rustc: fix pkgsLLVM #429745

RossComputerGuy · 2025-07-31T04:13:19Z

Things done

Fixes #311930

Works on aarch64-linux after stacking #409265

Add a 👍 reaction to pull requests you find important.

ccicnce113424 · 2025-07-31T04:33:24Z

I committed and tested commit 15470fe, and the build was successful without setting the rpath.

ccicnce113424 · 2025-07-31T07:22:49Z

pkgs/development/compilers/rust/rustc.nix

We don't need to set set the rpath here if we correct #320432 (comment).
We can also remove the fake libgcc here since it was already added in #330037.

pkgs/development/compilers/rust/rustc.nix

alyssais · 2025-08-04T12:21:17Z

pkgs/development/compilers/rust/rustc.nix

@@ -101,7 +101,9 @@ stdenv.mkDerivation (finalAttrs: {
        (stdenv.hostPlatform.isLinux && !withBundledLLVM && !stdenv.targetPlatform.isFreeBSD && useLLVM)
        "--push-state --as-needed -L${llvmPackages.libcxx}/lib -lc++ -lc++abi -lLLVM-${lib.versions.major llvmPackages.llvm.version} --pop-state"
    ++ optional (stdenv.hostPlatform.isDarwin && !withBundledLLVM) "-lc++ -lc++abi"
-    ++ optional stdenv.hostPlatform.isFreeBSD "-rpath ${llvmPackages.libunwind}/lib"
+    ++ optional (
+      with stdenv.hostPlatform; isFreeBSD || (isLinux && useLLVM && !withBundledLLVM)


Why not just check useLLVM though? If it's needed on both Linux and FreeBSD, it's probably not OS-specific, right?

Probably but what if someone sets useLLVM to false on FreeBSD? Or how does setting withBundledLLVM on BSD change things?

@rhelmot WDYT?

I'm really sorry I haven't had time to look into this. If I'm the only blocker, please merge it and I'll un-break FreeBSD when I get back to it.

Then let's be optimistic, and just set it to useLLVM && !withBundledLLVM.

blitz · 2025-08-13T18:43:52Z

@RossComputerGuy @alyssais It would be great to get this merged, because it blocks kernel builds on pkgsLLVM.

alyssais

Can the commit message please explain why each of these changes is being made?

RossComputerGuy · 2025-08-15T02:05:29Z

Can the commit message please explain why each of these changes is being made?

Done

alyssais · 2025-08-15T07:34:07Z

It doesn't really explain why still, just what changes are happening, which I can see from the diff. Why is each of these changes the correct thing to do?

RossComputerGuy · 2025-08-15T13:12:42Z

Idk how to explain why, it just does.

emilazy · 2025-08-15T14:19:13Z

I don't think it should be necessary to do any of this llvmPackages or llvmSharedFor stuff at all. Why wouldn't llvmPackages just work, even in pkgsLLVM?

I get the strong impression that there is something buggy with how the toolchain is wired up in pkgsLLVM that is creating the perceived need for all these ad-hoc unexplained workarounds for it across the tree. I know Darwin is a different case to pkgsLLVM in many ways, e.g. around the unwinder library and soon libc++, but it's similar enough that the fact that we don't need any of it there makes me strongly suspect the problem here is not with LLVM or Rust. (Though I wouldn't be surprised if there is a still a degree of "Linux = GCC" confusion in Rust's build system, but e.g. Chimera Linux uses an LLVM toolchain and has none of this stuff (other than perhaps this patch, but it seems to relate more to Musl).)

I am worried about adding to these hacks if we can't explain them, because when we start to move these yo principled toolchain conditions they will start to apply to Darwin and we will need to explain what they are actually conditioning on. I think there needs to be more work put into figuring out what the underlying problem with pkgsLLVM is before we extend workarounds that we already have no explanation for. I know there's at least one known issue there with the broken libgcc_s.so. If we're going to move our toolchain bootstrap and checks to be more granular and consistent then overriding stuff at random until it happens to build isn't the right approach.

