=== maclin1 is now known as maclin === cwayne_ is now known as cwayne [07:43] What is the proper name for the indicator pulldown thing? [07:51] indicator panel [08:19] Saviq: Ta [08:20] duflu: hey, with mir16 landed, is your qtmir work on better buffer management ready? [08:21] greyback_: I started from scratch due to the major qtmir reworks. And then stopped because your multi-monitor work was pending. When the dust settles I will have to rewrite it, from scratch [08:21] greyback_: But in other news I will try to propose touch performance improvements today [08:22] duflu: ok [08:27] greyback_: Hard to be sure but I think the phones are smoother with 0.16 [08:27] Because the change is in USC at least [08:38] dednick: ping [09:03] ltinkl: want to have a look at your needs fixing in https://code.launchpad.net/~nick-dedekind/unity8/lp1378821.time-translation/+merge/271452 or should i top approve? [09:10] greyback_, shall I add the qml cache branch to silo? [09:11] you got a review lined up? [09:11] Saviq: we in feature freeze now, right? Can we even land mousePointer now? [09:12] greyback_, we're not in main so we're fine, and anyway we can land to overlay already I believe [09:12] I mean wily+overlay [09:13] Saviq: even with OTA7 feature freeze? [09:13] ah, we finally got there? [09:13] we'd want the two qtmir fixes anyway I'd say, without mouse pointer potentially [09:13] * Saviq checks with sil [09:15] greyback_: Are apps usually upstarted or just the dash? [09:19] greyback__: Are apps usually upstarted or just the dash? === greyback__ is now known as greyback [09:19] duflu: dash has it's own dedicated upstart job. apps started via upstart too, but through upstart-app-launch [09:19] which is the same upstart job run for each app [09:19] duflu: get my "dash has it's own..." comment? [09:20] -' [09:20] greyback: OK, thanks. So the global upstart environment should affect all phone apps? [09:20] duflu, yes [09:21] duflu, the global *session* environment, yes [09:21] Words [09:21] In some order [09:21] duflu: yes. /usr/share/upstart/sessions/application-click.conf is the typical job for apps, which will inherit the session env [09:22] I often use "initctl set-env --global VAR=val" [09:25] Hmm, when all done my afternoon's work may come to a couple of lines of script [09:25] But still very useful script [09:41] greyback, ok, I moved mousePointer out into a separate train request, https://requests.ci-train.ubuntu.com/#/ticket/420 is what I want to land today [09:41] greyback, so we'll need a review on your branch [09:42] dednick: can you take? ^^ [09:46] greyback: sure [09:47] ta === marcusto_ is now known as marcustomlinson [10:18] Saviq: approved greyback's branch [10:33] dednick, tx [10:40] dednick: couple of extra tags appeared in your touch_tracing branch [10:40] for me anyway [10:41] greyback: hm. thought i just deleted them. let me check again [10:41] it might be me [10:43] greyback: removed [10:43] dont know where they came from [10:43] me neither [10:44] just deleted all of them a couple of hours ago! [10:44] greyback: huh. they just appeared on my other branch as well. [10:44] :/ [10:44] dednick, greyback, I've seen some tags being kept around locally even though they're considered bad remotely (for me it's likely because I use colocated repos) [10:45] Saviq: i deleted everything with ? manually [10:45] dednick: please merge trunk into touch_tracing too, there's a conflict [10:45] dednick, right, yeah [10:46] greyback, I forgot to remove unity-api from the silo? [10:46] Saviq: I think you removed the MR, but the packages aren't cleared from the PPA automatically. Need to ask trainguard to do that [10:47] greyback, that's what I meant, yeah [10:47] yep === marcusto_ is now known as marcustomlinson [10:59] greyback, Saviq: are the mousePointer branches in a silo? [11:00] ltinkl, had to pull them out to land some fixes [11:00] ltinkl, let me kick a build of those [11:01] Saviq, I see; I was wondering whether the liveCaption shouldn't go with them since I had to bump the API level and also rebase on top of them; if the mousePointer stuff gets landed separately and before the liveCaption, wouldn't that have to be done again? [11:01] ltinkl: want to have a look at your needs fixing in https://code.launchpad.net/~nick-dedekind/unity8/lp1378821.time-translation/+merge/271452 or should i top approve? [11:02] tsdgeos, a sec [11:03] tsdgeos, it's OK, just top approve [11:03] ltinkl, if ready to land, sure (we then only need a single api bump, too) [11:07] mzanetti, can you please coordinate with greyback and put all interesting MPs into