing and using push -f
- # [22:50] * Joins: eflores (AndChat@538BABFE.A073F3E.97BBD552.IP)
- # [22:50] <jlebar> See, these are the conflicting messages I get.
- # [22:50] <bhearsum> Pike: i'm pretty sure we're using the in-tree ones
- # [22:50] <gps> instead, perform a bunch of merges, then copy the branch and do a bunch of rebasing to produce your final patch
- # [22:50] <jlebar> gps, And if instead of one patch, I want a series of N patches?
- # [22:50] * philor thinks "browser_details.js | $0.99" is a pretty nice error message
- # [22:50] <khuey> vlad: I managed to get it working by copying the debug crt out of the noredist folder (instead of using the one the compiler installed on the machine) and putting it in the firefox directory
- # [22:51] <khuey> vlad: and rejiggering the symbol path order
- # [22:51] <Pike> bhearsum: can you help find which we're using for l10n osx repacks? nightly and dep? as the file names just confuse me
- # [22:51] <gps> well, you can merge patches selectively with $ git rebase -i
- # [22:51] <bhearsum> Pike: sure - i'll try to get back to you in the next 15min
- # [22:51] <Pike> thanks
- # [22:51] <jlebar> gps, but if I have csets A --> B --> C, I can't fold C into A.
- # [22:51] * Quits: edmorley (edmorley@moz-28ED10AB.range86-144.btcentralplus.com) (Ping timeout)
- # [22:52] <jlebar> gps, whereas in hg, I can hg qgoto A, touch some files, then hg qgoto B.
- # [22:52] <gps> jlebar: have you seen $ git rebase -i ? you can reorder commits, fold anything into anything
- # [22:52] * Joins: gunther (Mibbit@D428F312.8C1A2EFD.5642DB81.IP)
- # [22:52] <jlebar> gps, orly? /me did not know I could reorder them.
- # [22:52] * Joins: robhawkes (robhawkes@moz-33A339B7.dsl.cnl.uk.net)
- # [22:53] <mbrubeck> yeah, just use your editor to change the order of the lines
- # [22:53] <gps> jlebar: rebase -i is friggin awesome. see http://book.git-scm.com/4_interactive_rebasing.html for an overview
- # [22:53] * Quits: sfink (chatzilla@moz-4EAF93BB.dsl.pltn13.sbcglobal.net) (Client exited)
- # [22:53] * Joins: edmorley (edmorley@moz-28ED10AB.range86-144.btcentralplus.com)
- # [22:54] <bhearsum> Pike: ah, for mozilla-aurora nightly l10n we don't use one
- # [22:54] * Quits: gunther (Mibbit@D428F312.8C1A2EFD.5642DB81.IP) (Quit: http://www.mibbit.com ajax IRC Client)
- # [22:54] <gps> https://bitbucket.org/durin42/histedit is a HG extension that effectively adds 'rebase -i' to Mercurial
- # [22:54] <bhearsum> nightly desktop l10n is the last place we don't have mozconfigs AFAIK
- # [22:54] * Joins: dRdR (dRdR@moz-F979C50C.cpe.net.cable.rogers.com)
- # [22:54] <philor> gps: can you back out merges? that would be a really handy thing right now
- # [22:54] * Quits: rfkelly (rfkelly@moz-B619B37F.lns20.mel6.internode.on.net) (Quit: Leaving)
- # [22:54] <bhearsum> Pike: http://hg.mozilla.org/build/buildbotcustom/file/tip/process/factory.py#l3059 is what we use instead
- # [22:55] <gps> define "back out"
- # [22:55] <gps> do you care about history?
- # [22:55] <gps> if you don't care about history, $ git checkout -b new_branch_name REVISION_BEFORE_MERGE
- # [22:55] <philor> gps: I care about the mozilla-central tree not failing 211 tests
- # [22:55] * Joins: felipe2 (AndChat@moz-BBE3ABD.mv.mozilla.com)
- # [22:55] <bhearsum> Pike: does that answer your question?
- # [22:55] * Quits: felipe|tab (AndChat@moz-BBE3ABD.mv.mozilla.com) (Connection reset by peer)
- # [22:55] <gavin> you can back out merge with |hg revert| :)
- # [22:55] * Parts: benadida (ben@moz-BBE3ABD.mv.mozilla.com)
- # [22:55] * Joins: lsumar (lsumar@538BABFE.A073F3E.97BBD552.IP)
- # [22:55] <gps> if you do, what gavin says
- # [22:56] <mbrubeck> umm, wow
- # [22:56] <philor> gps: I'm hinting around about the fact that your merge broke the tree
- # [22:56] * Joins: caillon (caillon@moz-359E39FC.lightspeed.sntcca.sbcglobal.net)
- # [22:56] <philor> turns out that merging from a totally orange tree isn't aways a good idea ;)
- # [22:56] <gps> philor: huh?