RossComputerGuy · 2025-08-15T14:24:22Z

I don't think it should be necessary to do any of this llvmPackages or llvmSharedFor stuff at all. Why wouldn't llvmPackages just work, even in pkgsLLVM?

Because when rust is a dependency, it comes from buildPackages. When that happens, llvmPackages comes from another stage. The problem which made the special llvmPackages is because things were linking to libstdc++ instead of libc++. This is because things were being accessed too deep.

I did find an easier solution to the cc for one of the overridden dependencies which prompted the first change. However, the libunwind stuff broke since lib.optional was used which made libunwind not discoverable.

emilazy · 2025-08-15T14:28:44Z

If buildPackages uses libstdc++, then it's correct for the Rust compiler to link against it, right? Is this the problem where Rust can't distinguish between a build and host platform with the same target triple, or something else? It probably makes sense for pkgsLLVM to arrange for its buildPackages to use libc++, if it's intended to provide a "native" experience.

RossComputerGuy · 2025-08-15T14:31:40Z

If buildPackages uses libstdc++, then it's correct for the Rust compiler to link against it, right?

No because it's linking from something like pkgsLLVM.buildPackages.hostPackages.llvmPackages. At that stage, it is pre-LLVM so it ends up not linking to libc++.

ccicnce113424 · 2025-08-16T12:22:56Z

pkgs/development/compilers/rust/rustc.nix

@@ -101,7 +101,7 @@ stdenv.mkDerivation (finalAttrs: {
        (stdenv.hostPlatform.isLinux && !withBundledLLVM && !stdenv.targetPlatform.isFreeBSD && useLLVM)
        "--push-state --as-needed -L${llvmPackages.libcxx}/lib -lc++ -lc++abi -lLLVM-${lib.versions.major llvmPackages.llvm.version} --pop-state"
    ++ optional (stdenv.hostPlatform.isDarwin && !withBundledLLVM) "-lc++ -lc++abi"
-    ++ optional stdenv.hostPlatform.isFreeBSD "-rpath ${llvmPackages.libunwind}/lib"
+    ++ optional (useLLVM && !withBundledLLVM) "-rpath ${llvmPackages.libunwind}/lib"


In FreeBSD, we can try adding LLVM libunwind to buildInputs instead of setting rpath here. If moving libunwind to buildInputs works, we can remove it.

nixos-discourse · 2025-08-20T17:47:15Z

This pull request has been mentioned on NixOS Discourse. There might be relevant details there:

https://discourse.nixos.org/t/prs-ready-for-review/3032/5785

alyssais · 2025-08-20T18:12:28Z

I really don't think this is ready for review… It's blocked by a change request, and we're still waiting to find out why it works.

RossComputerGuy · 2025-08-20T18:19:58Z

I thought we already knew the why:

Funky splicing of buildPackages vs pkgs which causes useLLVM = false at some point, making libc++ not used
libunwind was being provided by lib.optional () [] which causes it to not propagate

alyssais · 2025-08-20T18:34:13Z

I thought we already knew the why:

Well last time I asked, I got "Idk how to explain why, it just does.", and the discussion since didn't seem particularly conclusive, still none of that information is in the commit message, and Emily's further probing on Matrix from earlier is still unresponded to, so I'm not sure how the reviewability is supposed to have changed here.

RossComputerGuy · 2025-08-21T01:41:25Z

Emily's further probing on Matrix from earlier is still unresponded to

When was that? It might've been when I was at work or asleep.