https://requests.ci-train.ubuntu.com/#/ticket/445 (stuff that's only gonna land after OTA7 freeze) [11:07] I'm gonna test https://requests.ci-train.ubuntu.com/#/ticket/420 [11:08] Saviq, when is the OTA-7 freeze? [11:08] mzanetti: now :) [11:08] oh :D [11:08] feathre freeze [11:08] ok... [11:15] Saviq, not exactly sure what you're asking for tbh... [11:16] what intersesting stuff? [11:16] mzanetti, anything that we should target for the next landing, like ltinkl's window title set [11:17] right... but we're not gonna land this silo within the next 2 wooks anyways, right? [11:17] weeks [11:17] mzanetti, more like days [11:18] I don't think we're freezing for that long any more? [11:18] sil2100, what's the expected OTA7 thaw date? [11:20] Saviq: nowadays we usually do a snapshot on final freeze and unblock the overlay PPA for future landings [11:21] right, so days, hopefully [11:21] Yeah, final freeze is next week, so around 1 week I would say [11:22] sil2100, by "doing a snapshot" do you mean actually copying the whole PPA? is that a potential landing target if fixes are needed for the OTA? [11:23] Saviq: more or less... the snapshot copies the whole PPA but we don't enable it as a target - if a change is needed for the given OTA that has been snapshotted we first request landing it to the overlay and then cherry-pick it to the snapshot [11:24] ack [11:24] To make sure we only have things that are really released :) [11:24] (the idea is that after snapshotting not too many changes need to be copied) [11:25] right, but has the added danger of having to pull in more changes than actually needed? if a project moved after the freeze was lifted? [11:25] or do you actually require rollbacks in that case? [11:30] sil2100, pointing up so you don't miss my question ↑ [12:09] larsu, hey, is it expected that indicator-session-service takes like 400MB RSS? [12:19] pstolowski, karni, I'm seeing telegram scoperunner hogging CPU and crashing, that known? [12:20] Saviq: we're aware there was an issue that it crashed, we have not found the problem yet (more on that shortly). however, I did not know it was actually hogging the CPU. [12:20] karni, I think that might be a symptom of the crash actually [12:20] Saviq: more: I rewrote the scope basically from scratch, and did see the same issue appear once [12:21] Saviq: could it be apparmor collecting info? [12:21] karni, or just before, yeah [12:22] Saviq: we'll be starting testing the v2 shortly, hopefully the situation with the scope will at least improve. if not, I'll want more eyes on this source, because it seems unprobable a brand new code would have the same issue. [12:22] karni, ktx [12:36] Saviq: err, probably not :) [12:36] larsu, ok, I'll try and gather more info and report [12:37] Saviq: doesn't for me right now, but yeah, please report [12:44] ltinkl: added the QLaint1Char [12:48] mzanetti, I don't know if you saw, but I also made a unity8 branch that consumes the new isTouchApp/canSuspend bits from qtmir: https://code.launchpad.net/~mterry/unity8/no-touch-no-lifecycle/+merge/272844 [12:48] mterry, already rebased my stuff on that one [12:48] mzanetti, awesome [13:17] mzanetti-: Saviq: some more free speedup https://code.launchpad.net/~aacid/unity8/fast_string_concatenation/+merge/272904 [13:23] tsdgeos, some stuff there is... unexpected :) [13:23] ltinkl: well it has the downside it doesn't autoconvert to qvariant anymore [13:24] which imho is already a bit sloppy on our side [13:24] so you need the casts to QString for the autoconvert to happen [13:24] tsdgeos, have you tried with QStringBuilder and the "%" operator? [13:24] it's exactly the same thing [13:25] tsdgeos, heh right, I didn't notice the add_definitions(-DQT_USE_QSTRINGBUILDER), sorry :) [13:27] no worries === mzanetti- is now known as mzanetti [14:17] dandrader, https://code.launchpad.net/~lukas-kde/unity8/liveCaption/+merge/272920 resubmitted and fixed [14:19] ltinkl, ok [14:20] ltinkl, what I mean is that we don't need this Connections {} in SurfaceContainer at all [14:21] dandrader, are you sure? [14:21] ltinkl, it's doing the same thing as the binding [14:21] ltinkl, I don't see the point [14:22] ltinkl, it will actually break the binding the first time it's executed. so it will effectively replace it [14:22] ltinkl, by "binding" I mean "property string name: surface ? surface.name : """ [14:22] dandrader, right, it makes no sense after removing the condition [14:26] dandrader, ok done, hopefully that's all there :) [14:27] ltinkl, all approved [14:28] dandrader, qtubuntu too? :) [14:29] dandrader, https://code.launchpad.net/~lukas-kde/qtubuntu/liveCaption/+merge/272756 