NixOS RISC-V | 241 Members | |
| NixOS on RISC-V https://wiki.nixos.org/wiki/RISC-V https://pad.lassul.us/NixOS-riscv64-linux | 73 Servers |
| Sender | Message | Time |
|---|---|---|
| 5 Jan 2025 | ||
In reply to @rosscomputerguy:matrix.orgNB: this doesn't mean it's bootable. | 07:53:46 | |
In reply to @alex:tunstall.xyzI know, I'll need to pull in the Starfive OpenSBI and u-boot according to the forums. | 14:42:25 | |
| It's at least progress, when I tried this in the past it wasn't even showing up in lsblk. | 14:43:00 | |
| Probably because the kernel you'd built at the time didn't have the patches needed for PCIe. | 16:15:32 | |
| Yeah, I'm currently trying to update to the VisionFive2 v5.13.1 software | 16:42:17 | |
| https://github.com/starfive-tech/VisionFive2/releases/tag/JH7110_VF2_515_v5.13.1 | 16:42:22 | |
| It seems the v3.x.x set fixed a few of the issues for NVMEs | 16:43:35 | |
| But it'll probably be better to work with the latest stuff | 16:43:52 | |
Huh, the files u-boot is trying to install don't exist | 16:53:13 | |
| Oh, the nvme doesn't show up if your USB C power supply isn't capable enough lol. | 19:23:43 | |
| I would love to have time to work on RISC-V stuff again, but that probably only happens if someone needs professional help :( | 22:39:20 | |
| Damn my config is too big to eval on the VF2, I had to make a 32GB swap file and it's starting to make more progress. | 23:50:25 | |
| Are you evaluating multiple versions of Nixpkgs? Needing more than 4 GiB for NixOS evaluation is very unusual... | 23:51:28 | |
| 6 Jan 2025 | ||
| I don't think so | 00:04:10 | |
| https://github.com/RossComputerGuy/dotfiles/blob/master/flake.nix#L213-L215 | 00:04:43 | |
| Dang, 30 - 40 minutes to eval until I got a GHC error | 00:21:47 | |
In reply to @rosscomputerguy:matrix.orgUsing nix flake metadata, I count 5 versions of Nixpkgs amongst your inputs (2 of which are coincidentally the same). | 00:26:37 | |
Does that matter when you use overlays only and don't use packages or legacyPackages of the inputs? | 00:29:27 | |
| I don't think so. Unfortunately I have no idea how to profile Nix evaluations... I've always been able to just throw more RAM at the problem (since it's only been a problem in VMs for me so far). | 00:30:35 | |
| Yeah, I'm not sure how to profile evals too | 00:31:06 | |
| It would really help if I knew how to for some of my other projects lol | 00:31:27 | |
| I guess the old binary search trick could work... Remove half of the code and see how it affects memory usage. Then focus your search onto the more memory-consuming half and repeat. | 00:35:54 | |
| https://github.com/RossComputerGuy/dotfiles/runs/35174538512 huh the x86_64-linux cross is able to evail within Garnix's 5GB limit but the full system cannot | 00:37:51 | |
| * https://github.com/RossComputerGuy/dotfiles/runs/35174538512 huh the x86_64-linux cross is able to eval within Garnix's 5GB limit but the full system cannot | 00:37:57 | |
In reply to @alex:tunstall.xyzMemory usage is one of the major pain points of Nix. 😞 | 01:03:36 | |
I cannot find where this etc file is defined lol | 02:02:49 | |
| Oh, it's inside the derivation | 02:04:15 | |
| Webkit requires GHC somewhere and it throws that error, I know there's a way to get around that but just trying to get the system to build purely | 02:05:09 | |
| Oh, shellcheck is required which uses GHC | 02:08:01 | |
| * Oh, shellcheck is required which uses GHC deep in the chain | 02:08:13 | |