!lheuhImcToQZYTQTuI:nixos.org

Nix on macOS

1158 Members
“There are still many issues with the Darwin platform but most of it is quite usable.” — http://yves.gnu-darwin.org186 Servers

Load older messages


SenderMessageTime
6 Oct 2025
@reckenrode:matrix.orgRandy EckenrodeApple merged LaunchPad with Spotlight in macOS 26.18:19:56
7 Oct 2025
@reckenrode:matrix.orgRandy EckenrodeFrom what I understand, they still “work”, but there’s no way to view them with applications. They show up as aliases.00:23:56
@eureka-cpu:matrix.orgeureka-cpuI just keep applications in the dock after opening them once. It's a bit annoying that they show up as a ? but after you click on them they show the icon.01:07:04
@Cu3PO42:matrix.orgCu3PO42

Can anyone build _1password-cli on current nixos-unstable on aarch64-darwin? I'm getting errors while generating the shell completions which I'm not totally sure are actually the derivation's fault. I unfortunately don't have another Mac at hand to test.

The full error suggests permission errors around the build directory, which I wasn't able to verify. I inserted an id -u and ls -la $TMPDIR to check. On older nixpkgs, I was able to build v2.31.1 (current is v2.32.0), but simply downgrading the sources for the derivation does not resolve the issue. It would seem something has indeed changed around the sandbox, but I'm not sure where else to look.

I'm also trying to bisect, but there's a lot of building LLVM that's taking a bit of time.

08:28:38
@toonn:matrix.orgtoonn Cu3PO42: FWIW, you shouldn't use nixos-* channels on Darwin. Nixpkgs-unstable has some extra tests for Darwin. 11:36:23
@Cu3PO42:matrix.orgCu3PO42Duly noted, thanks for the info. Unfortunately, it doesn't resolve this error. I've 'worked around it' by simply removing the shell completions so I can build my HM environment, but I'm still trying to figure out the root cause.11:41:35
@reeve567:matrix.orgXwy joined the room.16:14:58
@reeve567:matrix.orgXwyhad to take my laptop into the apple store for damage and was surprised to open my laptop up to tahoe the other day, worst surprise they could've thrown at me16:17:16
@skyraider256525:matrix.orgskyraiderI was thinking of upgrading once 26.01.2 was out, but now I will wait for the green signal here 😁17:07:40
@weethet:catgirl.cloudWeetHet samasaur: have you solved cpp interop? 19:16:34
@weethet:catgirl.cloudWeetHetIf not, do you think it's needed and does vlc/dotnet build?19:16:52
@weethet:catgirl.cloudWeetHetI can look into it a bit later if needed19:17:46
@samasaur:matrix.orgsamasaur
In reply to @weethet:catgirl.cloud
samasaur: have you solved cpp interop?
I have not, had a very busy weekend (was a groomsman in my friends' wedding!) and have not been able to take a look in a while
20:03:05
@samasaur:matrix.orgsamasaurI'm not sure whether it's needed, honestly20:03:16
@samasaur:matrix.orgsamasaurI'd like it to work because it worked with our swift 5.8 but I don't know if anything uses it, given that it was experimental in 5.820:03:44
@samasaur:matrix.orgsamasaurso it could perhaps be put off to a follow-up PR20:04:08
@reckenrode:matrix.orgRandy EckenrodeThe goal right now should be to get enough of a Swift working for 25.11 with the focus being on 6.2 (and newer) for 26.05.20:04:59
@samasaur:matrix.orgsamasaur emily: you were the one to add swift.tests.cxx-interop-test, was there a specific downstream dependent or was it just to ensure the feature was working? 20:05:37
@samasaur:matrix.orgsamasaur
In reply to @reckenrode:matrix.org
The goal right now should be to get enough of a Swift working for 25.11 with the focus being on 6.2 (and newer) for 26.05.
in that case it might be good as is!
20:11:20
@samasaur:matrix.orgsamasauri have a couple local changes for cxx interop that are not pushed (changing the flag name in the wrapper) but otherwise i think everything is up and as best i can tell all the other swift functionality works20:13:15
8 Oct 2025
@samasaur:matrix.orgsamasaurfascinating05:30:39
@samasaur:matrix.orgsamasaur

i figured i should make sure that the cxx interop example builds with a non-nixpkgs setup so I've been testing that. it works with swift 6.2 from Xcode. then i figured i should try swift 5.10.1, so I got it via swiftly and tried that, only to get the following error:

