| 1 Jun 2025 |
 emily | I see | 18:50:50 |
| emily | we don't have any of those building on Hydra, right? I guess I'll just try and figure out how to make repart not slow | 18:51:14 |
| emily | oh we have an sd_image | 18:51:44 |
 K900 | We do | 18:51:46 |
| K900 | https://hydra.nixos.org/job/nixos/trunk-combined/nixos.sd_image_new_kernel_no_zfs.aarch64-linux | 18:51:47 |
| emily | cool | 18:51:51 |
| K900 | sd_image will have 6.12 which is too old | 18:52:03 |
| K900 | To actually do anything useful at least | 18:52:20 |
| emily | right | 18:52:23 |
| K900 | It might boot to a serial console, I haven't checked | 18:52:30 |
| emily | what a pain | 18:53:03 |
| K900 | You could theoretically netboot it | 18:53:46 |
| K900 | If you want to have even more fun | 18:53:51 |
| emily | well if I want to do fun things I should probably figure out why systemd-repart takes over an hour to build an image. | 18:54:29 |
 @misuzu:matrix.org | Just read about bpi-r4 pro and now I want one | 19:22:28 |
| emily | the pro doesn't add anything I'd want I think thankfully | 19:37:35 |
| emily | I'd be more interested in a new WiFi chip or case they have actual shielding | 19:37:57 |
| emily | *that | 19:38:00 |
| @misuzu:matrix.org | I want those 2.5G ports | 19:38:31 |
| emily | USB-C UART is nice... but the 2.5G LAN isn't super relevant to me. admittedly including the 2.5G WAN alongside the 10G would solve my WAN SFP+ woes | 19:38:53 |
 matthewcroughan | emily: did you get the fan working? | 20:44:15 |
| emily | haven't really tried, only booted it into NixOS yesterday. does yours not work? | 20:44:31 |
| matthewcroughan | Never did, there's an overlay in openwrt that I was gonna play around with maybe | 20:44:46 |
| matthewcroughan | Are you using binfmt? | 20:45:36 |
| emily | yes but I specifically made sure that systemd-repart is not | 20:45:49 |
| emily | so no | 20:45:52 |
| matthewcroughan | Yeah weird, definitely isn't like that for me | 20:46:04 |
| matthewcroughan | is your storage slow maybe? | 20:46:09 |
| emily | it's in the macOS builder VM. IO is slow but not that slow. it's probably fakeroot. | 20:57:36 |
| matthewcroughan | macOS stuff is always buggy and slow | 21:06:16 |