- # [22:56] <khuey> oh beautiful beautiful crt symbols
- # [22:56] <philor> https://tbpl.mozilla.org/
- # [22:56] <Ms2ger> What's "orange"?
- # [22:56] <mbrubeck> https://tbpl.mozilla.org/?tree=Services-Central&rev=259262c87f23
- # [22:57] <Ms2ger> Is that something you need to know to land?
- # [22:57] <Ms2ger> It can't be, right?
- # [22:57] <Pike> bhearsum: that sounds like severe ouch, as that might be the issue keeping us from having mac nightlies
- # [22:57] <mbrubeck> I'm thinking of something orange: https://www.youtube.com/watch?v=TN-pXWHqAAM
- # [22:57] <bhearsum> Pike: if it is, we'll find a fix
- # [22:58] * Quits: bugpage (blah@71CD2039.B4CD08EA.79496794.IP) (Ping timeout)
- # [22:59] * juab|brb is now known as juanb
- # [22:59] <darktrojan> I like the way planet choked on your blog post, khuey
- # [22:59] * catlee-mtg is now known as catlee
- # [23:00] <gps> philor: I think it would be sufficient to revert https://hg.mozilla.org/mozilla-central/rev/259262c87f23
- # [23:00] <mbrubeck> gps, philor: So, maybe we should just back out 259262c87f23?
- # [23:00] * Quits: Ami_Ty_ (Amie@moz-BBE3ABD.mv.mozilla.com) (Connection reset by peer)
- # [23:00] <mbrubeck> jinx
- # [23:00] * Joins: pcwalton (pcwalton@moz-7B0110AD.mv.mozilla.com)
- # [23:00] * Joins: Ami_Ty_ (Amie@moz-BBE3ABD.mv.mozilla.com)
- # [23:00] <rnewman> mbrubeck: seems sane to me
- # [23:01] <darktrojan> actually, khuey, your rss feed has choked on your blog post
- # [23:01] * Quits: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [23:01] <reuben> if I click on this link and quickly press cmd+L, cmd+A, cmd+C to copy the URL while the page is still loading the 'http://' doesn't get prepended: http://www.amazon.com/gp/feature.html?ie=UTF8&docId=1000746611
- # [23:01] <khuey> darktrojan: yeah tumblr is dumb
- # [23:01] <reuben> (with browser.urlbar.trimURLs = true)
- # [23:01] <darktrojan> I see that
- # [23:02] <gavin> reuben: yes, known issue
- # [23:02] <Pike> bhearsum: I'm pretty sure, I moved the bug over for now. I can't get to the logs from today's en-US nightly through tbpl, so it's hard to confirm what it's actually using for build
- # [23:02] * Joins: anant_ (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [23:02] * zpao is now known as zpao|detached
- # [23:02] * Joins: Mossop_ (mossop@moz-BBE3ABD.mv.mozilla.com)
- # [23:02] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
- # [23:02] * Quits: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [23:02] * reuben is terrible at finding bugs :(
- # [23:02] <reuben> gavin, can you give me a bug # ?
- # [23:03] * Quits: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
- # [23:03] * Quits: Mossop (mossop@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [23:03] <jlebar> reuben, Have you tried Googling "site:bugzilla.mozilla.org what I want to search for"?
- # [23:03] * Quits: gkw (gkw@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [23:03] * Mossop_ is now known as Mossop
- # [23:03] * Joins: sworkman-tablet (sworkman@moz-BBE3ABD.mv.mozilla.com)
- # [23:04] * Quits: anant_ (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
- # [23:04] * Quits: Ms2ger (Ms2ger@CC7F0FF.207206FA.37724B0D.IP) (Quit: nn)
- # [23:04] * Quits: smooney (smooney@moz-BBE3ABD.mv.mozilla.com) (Quit: smooney)
- # [23:04] * rail is now known as rail-champagne
- # [23:04] <firebot> Check-in: http://hg.mozilla.org/mozilla-central/rev/f98c57415d8d - Matt Brubeck - Back out 259262c87f23 (bug 711263) because of test failures
- # [23:04] <reuben> jlebar, nope, I always use bugzilla's search
- # [23:05] * Quits: sworkman-tablet (sworkman@moz-BBE3ABD.mv.mozilla.com) (Quit: Bye)
- # [23:05] * philor is now known as philor|away
- # [23:05] <jlebar> reuben, that may be the problem. :)
- # [23:05] <gavin> reuben: I guess there isn't a bug covering that issue specifically, but it's a variant of bug 670531
- # [23:05] <gavin> any URL that isn't "loaded" doesn't get copied directly
- # [23:05] <reuben> hm, they fixed the specific case of copying it from the suggestion list
- # [23:06] * coop|mtg is now known as coop
- # [23:06] <reuben> which isn't loaded
- # [23:06] <gavin> did we?