RossComputerGuy · 2025-08-21T01:45:11Z

still none of that information is in the commit message,

Updated the commit message

alyssais · 2025-08-21T10:55:15Z

Emily's further probing on Matrix from earlier is still unresponded to

When was that? It might've been when I was at work or asleep.

https://matrix.to/#/!kxOJEqURGkuOHTRRQB:matrix.org/$jfJfEVIbG_BXOxjKQw4CMiQfwVWRVdrZrGkt_CVnCjk?via=matrix.org&via=nixos.dev&via=tchncs.de (it pinged you)

emilazy · 2025-08-21T11:09:03Z

We add llvmPackages.libunwind to buildInputs depending on stdenv.targetPlatform.isLLVM. The libgcc_s hack aside, I don’t see how that can be correct – it should surely be either stdenv.hostPlatform.isLLVM or be a libunwind for the target platform.

Other than that, it seems like this is likely downstream of either or both of issues wiring up llvmPackages, and the Rust derivation’s confusion about cross. pkgsLLVM.buildPackages.hostPackages doesn’t exist, but assuming we’re talking about targetPackages instead, perhaps these lines are the culprit?

      ccForBuild = ccPrefixForStdenv pkgsBuildBuild.targetPackages.stdenv;
      cxxForBuild = cxxPrefixForStdenv pkgsBuildBuild.targetPackages.stdenv;
      ccForHost = ccPrefixForStdenv pkgsBuildHost.targetPackages.stdenv;
      cxxForHost = cxxPrefixForStdenv pkgsBuildHost.targetPackages.stdenv;
      ccForTarget = ccPrefixForStdenv pkgsBuildTarget.targetPackages.stdenv;
      cxxForTarget = cxxPrefixForStdenv pkgsBuildTarget.targetPackages.stdenv;

RossComputerGuy · 2025-08-21T13:56:26Z

Emily's further probing on Matrix from earlier is still unresponded to

When was that? It might've been when I was at work or asleep.

https://matrix.to/#/!kxOJEqURGkuOHTRRQB:matrix.org/$jfJfEVIbG_BXOxjKQw4CMiQfwVWRVdrZrGkt_CVnCjk?via=matrix.org&via=nixos.dev&via=tchncs.de (it pinged you)

GOException: no routes for location: im.fluffychat://chat/! kxOJEqURGkuOHTRRQB:matrix.org

RossComputerGuy · 2025-08-24T02:28:41Z

Rebased messed up

RossComputerGuy · 2025-08-25T05:34:25Z

Managed to do some testing and the funky llvmPackages magic we had is no longer needed afaict.

Fixes `libunwind` from not being included due to improper usage of `lib.optional` by passing a list instead of a single item. Drops the hack for `llvmPackages` since just passing it through works now.

alyssais

Great to see the hacks going away, thank you!

RossComputerGuy mentioned this pull request Jul 31, 2025

Build failure: rustc on LLVM stdenv #311930

Open

nix-owners bot requested review from winterqt, Mic92, figsoda, zowoq and alyssais July 31, 2025 04:14

ccicnce113424 mentioned this pull request Jul 31, 2025

linux_cachyos-lto: adopt nixpkgs#402198 (wip) chaotic-cx/nyx#1115

Draft

ccicnce113424 suggested changes Jul 31, 2025

View reviewed changes

alyssais reviewed Jul 31, 2025

View reviewed changes

pkgs/development/compilers/rust/rustc.nix Outdated Show resolved Hide resolved

RossComputerGuy force-pushed the fix/rustc-llvm branch from 22cc1a3 to 530378f Compare July 31, 2025 16:10

ccicnce113424 approved these changes Jul 31, 2025

View reviewed changes

nixpkgs-ci bot added the 12.approvals: 1 This PR was reviewed and approved by one person. label Jul 31, 2025