Building for debugging...
error: emit-module command failed due to signal 6 (use -v to see invocation)
error: compile command failed due to signal 6 (use -v to see invocation)
Assertion failed: (T && "could not create debug info for type"), function getOrCreateStandaloneType, file CGDebugInfo.cpp, line 2358.
Please submit a bug report (https://swift.org/contributing/#reporting-bugs) and include the crash backtrace.
Stack dump:
0.      /Applications/Xcode-26.0.1.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX26.0.sdk/usr/include/c++/v1/__type_traits/void_t.h:30:41: at annotation token
Stack dump without symbol names (ensure you have llvm-symbolizer in your PATH or set the environment var `LLVM_SYMBOLIZER_PATH` to point to it):
0  swift-frontend           0x00000001075895d0 llvm::sys::PrintStackTrace(llvm::raw_ostream&, int) + 56
1  swift-frontend           0x00000001075889bc llvm::sys::RunSignalHandlers() + 112
2  swift-frontend           0x0000000107589c0c SignalHandler(int) + 304
3  libsystem_platform.dylib 0x000000019c0216a4 _sigtramp + 56
4  libsystem_pthread.dylib  0x000000019bfe7848 pthread_kill + 296
5  libsystem_c.dylib        0x000000019bef09e4 abort + 124
6  libsystem_c.dylib        0x000000019beefc18 err + 0
7  swift-frontend           0x0000000107eee1d4 clang::CodeGen::CGDebugInfo::addHeapAllocSiteMetadata(llvm::CallBase*, clang::QualType, clang::SourceLocation) (.cold.1) + 0
8  swift-frontend           0x0000000104e372b4 clang::CodeGen::CGDebugInfo::getOrCreateStandaloneType(clang::QualType, clang::SourceLocation) + 312
9  swift-frontend           0x00000001050c66b4 clang::RecursiveASTVisitor<(anonymous namespace)::PCHContainerGenerator::DebugTypeVisitor>::TraverseTypeAliasDecl(clang::TypeAliasDecl*) + 28
10 swift-frontend           0x00000001050c2a64 clang::RecursiveASTVisitor<(anonymous namespace)::PCHContainerGenerator::DebugTypeVisitor>::TraverseDecl(clang::Decl*) + 872
11 swift-frontend           0x00000001050c5698 clang::RecursiveASTVisitor<(anonymous namespace)::PCHContainerGenerator::DebugTypeVisitor>::TraverseTypeAliasTemplateDecl(clang::TypeAliasTemplateDecl*) + 36
12 swift-frontend           0x00000001050c29c4 clang::RecursiveASTVisitor<(anonymous namespace)::PCHContainerGenerator::DebugTypeVisitor>::TraverseDecl(clang::Decl*) + 712
13 swift-frontend           0x00000001050c41d4 clang::RecursiveASTVisitor<(anonymous namespace)::PCHContainerGenerator::DebugTypeVisitor>::TraverseNamespaceDecl(clang::NamespaceDecl*) + 60
14 swift-frontend           0x00000001050c28e4 clang::RecursiveASTVisitor<(anonymous namespace)::PCHContainerGenerator::DebugTypeVisitor>::TraverseDecl(clang::Decl*) + 488
15 swift-frontend           0x00000001050c41d4 clang::RecursiveASTVisitor<(anonymous namespace)::PCHContainerGenerator::DebugTypeVisitor>::TraverseNamespaceDecl(clang::NamespaceDecl*) + 60
16 swift-frontend           0x00000001050c28e4 clang::RecursiveASTVisitor<(anonymous namespace)::PCHContainerGenerator::DebugTypeVisitor>::TraverseDecl(clang::Decl*) + 488
17 swift-frontend           0x00000001050c0efc (anonymous namespace)::PCHContainerGenerator::HandleTopLevelDecl(clang::DeclGroupRef) + 164
18 swift-frontend           0x0000000105a1d544 clang::MultiplexConsumer::HandleTopLevelDecl(clang::DeclGroupRef) + 60
19 swift-frontend           0x0000000105b84cac clang::ParseAST(clang::Sema&, bool, bool) + 628
20 swift-frontend           0x00000001059eef28 clang::FrontendAction::Execute() + 100
21 swift-frontend           0x0000000105988680 clang::CompilerInstance::ExecuteAction(clang::FrontendAction&) + 808
22 swift-frontend           0x00000001059921a0 void llvm::function_ref<void ()>::callback_fn<compileModuleImpl(clang::CompilerInstance&, clang::SourceLocation, llvm::StringRef, clang::FrontendInputFile, llvm::StringRef, llvm::StringRef, llvm::function_ref<void (clang::CompilerInstance&)>, llvm::function_ref<void (clang::CompilerInstance&)>)::$_8>(long) + 128
23 swift-frontend           0x00000001074da2a0 llvm::CrashRecoveryContext::RunSafely(llvm::function_ref<void ()>) + 208
24 swift-frontend           0x00000001074da468 RunSafelyOnThread_Dispatch(void*) + 52
25 swift-frontend           0x00000001074da4c0 void* llvm::thread::ThreadProxy<std::__1::tuple<void (*)(void*), (anonymous namespace)::RunSafelyOnThreadInfo*>>(void*) + 24
26 libsystem_pthread.dylib  0x000000019bfe7bc8 _pthread_start + 136
27 libsystem_pthread.dylib  0x000000019bfe2b80 thread_start + 8

I get this error for other versions of swift installed from swiftly as well

05:33:37
@samasaur:matrix.orgsamasauryeah this is beyond me I think07:20:34
@samasaur:matrix.orgsamasaur WeetHet: here are the two extra commits that i think are necessary: https://github.com/Samasaur1/nixpkgs/tree/push-lrsvsqplzrnz 07:20:45
@samasaur:matrix.orgsamasaur setting NIX_DEBUG=1 on the interop test derivation shows that the wrapper is properly adding -Xcc -isystem -Xcc /nix/store/2zq6ygricncv5ffqgrv6mlwhqxyv5f8c-swift-5.10.1-dev/include/c++/v1 -I /nix/store/2zq6ygricncv5ffqgrv6mlwhqxyv5f8c-swift-5.10.1-dev/include/c++/v1 -Xcc -stdlib=libc++ when -cxx-interoperability-mode=default is present, but I'm still getting errors about missing APIs 07:23:06
@samasaur:matrix.orgsamasaurand I'm really pretty sure that that example should work, given that it works with Xcode-provided swift 6.207:23:52
@samasaur:matrix.orgsamasaurif you decide to take this on, good luck07:27:04
@samasaur:matrix.orgsamasauri fear the answer may be that I'm missing some build step that was added between 5.8 and 5.10.1, since I didn't go through the build script in detail07:31:06
@samasaur:matrix.orgsamasaurNEVER MIND08:16:35
@samasaur:matrix.orgsamasaurI GOT IT08:16:38

Show newer messages


Back to Room ListRoom Version: 6