**** BEGIN LOGGING AT Wed Nov 02 02:59:57 2022
Nov 02 08:19:44 <JaMa> pushed small kirkstone update which will keep you warm :)
Nov 02 08:20:19 <JaMa> Tofe: I would expect it to create remote branch called "origin/tofe/window-open", but I don't see it created on github, did you delete it afterwards?
Nov 02 08:45:00 <Herrie> Well I see this branch? https://github.com/webOS-ports/wam/tree/tofe/window-open
Nov 02 08:45:34 <JaMa> yes, I've seen this one, but not "origin/tofe/window-open" one
Nov 02 08:46:37 <JaMa> martin@jama /OE/projects/git-repos/webos-ports/webos-ports-setup $ git push origin origin/jansa/mickledore 
Nov 02 08:46:40 <JaMa> Total 0 (delta 0), reused 0 (delta 0), pack-reused 0
Nov 02 08:46:43 <JaMa> To github.com:webOS-ports/webos-ports-setup.git
Nov 02 08:46:45 <JaMa>  * [new reference]   origin/jansa/mickledore -> origin/jansa/mickledore
Nov 02 08:46:48 <JaMa> martin@jama /OE/projects/git-repos/webos-ports/webos-ports-setup $ git branch -a | grep mickledore
Nov 02 08:46:51 <JaMa>   jansa/mickledore
Nov 02 08:46:53 <JaMa>   mickledore
Nov 02 08:46:56 <JaMa>   remotes/origin/jansa/mickledore
Nov 02 08:46:58 <JaMa> same here, so somehow it doesn't create a branch
Nov 02 08:47:00 <JaMa>   remotes/origin/mickledore
Nov 02 08:47:53 <Herrie> Maybe some new "improvement" from MS ?
Nov 02 08:50:38 <JaMa> I don't know if it changes anything as long as you really don't want to create a branch with such confusing name
Nov 02 08:54:18 <JaMa> it just stripps origin/ from :dst param and pushes origin/jansa/mickedore as :src
Nov 02 08:54:37 <JaMa> if I rebase my local jansa/mickledore branch:
Nov 02 08:54:38 <JaMa> martin@jama /OE/projects/git-repos/webos-ports/webos-ports-setup $ git push origin origin/jansa/mickledore 
Nov 02 08:54:41 <JaMa> Everything up-to-date
Nov 02 08:54:44 <JaMa> martin@jama /OE/projects/git-repos/webos-ports/webos-ports-setup $ git push origin jansa/mickledore:origin/jansa/mickledore 
Nov 02 08:54:47 <JaMa> To github.com:webOS-ports/webos-ports-setup.git ! [rejected]        jansa/mickledore -> origin/jansa/mickledore (non-fast-forward)
Nov 02 08:55:45 <JaMa> the remote ref needs to exist in origin, otherwise it fails with reasonable error message showing it's src refspec:
Nov 02 08:55:48 <JaMa> martin@jama /OE/projects/git-repos/webos-ports/webos-ports-setup $ git push origin origin/jansa/mickledore-typo
Nov 02 08:55:51 <JaMa> error: src refspec origin/jansa/mickledore-typo does not match any
Nov 02 08:56:47 <JaMa> so the only issue Tofe could have, is that the push didn't push his updated local "tofe/window-open" branch, easily fixed by another push without the origin/ prefix
Nov 02 12:48:15 <Tofe> JaMa: ok, well, then all is good
Nov 02 18:44:53 <Tofe> Herrie: ok, when inspecting with Chromium's devtool, I can see that systemui does have the device pixel ratio applied for its primary webpage, which has no window
Nov 02 18:45:07 <Tofe> but the next window does not
Nov 02 18:45:39 <Herrie> Ah so could be your PR needs some extending to cover next window(s) too?
Nov 02 18:46:51 <Tofe> not sure yet
Nov 02 18:47:32 <Tofe> for instance, it doesn't explain why the Messaging app, which only has one window, is directly rendered with 720x1440 (i.e. no scaling)
Nov 02 18:47:39 <Herrie> At least having this is a lead so that's good
Nov 02 18:49:40 <Tofe> well it's still quite a weak lead for the moment
Nov 02 18:50:37 <Herrie> It could be verified with email and calendar which have noWindow too?
Nov 02 18:50:54 <Tofe> for calendar we have the same issue yes
Nov 02 18:54:46 <Herrie> If I recall email was noWindow too (and legacy messaging app too)
Nov 02 19:42:34 <Tofe> mmh weird, but may be a lead too: with a forced scaling of 3.0, starting a Calendar window fails; in the logs, we see that webruntime tries to create an output buffer of 2160x4320, i.e. 3 times too large and too high
Nov 02 19:46:02 <Tofe> I guess there is a place in the code where we create a window of size (screen_width,screen_height), instead of (screen_width/scale,screen_height/scale)
**** ENDING LOGGING AT Thu Nov 03 02:59:56 2022