- # [23:06] <gavin> I forget
- # [23:06] <reuben> bug 668019
- # [23:07] <reuben> maybe the fix for the bug you mentioned will eliminate the need for that patch
- # [23:07] <gavin> yeah, I reviewed that
- # [23:07] * Joins: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [23:08] * Quits: peregrino (peregrino@moz-96A63A2D.telecom.net.ar) (Quit: peregrino)
- # [23:08] <mbrubeck> gps, rnewman: Looks like this does leave us with a new xpcshell failure on m-c, though.
- # [23:08] * philor|away is now known as philor
- # [23:08] <mbrubeck> It'd be nice if that could be fixed soon. (Or alternately we could revert the whole merge, but that would get messy.)
- # [23:08] * Joins: sworkman-phone (sworkman@moz-BBE3ABD.mv.mozilla.com)
- # [23:09] * Joins: stfl|away (stfl@moz-22F37930.hsi7.kabel-badenwuerttemberg.de)
- # [23:09] <rnewman> mbrubeck: yes, I'm sure gps will find a fix that doesn't cause bustage :/
- # [23:11] * Quits: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com) (Quit: nhirata)
- # [23:11] * Quits: bholley (bholley@moz-BBE3ABD.mv.mozilla.com) (Quit: bholley)
- # [23:11] * Quits: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com) (Quit: mccr8)
- # [23:11] * adam is now known as adam-afk
- # [23:11] * Joins: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com)
- # [23:12] * Quits: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP) (Quit: Konversation terminated!)
- # [23:12] * Joins: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP)
- # [23:12] * Quits: Mossop (mossop@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [23:13] * Quits: bbondy (bbondy@moz-28CF6D1C.home.cgocable.net) (Ping timeout)
- # [23:13] * Quits: myk (myk@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Ping timeout)
- # [23:14] * Quits: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP) (Quit: Konversation terminated!)
- # [23:14] * Joins: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP)
- # [23:18] * Quits: zandr (zandr@moz-891BD824.milewski.org) (Max SendQ exceeded)
- # [23:19] * rail-champagne is now known as rail-alcohol
- # [23:19] * Joins: myk (myk@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [23:19] * bc|afk is now known as bc
- # [23:19] * rail-alcohol is now known as rail
- # [23:20] * coop is now known as coop|away
- # [23:20] * Joins: zandr (zandr@moz-891BD824.milewski.org)
- # [23:20] * bear-buildduty is now known as bear
- # [23:21] * Quits: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net) (Input/output error)
- # [23:21] * Joins: anant (anant@moz-C03D0C61.vlan426.asr1.sfo1.gblx.net)
- # [23:21] * Joins: sworkman (sworkman@moz-8454A02.mycingular.net)
- # [23:21] <philor> charming, apparently my internet connection disappeared last night minutes after I pushed to both m-a and m-c because some idiots stole a 600' chunk of fiber optic cable
- # [23:21] * Quits: lmandel (lmandel@F2D29657.F60B0462.67AC9B1.IP) (Quit: lmandel)
- # [23:21] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
- # [23:21] * Quits: sworkman-phone (sworkman@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [23:21] <philor> and here I thought it was just the punishment of a vengeful god
- # [23:21] * sworkman is now known as sworkman-phone
- # [23:21] * Joins: sworkman (sworkman@moz-BBE3ABD.mv.mozilla.com)
- # [23:22] * Quits: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP) (Quit: Konversation terminated!)
- # [23:22] * Joins: sewardj (sewardj@moz-C79D927B.cable.virginmedia.com)
- # [23:22] * Joins: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP)
- # [23:22] <mbrubeck> it could be both, if it was that thing from "Super 8" using fiber-optics to build its spaceship
- # [23:22] * Quits: sworkman-phone (sworkman@moz-8454A02.mycingular.net) (Quit: Bye-bye!)
