| 12 Jul 2023 |
 misuzu | In reply to @thefossguy:matrix.org
I noticed that the image (https://github.com/misuzu/nixos-vf2) didn't boot with all combinations of DIP switches so me being extra smart, I dd-ed uboot SPL and SBI firmware on /dev/sda1 and /dev/sda2 respectively and mvoed /boot to /dev/sda3 and / to /dev/sda4. Now, your derivation doesn't work because it expects "/dev/sda1" to be EFI? https://github.com/misuzu/nixos-vf2/tree/master#u-boot-on-an-sd-card follow this if you want u-boot on an sdcard | 19:29:40 |
 Alex | In reply to @misuzu:matrix.org
https://github.com/misuzu/nixos-vf2/tree/master#u-boot-on-an-sd-card follow this if you want u-boot on an sdcard Oh neat, that means I can cancel building my dodgy attempt at tweaking your make-efi-image script. | 19:30:52 |
| misuzu | In reply to @alex:tunstall.xyz
Oh neat, that means I can cancel building my dodgy attempt at tweaking your make-efi-image script. Try it on spare sdcard first, maybe I've written something wrong 🤡 | 19:34:08 |
| misuzu | The image uses labels to mount partitions, so partition numbers shouldn't matter | 19:36:37 |
| Alex | In reply to @misuzu:matrix.org
The image uses labels to mount partitions, so partition numbers shouldn't matter Doesn't that mean I could shrink my root partition a bit and put u-boot at the end of the disk? | 19:37:29 |
| misuzu | In reply to @alex:tunstall.xyz
Doesn't that mean I could shrink my root partition a bit and put u-boot at the end of the disk? Root partition - no, efi partition - maybe | 19:38:15 |
 raitobezarius | In reply to @0x4a6f:matrix.org
Not enough funds, maybe NixOS foundation could chim in. Maybe raitobezarius has some better understanding of this platform. I'd wait | 23:39:14 |
| raitobezarius | As said in another channel | 23:39:24 |
| raitobezarius | The speed of innovation in RISC V is sufficiently fast that by the time you get your MilkV, you may have to wait ~6-8 mo before you get the next generation which will probably much better at that moment | 23:40:07 |
| 13 Jul 2023 |
 Pratham Patel | In reply to @misuzu:matrix.org
It's not enough to flash u-boot to reset u-boot settings Okay yeah, that worked. I thought reflashing the firmware would do the trick but now that I think of it, it only makes sense to keep the data separate :D | 04:04:03 |
| Pratham Patel | In reply to @alex:tunstall.xyz
Doesn't that mean I could shrink my root partition a bit and put u-boot at the end of the disk? If you mean the SPL or the flat binary, then no. | 04:05:45 |
| Pratham Patel | In reply to @alex:tunstall.xyz
I'm currently compiling with a few patches applied to try to fix the compilation problems without disabling DRM, since I'd really like to get working HDMI on my VF2.
I'll share the configuration if I'm able to get it working. Given the similarities between the two SBCs (and their respective Linux fork), it might work for the Star64 too. The serial console stops displaying messages after these:
EFI stub: Booting Linux Kernel...
EFI stub: Loaded initrd from command line option
EFI stub: Using DTB from configuration table
EFI stub: Exiting boot services...
clk u2_dw_i2c_clk_core already disabled
clk u2_dw_i2c_clk_apb already disabled
clk u5_dw_i2c_clk_core already disabled
clk u5_dw_i2c_clk_apb already disabled
the kernel doesn't even start...
| 04:22:48 |
 artemis | trofi: i still get the libtool error after cleaning the nix store as discussed and starting over with sandbox-fallback=false from the start | 08:55:46 |
 trofi | Just to clarify: you do use sandbox = true, right? | 08:56:21 |
| artemis | only if that is default | 08:56:36 |
| artemis | this is my entire nix.conf http://dpaste.com/5KQ988BL5 | 08:56:57 |
| trofi | nod, nix show-config should tell you. | 08:57:04 |
| artemis | sandbox = true
sandbox-build-dir = /build
sandbox-dev-shm-size = 50%
sandbox-fallback = false
sandbox-paths = /bin/sh=/usr/bin/busybox-nix-sandbox-shell /usr/bin/qemu-arm | 08:57:23 |
| artemis | i have not yet updated to your latest nix ebuild fwiw so i can do that if you think it's relevant | 09:00:15 |
| trofi | no, it looks fine. | 09:02:47 |
| trofi | is it a systemd or openrc system? i'll try to reproduce it in a cleaner VM | 09:03:07 |
| artemis | openrc | 09:03:32 |
| artemis | make.profile -> ../../var/db/repos/gentoo/profiles/default/linux/amd64/17.1/ | 09:04:04 |
| artemis | actually here is my whole `emerge --info` https://dpaste.com/22NWU74FE while i am at it | 09:05:52 |
| trofi | I'll try to reproduce locally and if I fail we'll try to compare the build logs for difference. | 09:15:05 |
| trofi | Built file locally for me: https://bpa.st/raw/SI6ZU. What log do you get with $ nix build --no-link github:NixOS/nixpkgs/9af373a61647257d16ae6062cddaa9094d24920c#pkgsCross.riscv64.libtool? | 10:11:29 |
| artemis | In reply to @trofi:matrix.org
Built file locally for me: https://bpa.st/raw/SI6ZU. What log do you get with $ nix build --no-link github:NixOS/nixpkgs/9af373a61647257d16ae6062cddaa9094d24920c#pkgsCross.riscv64.libtool? http://dpaste.com/6SMJ92MK9 | 20:11:15 |
|  zuckerberg changed their profile picture. | 22:03:45 |
| trofi | Ended up being a libtool deficiency (current unstable is less unstable): https://github.com/NixOS/nixpkgs/pull/167071#issuecomment-1634992962 | 22:24:18 |
|  sorear joined the room. | 23:48:33 |
