| 23 Nov 2025 |
 Tristan Ross | However, I see Arm having a higher tier support first before RISC-V | 22:18:42 |
| Tristan Ross | The problem is the HW is just now starting to get better performance. Like the ESWIN EIC7700X is significantly better than the JH7110 but still quite lacking. | 22:19:26 |
 kwinz | sorry for the beginner questionL to summarize i have those options right now to do a nixos-rebuild:
- build 25.05 or staging myself which takes probably 3 days to finish
- cross compile staging or 25.05
- wait until the PIE fix for go from staging is merged into unstable and ztier starts caching it.
| 22:19:34 |
| kwinz | * sorry for the beginner question: to summarize i have those options right now to do a nixos-rebuild:
- build 25.05 or staging myself which takes probably 3 days to finish
- cross compile staging or 25.05
- wait until the PIE fix for go from staging is merged into unstable and ztier starts caching it.
| 22:19:45 |
| Tristan Ross | I think 2 is the safest option you have | 22:19:55 |
| Tristan Ross | I do not recommend native compilation unless you actually need it | 22:20:07 |
| kwinz | appreciate it! | 22:20:28 |
| kwinz | HiFive Premier P550 is probably twice as fast as the JH7110, but nothing ground breaking it seems | 22:22:32 |
| Tristan Ross | Actually, the Sifive P550 cores are actually pretty good imo. It's nothing to scoff at. It's actually got some nice features like branch prediction and out of order execution. | 22:23:58 |
| Tristan Ross | It's just when you compare it to non RISC-V cores, it's not as exciting in terms of performance. | 22:24:29 |
| 24 Nov 2025 |
 Alex | In reply to @rosscomputerguy:matrix.org
I think 2 is the safest option you have Depends on what you have installed.
For a bare system, it works fine, but various random things just don't work in cross. | 00:55:17 |
| Tristan Ross | That is true, but I've also ran into various things being broken natively | 00:56:04 |
 dramforever | In reply to @dramforever:matrix.org
it will be fixed when this lands https://github.com/NixOS/nixpkgs/pull/458867 landed | 04:35:35 |
| dramforever | this is in nixos-unstable | 04:35:45 |
| Tristan Ross | Oh sweet | 04:36:07 |
| Tristan Ross | I couldn't use nixpk.gs to check since 504 so I decided to give it a shot like 5 minutes ago | 04:36:42 |
| Tristan Ross | Having to wait for a bunch of other builds | 04:36:50 |
| Tristan Ross | At least I get to put the 512GB of RAM I have to good use | 04:37:01 |
| Tristan Ross | Oh yes, that indeed is working. Thank you for fixing that. | 04:38:12 |
| dramforever | i literally did nothing lol | 04:38:27 |
| Tristan Ross | Oh lol, I misremembered | 04:38:52 |
| Tristan Ross | I even looked at the PR like 2 seconds ago... 🤦♂️ | 04:39:08 |
| dramforever | anyway i'm going to try ubuntu riscv64 and see | 04:39:36 |
| Tristan Ross | Alright | 04:39:54 |
| dramforever | like how on earth they're not running into the base address problem for PDEs | 04:40:09 |
| Tristan Ross | Idk, it somehow works for them lol | 04:41:12 |
| 25 Nov 2025 |
| Tristan Ross | EFI stub: Booting Linux Kernel...
EFI stub: EFI_RNG_PROTOCOL unavailable
EFI stub: Loaded initrd from LINUX_EFI_INITRD_MEDIA_GUID device path
EFI stub: Using DTB from configuration table
EFI stub: Exiting boot services...
[ 0.000000] OF: fdt: Reserved memory: failed to reserve memory for node 'lpcpures@DFFE0000': base 0x00000000dffe0000, size 0 MiB
[ 0.000000] Kernel panic - not syncing: RISC-V system with no 'timebase-frequency' in DTS
[ 0.000000] CPU: 0 PID: 0 Comm: swapper/0 Not tainted 6.6.18 #1-NixOS
[ 0.000000] Hardware name: Deepcomputing FML13V03 (DT)
[ 0.000000] Call Trace:
[ 0.000000] [<ffffffff8000610e>] dump_backtrace+0x28/0x30
[ 0.000000] [<ffffffff81115712>] show_stack+0x38/0x44
[ 0.000000] [<ffffffff81135d9a>] dump_stack_lvl+0x44/0x5c
[ 0.000000] [<ffffffff81135dca>] dump_stack+0x18/0x20
[ 0.000000] [<ffffffff81115a7a>] panic+0x120/0x30e
[ 0.000000] [<ffffffff812056be>] time_init+0x44/0xc8
[ 0.000000] [<ffffffff812011fa>] start_kernel+0x842/0xb40
[ 0.000000] ---[ end Kernel panic - not syncing: RISC-V system with no 'timebase-frequency' in DTS ]---
Strange error while booting this kernel: https://github.com/DC-DeepComputing/fml13v03_linux
| 06:29:34 |
| Tristan Ross | I might've forgot to set the device tree | 06:44:25 |
| Tristan Ross | Or the problem is I am using the older kernel branch | 06:44:35 |
| Tristan Ross | Or the problem is both of those things | 06:44:42 |