- # [23:22] <froydnj> could just be the wrath of angry mozilla developers
- # [23:22] * jmaher is now known as jmaher|afk
- # [23:22] * adam-afk is now known as adam
- # [23:23] * bear is now known as bear-afk
- # [23:23] <mounir> karl: ping
- # [23:24] * Quits: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [23:24] * Quits: felipe2 (AndChat@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [23:24] * Joins: kaze (kaze@moz-79F78BFC.w90-46.abo.wanadoo.fr)
- # [23:25] * Joins: davehunt (davehunt@moz-E2929564.dsl.in-addr.zen.co.uk)
- # [23:25] <karl> mounir: hi
- # [23:25] * Joins: felipe|tab (AndChat@moz-BBE3ABD.mv.mozilla.com)
- # [23:25] * Quits: Sander (chatzilla@moz-B871F4D3.direct-adsl.nl) (Quit: And back he spurred like a madman, shrieking a curse to the sky.)
- # [23:25] <mounir> philor: fiber optic cable can be sold for a lot of money?!
- # [23:25] <mounir> karl: hi :)
- # [23:25] * Quits: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com) (Quit: mccr8)
- # [23:25] <mounir> karl: by any chance, have you heard of marshallers in DBus?
- # [23:26] <philor> dunno, last I heard it was worthless, and only stolen by people looking for copper and too dumb to know the difference
- # [23:26] * Joins: bholley (bholley@moz-7B0110AD.mv.mozilla.com)
- # [23:26] * Joins: Mossop (mossop@moz-BBE3ABD.mv.mozilla.com)
- # [23:26] <mounir> philor: yeah, stilling copper is common
- # [23:26] <karl> mounir: no, not in dbus; glib uses marshallers quite a bit
- # [23:27] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
- # [23:27] * Joins: bc_ (bc@280C34BB.72D62797.F0135836.IP)
- # [23:27] <mounir> karl: oh, might be that actually
- # [23:27] * Quits: sheppy (sheppy@moz-F39D62DA.dhcp.kgpt.tn.charter.com) (Quit: sheppy)
- # [23:27] <mounir> I mean, the same thing, given that I'm using dbus-glib
- # [23:27] <karl> ok
- # [23:27] <mounir> karl: AFAIUI, I have to generate them
- # [23:27] <mounir> do we currently have that in the tree?
- # [23:27] * Quits: sewardj (sewardj@moz-C79D927B.cable.virginmedia.com) (Ping timeout)
- # [23:28] * Quits: joey (chatzilla@moz-EFCB4CBF.princetowncable.com) (Broken pipe)
- # [23:28] <karl> mounir: often there already exists a marshaller of the correct type that you can use
- # [23:28] <mounir> karl: where can I check that?
- # [23:29] <mounir> I had a hard time to find any information about that actually :(
- # [23:29] * Quits: stevee (Miranda@moz-BEBDF855.cable.virginmedia.com) (Quit: Miranda IM - Multi protocol instant messenger @ www.miranda-im.org)
- # [23:29] * Quits: bc_ (bc@280C34BB.72D62797.F0135836.IP) (Quit: leaving)
- # [23:30] <karl> mounir: what told you that you needed one/them?
- # [23:30] * Joins: nattokirai (nattokirai@moz-C97EE7C8.dynamic.ppp.asahi-net.or.jp)
- # [23:30] * Quits: stefanh (stefanh@moz-3EED0162.customers.ownit.se) (Quit: ChatZilla 0.9.87 [SeaMonkey 2.5/20111121020919])
- # [23:30] * Joins: nhirata (nhirata.bu@moz-BBE3ABD.mv.mozilla.com)
- # [23:30] <mounir> karl: I'm trying to add/connect to a signal and when doing so, I get a warning about no marshaller and I don't caught the signal
- # [23:31] <mounir> calling dbus_g_object_register_marshaller with the appropriate arguments do the trick
- # [23:31] * Quits: bnicholson (bnicholson@moz-F7C50328.rcmdva.fios.verizon.net) (Quit: Leaving)
- # [23:31] <mounir> (no warning and things work)
- # [23:31] <mounir> it seems that dbus-glib needs a marshaller when trying to connect to a signal with more than one parameter
- # [23:31] <karl> a marshaller is used by a signal generator to call a function (or even some other language) in an appropriate manner
- # [23:32] <karl> i.e. with appropriate arguments
- # [23:32] * mbrubeck is mid-air-colliding with philor on bug 689856 and it wasn't even the same log
- # [23:32] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Connection reset by peer)
- # [23:33] <philor> yeah, I was second in line earlier today
- # [23:33] * Joins: merinui (merinui@moz-61C7235E.osk2.eonet.ne.jp)
- # [23:33] * Joins: whimboo (whimboo@moz-216F9AFA.superkabel.de)
- # [23:33] * Quits: AutomatedTester (AutomatedT@moz-352F716D.ppp.as43234.net) (Ping timeout)
- # [23:34] * Quits: merinui (merinui@moz-61C7235E.osk2.eonet.ne.jp) (Quit: Leaving...)
