| 25 Mar 2025 |
FliegendeWurst | try kcmshell6 kcm_kded | 22:27:45 |
FliegendeWurst | I can also still find the page via the kickoff search | 22:28:35 |
| 26 Mar 2025 |
K900 ⚡️ | Intentionally hidden upstream | 01:46:33 |
K900 ⚡️ | As it's not intended to be user facing | 01:46:45 |
K900 ⚡️ | Which is to say You Should Not Fuck With It | 01:46:56 |
| raboof left the room. | 02:18:10 |
K900 ⚡️ | Qt 6.9 is hopefully next week | 10:13:27 |
K900 ⚡️ | It should line up with the staging cycles roughly | 10:13:36 |
K900 ⚡️ | The question is do we go for it this close to freeze | 10:14:14 |
K900 ⚡️ | I think yes | 10:14:23 |
K900 ⚡️ | But if they delay again | 10:15:35 |
K900 ⚡️ | Then we're not going to have a good time | 10:15:44 |
| Marie joined the room. | 12:16:08 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 | I was asking about it when I wanted to disable kde-gtk-settings due to going from normal scaling to dpi scaling (Qt's upstream unwillingness to rework font rendering doesn't really leave any other option to my eyes) and the answer was it's to prevent uneducated users from ruinng their systems when they try to make the system more lightweight | 14:40:17 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 | * I was asking about it when I wanted to disable kde-gtk-settings due to going from normal scaling to dpi scaling (Qt's upstream unwillingness to rework font rendering doesn't really leave any other option to my eyes) and the answer was it's to prevent uneducated users from ruining their systems when they try to make the system more lightweight | 14:41:36 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 | i'd expect build failures and QThread-related crashes | 14:42:17 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 | But if you want, you can still do it via configs | 14:50:53 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 | I don't remember exactly which config it is though, as the machine where I was doing this is broken | 14:51:38 |
K900 ⚡️ | Well KDE folks at least run 6.9 betas in CI | 15:34:21 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 | https://lists.qt-project.org/pipermail/development/2025-March/046250.html | 15:35:15 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 |
[17:07:52] <jaheikki3_> Qt 6.9.0 release blocker list is empty and Qt 6.9.0 release content is frozen [17:08:03] <jaheikki3_> Release package creation is ongoing [17:08:06] <thiago> sorry, I found one yesterday [17:08:18] <jaheikki3_> thiago: ID? [17:08:22] <thiago> the qthread changes ended up with a no-op [17:08:51] <thiago> https://codereview.qt-project.org/c/qt/qtbase/+/634258 [17:09:25] <thiago> the task itself was reported as a memory leak, but a) the leak can't be fixed in the library and b) it was a symptom that led me to find the real issue [17:11:10] <jaheikki3_> and this is really must have for Qt 6.9.0? Because re-creating packages at this point causes a risk for the release schedule... [17:11:24] <jaheikki3_> vohi[M]: any opinion? [17:11:29] <thiago> I would prefer [17:11:42] <thiago> because we're leaking the main thread QAdoptedThread, so we'll get a bunch of bug reports about that [17:11:54] <thiago> more importantly, the fix may introduce issues and I'd rather people see them in .0 than in 6.9.1 [17:12:39] <vohi[M]> yeah, would be nice to get this in; autotests should catch if this breaks something, I think [17:12:51] <vohi[M]> thiago: move the change to 6.9.0 and pick it to dev [17:13:34] <jaheikki3_> yeah, that way we can speed up the progress. [17:14:18] <thiago> I can do that [17:14:20] <thiago> moving all three changes [17:15:28] <jaheikki3_> Ok, so let's take that still in. Hoping CI behaves and we can get packages for testing tomorrow [17:15:51] <jaheikki3_> vohi[M]: please review changes and I'll stage those in after approval
| 15:35:35 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 | GNOME nixpkgs PRs seem to get open early at beta releases of gtk and other components | 15:40:05 |
K900 ⚡️ | Yeah I tried building 6.9 locally and it was fine | 16:33:45 |
K900 ⚡️ | As in it built | 16:33:49 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 | The point is testing dependent packages and reporting failures eary though? | 17:39:40 |
𝙼𝚛. 𝙱𝚘𝚠𝚕𝚎𝚛 𝚠𝙷𝚊𝚝 | * The point is testing dependent packages and reporting failures early though? | 17:39:45 |
K900 ⚡️ | Well GNOME is a lot less API stable so things break more | 17:44:20 |
K900 ⚡️ | I can push my 6.9 beta branch if you're interested | 17:44:32 |
emily | it looks like the change in question was only merged hours ago | 17:46:25 |
K900 ⚡️ | Yeah delaying for last minute fixes is very classic Qt behavior | 17:47:26 |