mozilla :: #servo

19 Mar 2017
01:49stshineSimonSapin: hooray for AtomicRefCell, which is exactly what I was thinking about when I saw the RwLock thing \o/
03:10nomayoQuick question. I see in one of the issues that gecko glue code was mentioned and I'd like to know if that's what the helpers generate that are mentioned in the stylo hacking guide?
10:48emiliocrowbot: tell heycam lol @ the hashtable patches ending up being needed :)
11:18heycamemilio: and then thought I may as well look at the remainder of the patches in the bug there...
11:18heycamcrowbot: didn't you have something to tell me?
11:18crowbotPlease specify a nick and a message
11:19emilioheycam: yeah, that was the only one that was hard to review tbh :)
11:19emilioheycam: thanks for the review!
11:19heycamnp :)
11:20heycamdown to 4 things in my review queue now (all ignorable)
11:20emilioheycam: well, that's because you don't check your Servo review queue (#15890) :P
11:20crowbotPR #15890: selectors: Check the bloom filter at most once per complex selector. -
11:21heycamof my 47 needinfos, two are from people who have left mozilla :o
11:22heycamemilio: I don't seem to have received mail from that PR :(
11:22* heycam isn't really sure what's going on with GH notifications for him, but would quite like something similar to bugzilla's request queue page for it
11:23emilioheycam: yeah, I've also seen github sometimes losing some notifications for w/e reason :(
11:23heycamemilio: I'll email myself to look at it tmr
11:23emilioheycam: The closest you can get is probably, with the S-awaiting-review label too probably
11:23emilioheycam: thanks!
11:24heycamemilio: oh yeah not bad
11:53emilioheycam: oh, btw, that switch happens to fix the performance problems I told you about the UndisplayedContentMap, since LinkedList::insertBack is O(1) now
11:53emilioheycam: that is, assuming we don't have to care about duplicates, which I think we don't.
12:09heycamemilio: sounds good... I don't remember the perf problems you mentioned hto
12:35SimonSapinemilio: anything else, for #16014 ? :)
12:35crowbotPR #16014: Per-process lock for CSSOM objects -
12:45noxemilio, heycam|away: Also,
13:13canaltinovaManishearth: I cc'ed you for will-change in m-c side in case xidorn doesn't have time to look
13:13xidorncanaltinova: I'll have a look tomorrow. bed time now
13:14canaltinovaxidorn: ok, good night :)
13:14canaltinovaI finished this in 4am but forgot to push yesterday :)
13:51CajetanPhelllo !
13:52CajetanPCould somebody here guide me with solving the issue #16021
13:52crowbotIssue #16021: WebGL rendering context should not unwrap every send -
13:53CajetanPI already posted a comment with my question there
15:15stshineemilio: the space looks ...
15:37noxacrichto: Are features correctly respected when replacing a crate?
15:39emiliostshine: ugh, indeed. Not on my computer r/n, but feel free to fix it up and r=me. And presumably we should file an issue so tidy doesn't allow that
15:41stshineemilio: let me get a general mind about rule tree first :)
17:34Aryxhi, the push for 15604 (zploskey) causes some mochitest failures: would apprciate if anybody can take a look
17:45KiChjangheycam|away, crowbot doesn't deliver messages until you relogin to IRC
19:08* KiChjang|mobile -> plane to HK
20:21canaltinovaemilio: oops, I wrote wrong bug id to commit
20:22emiliocanaltinova: I already updated the expectations at We should investigate that failure, since it's :-moz-broken related, before claiming to have them all implemented.
20:25canaltinovaemilio: hmm, I just look at the test
20:26emiliocanaltinova: I suspect bug 1338982, but worth investigating more I guess.
20:26firebot ASSIGNED, stylo: NODE_HAS_SLOW_SELECTOR_LATER_SIBLINGS is not set correctly
20:28canaltinovaemilio: yeah
20:28emiliocanaltinova: nope, it's not it.
21:18emilioSimonSapin: sorry for the delay in reviewing your fixups btw, today I wasn't at my computer for most of the day :(
21:27canaltinovaemilio: ooh, it looks like one of my changeset hasn't pushed to autoland in bug 1341642. And I just overwrote that changeset with failed expectation update changset...
21:27firebot FIXED, stylo: Need support for the various -moz stuff used for alt text
21:27canaltinovaemilio: I thought pulsebot shows last commit that pushed
21:28emiliocanaltinova: can you send the patch here?
21:28canaltinovaemilio: see
21:28canaltinovaemilio: resent to bug
21:29SimonSapinemilio: no worries
21:29emiliocanaltinova: can you do a try run and update the expectations as needed?
21:29canaltinovaemilio: sure
21:58math-alphaI am a CSC student and would like to participate to round 14 outreachy
21:59math-alphaI will be glad to talk to a mentor :)
22:53jntrnris there a way with cargo to update one of the dependencies without updating all of them?
22:54jntrnrI tried a 'cargo update' with servo, but it seems you can't do that willy nilly
22:56gwjntrnr: ./mach update-cargo -p <package name>
23:16AryxSimonSapin: hi, yt? issue 16014 needs a revendoring, autoland is busted
23:16crowbotPR #16014: Per-process lock for CSSOM objects -
23:16SimonSapinAryx: isnt autosync supposed to revendor automatically now?
23:17Aryxunless it has happened in the last week, no. and sheriffs have not be informed about this
23:17canaltinovaemilio: well, there is no passing test on the try :/ it&#39;s a bit late here, I&#39;ll continue to investigate this tomorrow.
23:18SimonSapinAryx: Im making a patch
23:18gwstandups: Fix rendering bug with multiple inset box shadows with offsets -
23:18crowbotPR #992: Ensure that inset box shadows don&#39;t fetch texels outside cache rect. -
23:18standupsOk, submitted #43895 for
23:19Aryxthank you
23:25SimonSapinAryx: I dont have push access
23:25Aryxok, will take it and push it to autoland. thank you
23:25SimonSapin(beyond L1)
23:26SimonSapinI knew this would happen, sorry I forgot again to watch for it at the time of landing
23:48Manishearthemilio: fwiw the GetDocumentState API is one that bz wanted after discussing this with him, too
20 Mar 2017
No messages
Last message: 155 days and 17 hours ago