- # [23:34] * Joins: AutomatedTester (AutomatedT@moz-8DECC833.as13285.net)
- # [23:34] <mounir> karl: what seems really odd is that the marshaller has to be registered in a separate method and it doesn't have to be passed to the connect method
- # [23:34] * Parts: whimboo (whimboo@moz-216F9AFA.superkabel.de)
- # [23:34] <mounir> that's a quite unintuitive API
- # [23:34] * Joins: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP)
- # [23:35] <karl> mounir: yes, sounds odd to me too
- # [23:35] <mounir> karl: anyway, I would be interesting to know how I could check if there is already a marshaller of the type I'm want
- # [23:35] <karl> looking
- # [23:36] * Quits: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP) (Quit: Konversation terminated!)
- # [23:36] * Joins: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP)
- # [23:36] <mounir> otherwise, would you recommend using glib-genmarshall in Makefile or just adding the generated files in m-c? (I would tend to do the former even if a bit more onnoying)
- # [23:36] * Quits: davehunt (davehunt@moz-E2929564.dsl.in-addr.zen.co.uk) (Input/output error)
- # [23:37] <karl> mounir: see if something here suits: http://developer.gnome.org/gobject/stable/gobject-Closures.html
- # [23:37] * Quits: clee (clee@moz-BBE3ABD.mv.mozilla.com) (Quit: clee)
- # [23:38] * Joins: clee (clee@moz-BBE3ABD.mv.mozilla.com)
- # [23:38] * rail is now known as rail_away
- # [23:38] <mounir> karl: when seeing those, I understand why one is needed: they all take 1 param while the signal I'm conncting to takes 3 params
- # [23:39] * Joins: pnemsak_ (Miranda@moz-7409BD53.orange.sk)
- # [23:40] <gps> mbrubeck: I'm working on a fix for the busted merge now. might take a while, since I'll probably need to invoke a try build
- # [23:40] <karl> mounir: i'm not familiar with generating marshallers; if the generate code is small, i'd just copy that, if it's big then i see the advantage of using genmarshall
- # [23:40] <mbrubeck> gps: cool
- # [23:40] * Quits: pnemsak (Miranda@moz-7409BD53.orange.sk) (Ping timeout)
- # [23:40] <gps> you'll see a few oranges for xpcshell failures on Windows, and possibly other arches
- # [23:40] <mounir> karl: I only need one marshaller so it's not really big for the moment
- # [23:40] * Quits: rwaldron (rwaldron@moz-BDCCF091.hfc.comcastbusiness.net) (NickServ (GHOST command used by rick))
- # [23:40] * Joins: rick (rwaldron@moz-CE27855E.hfc.comcastbusiness.net)
- # [23:41] * Quits: evilpie (chatzilla@moz-671907A1.pools.arcor-ip.net) (Quit: ChatZilla 0.9.87 [Firefox 11.0a1/20111218031140])
- # [23:42] * Joins: gkw (gkw@moz-BBE3ABD.mv.mozilla.com)
- # [23:42] * Quits: tanvi1 (tanvi@moz-BBE3ABD.mv.mozilla.com) (Quit: Leaving.)
