!xmLtiCaAJxfhURjrXl:matrix.org

NixOS RISC-V

236 Members
NixOS on RISC-V https://wiki.nixos.org/wiki/RISC-V https://pad.lassul.us/NixOS-riscv64-linux https://github.com/orgs/NixOS/teams/risc-v71 Servers

Load older messages


SenderMessageTime
17 Mar 2026
@alex:tunstall.xyzAlex I'm in the middle of building b0515cbf9eb44508f90c8cf5b39654dd9af16b80.
It is currently at 537 derivations completed with no failures so far, so perhaps this version is OK?
22:21:19
18 Mar 2026
@colemickens:matrix.orgcolemickens and it seems like native ghc is still not a thing. I'm getting libffi errors when I try to crossPkgs it? #error "libffi was configured for a RISC-V target but this does not appear to be a RISC-V compiler." :'( 01:47:06
@colemickens:matrix.orgcolemickensI left this stuff behind almost 3 years ago. I definitely was naive and assumed native ghc would be here by now01:47:22
@alex:tunstall.xyzAlex It is possible to go via x86_64 cross and, if you're lucky, pkgsCross.riscv64.haskell.compiler.ghc94 will be in Nixpkgs cache (otherwise you'll need to copy it). 01:48:58
@colemickens:matrix.orgcolemickensoh, right, I switched by testing apparatus to a flake that has an older nixpkgs. I might need to bump and try again.01:49:54
@alex:tunstall.xyzAlex

I'm getting libffi errors when I try to crossPkgs it?

Can you share what exactly you're doing? It should just work.

01:49:56
@colemickens:matrix.orgcolemickensoh actually this is on nixos-unstable, one second, let me still bump it anyway and then I'll show you if it still errors01:50:33
@colemickens:matrix.orgcolemickens okay I have this in a flake check attr and I'm just doing nix flake check in CI:
inputs.nixpkgs.legacyPackages.x86_64-linux.pkgsCross.riscv64.cachix.
01:52:32
@alex:tunstall.xyzAlex

Cross-compiling GHC and cross-compiling Haskell packages are two slightly different things.

[nixos@nixos:~]$ uname -a
Linux nixos 6.6.0 #1-NixOS SMP Tue Jan  1 00:00:00 UTC 1980 riscv64 GNU/Linux

[nixos@nixos:~]$ /nix/store/lcrs211p33jp89jm8a08m7d5vpi7533k-ghc-9.6.6/bin/ghc --version
The Glorious Glasgow Haskell Compilation System, version 9.6.6
01:54:12
@colemickens:matrix.orgcolemickensuh sorry I thought I sent a clarifying message, yes, I XY'd the problem a bit.01:55:07
@colemickens:matrix.orgcolemickensI'm actually after cachix. Sorry, important context.01:55:16
@colemickens:matrix.orgcolemickens* uh sorry I thought I sent a clarifying message (I definitely didn't from what I can see...), yes, I XY'd the problem a bit.01:55:38
@alex:tunstall.xyzAlex pkgsCross.riscv64.cachix evals for me on latest unstable. What commit of Nixpkgs is pinned in flake.lock? 01:56:06
@colemickens:matrix.orgcolemickens all the same, this is the derivation that fails: riscv64-unknown-linux-gnu-ghc-9.10.3 01:56:30
@colemickens:matrix.orgcolemickens I updated so it's latest nixos-unstable. I also confirmed by tracing the nodes in flake.lock and it reports 5b2c2d84341b2afb5647081c1386a80d7a8d8605 01:57:47
@alex:tunstall.xyzAlexDo you have a reproducer that only requires the Nix CLI?01:59:33
@colemickens:matrix.orgcolemickensI've just replaced my flake.{nix,lock}. Going to send it through the expensinve fat github runners to confirm output first, then I gist the flake.{nix,lock}02:08:46
@colemickens:matrix.orgcolemickensokay, it reprod: https://gist.github.com/colemickens/78af9a28c01b888726322ca6628b27f502:12:22
@alex:tunstall.xyzAlex Oh sorry, I thought you were talking about an error during eval not during realisation.
Yes, I can definitely reproduce it now.
02:24:15
@alex:tunstall.xyzAlex I've asked about it in #haskell:nixos.org.
I will see if I can fix it or at least track down the cause.
02:29:37
@jammie:matrix.orgJamieMageeI bumped into another issue while building OpenSSL https://github.com/NixOS/nixpkgs/pull/50106116:07:37
@dramforever:matrix.orgdramforeverwhat platform is this? are you building on qemu-user?16:32:17
@jammie:matrix.orgJamieMageeYes, it's on QEMU16:32:44
@dramforever:matrix.orgdramforeverthe binfmt thing, right?16:36:27
@dramforever:matrix.orgdramforeveri don't think i can in good conscience accept this. so many packages fail tests in qemu-user that we'll keep adding these, until the real riscv machine gets good then these would be obsolete16:40:52
@dramforever:matrix.orgdramforever * 16:41:04
@dramforever:matrix.orgdramforever as an example, a good while back even coreutils fails tests under qemu-user 16:41:20
@dramforever:matrix.orgdramforevera while back i wrote a hydra hack to schedule builds based on derivation meta16:43:15
@dramforever:matrix.orgdramforeverit still needs some weird extra hacks on top but it avoids changing hash16:43:32
@dramforever:matrix.orgdramforeverhttps://github.com/dramforever/hydra/tree/description-feature-hack16:44:16

Show newer messages


Back to Room ListRoom Version: 10