that one is a no brainer imo [14:30] @unity: standup [14:30] @dednick [14:31] ltinkl, forgot about this one. so many branches. done [14:32] dandrader, great, thanks [14:42] ltinkl, hey, you have the liveCaption branches based on top of mouseCursor === dandrader is now known as dandrader|afk [14:42] ltinkl, except the qtmir branch... I think you want to have that too [14:42] mzanetti, yea... I guess I'll have to do that as well [14:43] Saviq, added those branches to the silo... but when I pressed save I saw you clicked build in the meantime :) [14:43] you might want to retrigger that at some point [14:43] Saviq, first I think ltinkl needs to resubmit one of the branches [14:44] mzanetti, Saviq: on it now [14:45] mzanetti, Saviq: https://code.launchpad.net/~lukas-kde/qtmir/liveCaption/+merge/272921 (someone might want to re-top approve) [14:46] mzanetti, oh yeah, we will rebuild it a few times still [14:46] Saviq, yep... just saying as I changed it *while* you were building [14:48] ltinkl, this does not bump the version: https://code.launchpad.net/~lukas-kde/unity-api/liveCaption/+merge/272769 [14:49] mzanetti, because the other does [14:49] are we ok with that? [14:50] I guess in practice we could be... [14:50] in theory it would mess with reverting one or the other if we ever would [14:51] when reverting we'd probably bump it even higher [14:51] unless reverting the whole release [14:52] it could bite us if we only landed the bump and then separately landed the other one [14:52] right... [14:52] ok well... wfm [15:05] greyback, testing silo 15 on wily/mako, connected external screen, unity8 went away and crashed (external screen black), that expected? [15:05] no [15:06] external screen will remain black [15:06] but no crash [15:06] yeah, crash here [15:06] Saviq: had you tapped on the screen at all? [15:06] greyback, don't think so [15:06] backtrace please [15:07] Saviq: hang on, silo15 doesn't have multimonitor [15:07] greyback, no [15:07] without multimonitor, unity8 always crashed on external screen connection [15:07] greyback, but qtmir test plan includes connecting a monitor [15:07] ok, so expected [15:08] Saviq: ah yeah, I added that when I was trying to land MM a while ago [15:08] good enough for me [15:09] bbl === dandrader|afk is now known as dandrader [15:16] ltinkl, in https://code.launchpad.net/~lukas-kde/qtmir/liveCaption/+merge/272921 you added mousePointer as a prereq but you didn't rebase the branch on top of it, or merged mousePointer [15:20] dandrader, on it, thx [15:26] dandrader, done === pat_ is now known as Guest33589 === alexabreu is now known as alex-abreu [16:05] greyback: is silo 15 ready for testing again? [16:06] mhall119: if you're looking for mouse pointer, it's not there any more. [16:06] dang it [16:07] Saviq: any idea what the silo22 failure is about? "ERROR qtmir None not found in Silo PPA. unity8 None not found in Silo PPA" [16:08] greyback: so I'm going to be running a booth at FOSSETCON in November, and I'm really hoping to demo convergence on my Nexus 4, any chance of having an "MWC demo" style silo again for that purpose? [16:08] mhall119: by Nov, we should have many of those bits landed [16:09] cool, well I'll just keep checking back between now and then [16:09] mhall119: yeah, that would be best. Things moving quickly atm (aside from the FF ofc) [16:10] yeah, darn freezes, keeping my phone stable for days at a time :( [16:13] ltinkl: if you're still there, could you please rebase https://code.launchpad.net/~lukas-kde/qtmir/liveCaption/+merge/272921 on top of https://code.launchpad.net/~alan-griffiths/qtmir/small-refactoring-of-MirWindowManager [16:13] ltinkl: actually no, please ignore that [16:14] train appears to apply branches in peculiar order [16:15] dednick: can you please merge trunk into https://code.launchpad.net/~unity-team/qtmir/touch_tracing/+merge/271655 [16:15] greyback: sure [16:16] ah. forgot to push. thought i did it already. [16:16] yeah I said before, but wasn't sure you got my mesage [16:16] greyback: done. [16:17] dednick: thanks. Let's see how much it conflicts with everything else in the MR queue [16:17] :) === dandrader is now known as dandrader|afk [17:32] greyback, I think train just got dumb, rebuild should help [17:36] greyback_, I think train just got dumb, rebuild should help [17:36] greyback_, alan's refactoring branch is conflicting? again? or maybe with some other branch in the queue [17:37] Saviq: conflict with Lukas' branch I think === dandrader|afk is now known as dandrader === ljp is now known as lpotter === boiko__ is now known as boiko