- # [23:42] * Joins: tanvi (tanvi@moz-BBE3ABD.mv.mozilla.com)
- # [23:43] * Quits: stfl|away (stfl@moz-22F37930.hsi7.kabel-badenwuerttemberg.de) (Ping timeout)
- # [23:43] * Quits: jfkthame (jfkthame@CE8EB10E.90783722.9542EC20.IP) (Quit: jfkthame)
- # [23:43] * Joins: bmoss (bmoss@moz-BBE3ABD.mv.mozilla.com)
- # [23:44] <karl> mounir: also check gtkmarshal.h
- # [23:44] <smaug> jst: ping (I know, you're probably on vacation)
- # [23:45] * Quits: jhk (jiggy@8E6C34C1.A3F9767A.1C37C358.IP) (Ping timeout)
- # [23:45] * Joins: mccr8 (mccr8@moz-BBE3ABD.mv.mozilla.com)
- # [23:45] <mounir> smaug: fwiw, he is in Europe
- # [23:45] <smaug> yes, he is in Vaasa, I think
- # [23:45] <smaug> in Finland
- # [23:46] <mounir> smaug: oh, indeed you two are from the same country :)
- # [23:46] * Joins: jgilbert_ (jgilbert@moz-BBE3ABD.mv.mozilla.com)
- # [23:47] * Quits: jgilbert (jgilbert@moz-BBE3ABD.mv.mozilla.com) (Ping timeout)
- # [23:47] <Pike> ted: any idea why `export realchrome:: ../res/values/strings.xml` wouldn't work in embedding/android/locales, but having them in two separate rules totally does? Also, mobile/android/base/locales didn't seem to have that issue
- # [23:47] <mounir> karl: oh, that was close but unfortunately not :(
- # [23:47] * Quits: graememcc (chatzilla@moz-4BCD914C.range86-148.btcentralplus.com) (Quit: ChatZilla 0.9.87 [Firefox 8.0/20111115204015])
- # [23:48] * Cwiiis is now known as CwiiisAway
- # [23:48] * Joins: dholbert_ (dholbert@moz-87D6E13A.tmodns.net)
- # [23:48] * Joins: Fallen|mac (kewisch@moz-6F64C71A.adsl.alicedsl.de)
- # [23:49] * Joins: dholbert__ (dholbert@moz-294A858C.static.cruzio.com)
- # [23:49] * Quits: dholbert__ (dholbert@moz-294A858C.static.cruzio.com) (Connection reset by peer)
- # [23:50] * Quits: dholbert_ (dholbert@moz-87D6E13A.tmodns.net) (Ping timeout)
- # [23:51] * Quits: jimb (user@moz-F4EC06CC.hsd1.or.comcast.net) (Input/output error)
- # [23:51] <mounir> karl: I think we could create a mozilla_marshal list of marshalers like gtk_
- # [23:51] * heycam|away is now known as heycam
- # [23:51] <mounir> karl: would you recommend a place to put those?
- # [23:52] * Quits: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP) (Quit: Konversation terminated!)
- # [23:52] * Joins: bjacob (bjacob@F2D29657.F60B0462.67AC9B1.IP)
- # [23:53] * jhopkins is now known as jhopkins|afk
- # [23:53] <karl> mounir: i don't have any good suggestions really. widget/src/gtk would be the first place i'd look
- # [23:54] * Quits: florian (florian@moz-87C33FDA.kimsufi.com) (Quit: Instantbird 1.2a1pre)
- # [23:54] * Joins: smooney (smooney@moz-BBE3ABD.mv.mozilla.com)
- # [23:54] * Quits: victorporof (victorporo@5D14C11A.70F4578F.79933D60.IP) (Ping timeout)
- # [23:55] * Joins: alastair (alastair@moz-9E58F371.dynamic.dsl.as9105.com)
- # [23:55] * Joins: stfl|away (stfl@moz-22F37930.hsi7.kabel-badenwuerttemberg.de)
- # [23:55] * Joins: dao (dao@moz-2DD4FBFB.superkabel.de)
- # [23:55] * Joins: bnicholson (bnicholson@moz-CA78B821.rcmdva.fios.verizon.net)
- # [23:56] * armenzg is now known as armenzg_away
- # [23:57] * Quits: protz (jonathan@moz-7F6750F6.xulforum.org) (Quit: Leaving)
- # [23:57] * Quits: stfl|away (stfl@moz-22F37930.hsi7.kabel-badenwuerttemberg.de) (Connection reset by peer)
- # [23:58] * Joins: njn (chatzilla@32788AE8.60066DFE.8660782D.IP)
- # [23:58] * Quits: mconley (mconley@F2D29657.F60B0462.67AC9B1.IP) (Input/output error)
- # [23:59] * Quits: lsumar (lsumar@538BABFE.A073F3E.97BBD552.IP) (Quit: Leaving)
- # [23:59] * Quits: gabor (gabor@moz-3B57BCD1.catv.pool.telekom.hu) (Ping timeout)
- # Session Close: Tue Dec 20 00:00:00 2011
The end :)