RossComputerGuy mentioned this pull request Jul 31, 2025

llvmPackages_{20,21,git}.compiler-rt: no fmv on aarch64 without libc #409265

Open

13 tasks

ccicnce113424 mentioned this pull request Aug 1, 2025

pkgsLLVM.linuxPackages: fix kernel modules building by setting makeFlags #430035

Open

13 tasks

alyssais reviewed Aug 4, 2025

View reviewed changes

RossComputerGuy force-pushed the fix/rustc-llvm branch from 530378f to 42ddf94 Compare August 14, 2025 15:03

alyssais requested changes Aug 14, 2025

View reviewed changes

RossComputerGuy force-pushed the fix/rustc-llvm branch from 42ddf94 to 2691fcc Compare August 15, 2025 02:03

ccicnce113424 reviewed Aug 16, 2025

View reviewed changes

RossComputerGuy force-pushed the fix/rustc-llvm branch from 2691fcc to 94540df Compare August 21, 2025 01:44

wegank added the 2.status: merge conflict This PR has merge conflicts with the target branch label Aug 23, 2025

RossComputerGuy closed this Aug 24, 2025

RossComputerGuy force-pushed the fix/rustc-llvm branch from 94540df to b300c0a Compare August 24, 2025 02:27

RossComputerGuy reopened this Aug 24, 2025

nixpkgs-ci bot removed the 2.status: merge conflict This PR has merge conflicts with the target branch label Aug 24, 2025

RossComputerGuy force-pushed the fix/rustc-llvm branch from c0f3aec to 7a4a629 Compare August 25, 2025 05:33

RossComputerGuy force-pushed the fix/rustc-llvm branch from 7a4a629 to 6119905 Compare August 25, 2025 06:02

RossComputerGuy requested a review from alyssais August 25, 2025 06:03

emilazy approved these changes Aug 25, 2025

View reviewed changes

rustc: fix pkgsLLVM

37c2603

Fixes `libunwind` from not being included due to improper usage of `lib.optional` by passing a list instead of a single item. Drops the hack for `llvmPackages` since just passing it through works now.

RossComputerGuy force-pushed the fix/rustc-llvm branch from 6119905 to 37c2603 Compare August 25, 2025 06:06

nixpkgs-ci bot added 12.approvals: 2 This PR was reviewed and approved by two persons. and removed 12.approvals: 1 This PR was reviewed and approved by one person. labels Aug 25, 2025

alyssais approved these changes Aug 27, 2025

View reviewed changes

nixpkgs-ci bot added 12.approvals: 3+ This PR was reviewed and approved by three or more persons. and removed 12.approvals: 2 This PR was reviewed and approved by two persons. labels Aug 27, 2025

Uh oh!

rustc: fix pkgsLLVM #429745

Are you sure you want to change the base?

rustc: fix pkgsLLVM #429745

Conversation

RossComputerGuy commented Jul 31, 2025

Things done

Uh oh!

ccicnce113424 commented Jul 31, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

blitz commented Aug 13, 2025

Uh oh!

alyssais left a comment

Choose a reason for hiding this comment

Uh oh!

RossComputerGuy commented Aug 15, 2025

Uh oh!

alyssais commented Aug 15, 2025

Uh oh!

RossComputerGuy commented Aug 15, 2025

Uh oh!

emilazy commented Aug 15, 2025

Uh oh!

RossComputerGuy commented Aug 15, 2025

Uh oh!

emilazy commented Aug 15, 2025

Uh oh!

RossComputerGuy commented Aug 15, 2025

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

nixos-discourse commented Aug 20, 2025

Uh oh!

alyssais commented Aug 20, 2025

Uh oh!

RossComputerGuy commented Aug 20, 2025

Uh oh!

alyssais commented Aug 20, 2025

Uh oh!

RossComputerGuy commented Aug 21, 2025

Uh oh!

RossComputerGuy commented Aug 21, 2025

Uh oh!

alyssais commented Aug 21, 2025

Uh oh!

emilazy commented Aug 21, 2025

Uh oh!

RossComputerGuy commented Aug 21, 2025

Uh oh!

RossComputerGuy commented Aug 24, 2025

Uh oh!

RossComputerGuy commented Aug 25, 2025

Uh oh!

alyssais left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

ccicnce113424 commented Jul 31, 2025 •

edited

Loading