Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- darkowlzz is now known as darkowlzz|afk.
- [4:30pm] bdahl joined the chat room.
- [4:30pm] rclick joined the chat room.
- [4:31pm] dbaron: jya, there seem to be a lot of test failures in the last few pushes that are timeouts on media mochitests on Android -- any chance that's related to your pushes?
- [4:31pm] bbondy joined the chat room.
- [4:32pm] dzbarsky joined the chat room.
- [4:32pm] mixedpuppy joined the chat room.
- [4:35pm] jya: dbaron: it shouldn’t.. my changes do not touch any android code path
- [4:35pm] TYLin joined the chat room.
- [4:35pm] jya: was all green in try
- [4:36pm] grizzledorfin left the chat room. (Quit: Going offline, see ya! (www.adiirc.com))
- [4:37pm] philor: https://treeherder.mozilla.org/#/jobs?repo=mozilla-inbound&filter-searchStr=Android%20armv7%20API%209%20mozilla-inbound%20opt%20test%20mochitest-9
- [4:37pm] • philor backspaces through several descriptions of the before state as being too graphic
- [4:38pm] glandium: botond_: seems to me you got more than what was breaking, and that it's a big patch to land r=bustage...
- [4:39pm] botond_: glandium: i did a local build to make sure to get all bustages. all the changes i made were necessary to fix a Werror
- [4:39pm] glandium: botond_: I don't know how you did your local build, but you clearly didn't do it like automation does, because -Werror is not used in mozglue/build
- [4:40pm] glandium: so Nuwa.cpp didn't need a change
- [4:40pm] botond_: glandium: hmm. that's strange
- [4:40pm] jya: philikon: the test_mixingRules is known to be buggered, and that’s what bholley changed just fixed
- [4:40pm] glandium: in fact, that file is probably one of the first things to be built
- [4:41pm] botond_: glandium: my patch _introduces_ the Werror flag on b2g - could it be that i introduced it in more places than desirable?
- [4:41pm] jya: philor: ^
- [4:42pm] botond_: glandium: here's how i did it: https://hg.mozilla.org/integration/mozilla-inbound/rev/42e5e10c44f5
- [4:42pm] jya: looking at the try test, it’s the only test failing currently, everything else are either build error (nothing to do with my changes) or unable to connect stuff
- [4:42pm] glandium: botond_: 1073003? no, it looks fine. although since when is mwu a build config peer?
- [4:43pm] glandium: ... and that's not how we enaaaaaaaaaaaaable it
- [4:43pm] glandium: because there you force it for *everyone*
- [4:43pm] glandium: we don't do that
- [4:43pm] glandium: although the toolchain is more controlled in b2g caaaaaaaaaaaaase
- [4:45pm] bbondy left the chat room. (Connection closed)
- [4:46pm] botond_: glandium: you're right, mwu is not a build config peer. that was my bad - the patch was originally going to be to <b2g>/gonk-misc/default-gecko-config, and then mwu suggested changing b2g/confvars.sh instead, and i failed to re-evaluate who should review that
- [4:46pm] jya: philor: test_BufferingWait.html is the only one failing with timeout in those tests. and it was disabled in https://hg.mozilla.org/mozilla-central/rev/0532f2509f3f (which was committed after my changes)
- [4:46pm] botond_: glandium: how would you suggest enabling it instead?
- [4:46pm] boris left the chat room. (Quit: Leaving.)
- [4:46pm] glandium: botond_: b2g/config/mozconfigs/*/*
- [4:47pm] Jan\ joined the chat room.
- [4:47pm] bwc joined the chat room.
- [4:47pm] philor: jya: I know what you are intending to say to me, but all I'm hearing is "completely and utterly ignore every media test, ALWAYS"
- [4:47pm] philor: good lord
- [4:47pm] jrgill joined the chat room.
- [4:48pm] elin left the chat room. (Client exited)
- [4:48pm] jya: it’s certainly not what I’m trying to say I don’t see anything in any of the try tests there indicating an issue caused by my changes.
- [4:48pm] philor: "I screwed up addressing a review comment, but when I properly address it a crap test that fails intermittently becomes permaorange, so I'll just disable it, referencing the intermittent bug"
- [4:48pm] glandium: botond_: https://hg.mozilla.org/integration/mozilla-inbound/diff/ad0bb596da2c/dom/media/omx/mediaresourcemanager/moz.build did you *really* mean to add that FINAL_LIBRARY?
- [4:49pm] glandium: probably not, since there's already one 10 lines below
- [4:49pm] botond_: glandium:
- [4:49pm] • philor takes back his a=
- [4:49pm] botond_: glandium: i did not
- [4:49pm] bholley joined the chat room.
- [4:49pm] jya: bholley: ping
- [4:49pm] bholley: jya: hi
- [4:50pm] glandium: botond_: that's why there's code review. IMHO, you should back out your stuff, split that patch, and get review for each part
- [4:50pm] Waldo left the chat room. (Quit: ChatZilla 0.9.87-8.1450hg.fc20 [XULRunner 32.0/20140902134853])
- [4:50pm] bholley: jya: better be quick, otherwise it's back to my glass of wine
- [4:51pm] jya: can you address philor concerns above? this is in regards to media mochitest failing with timeout (test_BufferingWait.html)
- [4:51pm] mixedpuppy left the chat room. (Client exited)
- [4:51pm] Mossop joined the chat room.
- [4:51pm] jya: due to the poor timing coincidence of commits
- [4:51pm] wesley_huang joined the chat room.
- [4:51pm] glandium: botond_: I'm not particularly thrilled by the -Wno-error added here and there
- [4:51pm] Mossop left the chat room. (A TLS packet with unexpected length was received.)
- [4:51pm] jya: bholley: obviously it makes it look like my AnnexB changes caused it
- [4:51pm] mixedpuppy joined the chat room.
- [4:51pm] bholley: jya: "above"?
- [4:51pm] botond_: glandium: it's just more of the same as in the reviewed patches of bug 1073003 and bug 1073081
Advertisement
Add Comment
Please, Sign In to add comment