| 19 Apr 2026 |
 Randy Eckenrode | emily Let me know if there’s any follow-up to my replies to your questions. Also regarding the 26.4 SDK fix, another reason I wanted to avoid the cctools-port fix is it mentioned decompilation in the PR. I’m not sure how it was used, but I figured it was better to finish my work and independently derive a fix. | 13:51:27 |
 WeetHet | You still need to force it with GSK_RENDERER=vulkan | 13:51:43 |
| WeetHet | Otherwise it uses GL | 13:51:49 |
 emily | yep, will take a look soon (I might have also forgotten about other PRs pending on me if there are any) | 13:52:16 |
| Randy Eckenrode |
- https://github.com/NixOS/nixpkgs/pull/511211
- https://github.com/NixOS/nixpkgs/pull/511070
- https://github.com/NixOS/nixpkgs/pull/485980
| 13:53:45 |
| Randy Eckenrode | Those are my active ones. The 26.4 SDK one will be coming shortly. | 13:53:53 |
| Randy Eckenrode | emily, here’s the 26.4 SDK: https://github.com/NixOS/nixpkgs/pull/511399 | 13:58:46 |
|  qwrdd joined the room. | 13:59:56 |
| qwrdd | Hello! As the next release will drop x86_64-darwin, what your thoughts about adding rosetta flag into linux-builder? | 14:01:21 |
| Randy Eckenrode | https://github.com/NixOS/nixpkgs/pull/511400 fixes Rubberband. It was the only fallout I encountered from the darwin.libcxx update. | 14:02:15 |
| emily | we did most of the libc++ 21 fixing already 😅 | 14:05:21 |
| Randy Eckenrode | Somehow that got missed. | 14:05:48 |
| Randy Eckenrode | Are we going to update to LLVM 22 by default for 26.05, or has the window passed for that? | 14:06:26 |
| emily | technically we have a week, pragmatically we don't | 14:09:18 |
| emily | https://nixos.github.io/release-wiki/Release-Critical-Packages.html oh actually no | 14:09:27 |
| emily | LLVM is on there, it's too late | 14:09:35 |
| Randy Eckenrode | I guess Swift may slip to post 26.05, unfortunately. | 14:11:35 |
| Randy Eckenrode | Biggest help for Swift right now would be using my branch to build things and find problems. That’s how the issue with in-tree SwiftPM dependencies was uncovered. | 14:13:02 |
| Randy Eckenrode | https://github.com/NixOS/nixpkgs/actions/runs/24630997267/job/72018257578#step:3:1 | 14:27:34 |
| Randy Eckenrode | Did the Git error just
hit CI? | 14:27:43 |
| Randy Eckenrode | No, GitHub appears to be having issues. | 14:29:58 |
| Randy Eckenrode | https://github.com/NixOS/nixpkgs/actions/runs/24630997244/job/72018257140?pr=511211#step:4:39 in another PR. | 14:30:15 |
 viraptor | Looks like that's my next blocker. I'm stuck on failed to build module 'Swift'; this SDK is not supported by the compiler (the SDK is built with 'Apple Swift version 5.10 (swiftlang-5.10.0.12.5 clang-1500.3.9.1.1)', while this compiler is 'Swift version 6.2.4 (swift-6.2.4-RELEASE)') currently. But at least I've got all the other required changes to merge in the meantime. | 23:25:48 |
| Randy Eckenrode | Is that using the current Swift? | 23:44:14 |
| Randy Eckenrode | * | 23:44:21 |
| viraptor | From your pre-rebase mk2 branch - https://github.com/viraptor/nixpkgs/tree/hammerspoon-swbuild | 23:46:01 |
| viraptor | I'll check today if anything changed post-rebase. | 23:47:54 |
| 20 Apr 2026 |
| viraptor | * From your pre-rebase mk2 branch - https://github.com/viraptor/nixpkgs/tree/hammerspoon-swbuild
(all the changes are "YOLO, just make it work" level, it will be cleaned up a lot before submitting anywhere) | 00:25:30 |
| Randy Eckenrode | Is it using the wrong SDK? Swift is really sensitive to SDK version. | 00:44:29 |
| Randy Eckenrode | It should be propagating the 26.x SDK though. | 00:44:38 |
