| 21 Apr 2025 |
 emily | which was approved but failed to make complete contact with reality | 21:59:59 |
| emily | which is something like x86_64-linux Tier 1, aarch64-linux Tier 1.5, aarch64-darwin Tier 2, x86_64-darwin Tier 2.5, i686-linux Tier 3, everything else tier nothing | 22:00:48 |
| emily | RFCs are immutable so the appendix will never be updated, but in principle we have accepted the tiers themselves | 22:01:15 |
| emily | and they have quite detailed descriptions | 22:01:21 |
| emily | in practice the biggest tier distinction is just "used on NixOS infra" vs. "not used on NixOS infra" | 22:01:54 |
| emily | there are x86_64-linux, aarch64-linux, and aarch64-darwin machines in the infra so there's no choice but to keep those platforms in good shape. x86_64-darwin comes along for the ride on aarch64-darwin for the most part | 22:02:33 |
| emily | and a subset of i686-linux is required to build e.g. NixOS tests at present, so that is also kept in working order | 22:02:47 |
| emily | everything else is pretty much if it breaks you can keep both pieces | 22:02:53 |
 Tristan Ross | Gotcha, I was thinking it'd be easier if it were in the manual. It'd be mutable then. | 22:10:01 |
| emily | well, it basically is, they're just described more vaguely :) | 22:12:13 |
| emily | but yes if we want to assign actual numbers to the platforms then the manual would be the place | 22:12:27 |
| emily | the .5s sort of make it awkward to describe reality in more precise terms than is already there though | 22:12:42 |
| emily | probably Tier 1 x86_64-linux, Tier 2 aarch64-linux aarch64-darwin x86_64-darwin would be adequate | 22:14:04 |
| emily | x86_64-darwin is sort of tier 3 in practice but "No channel-blocking jobs on Hydra." isn't true | 22:14:30 |
| Tristan Ross | I'd say aarch64-linux is tier 1 by now lol | 22:23:29 |
 Randy Eckenrode | When it comes down to it, aarch64-linux is not going to be allowed to block like x86_64-linux is. | 22:28:47 |
| emily | yeah it definitely does not meet the criteria outlined for tier 1, but it is "tier 1.5"-ish | 22:30:27 |
| emily | mostly insofar as it piggy-backs on x86_64-linux though, in teh same way that x86_64-darwin does on aarch64-darwin | 22:30:40 |
| emily | * mostly insofar as it piggy-backs on x86_64-linux though, in the same way that x86_64-darwin does on aarch64-darwin | 22:30:45 |
| Tristan Ross | In reply to @reckenrode:matrix.org
When it comes down to it, aarch64-linux is not going to be allowed to block like x86_64-linux is. Why? | 22:57:07 |
| Randy Eckenrode | I can’t see x86_64-linux users accepting they have to wait on it. | 22:57:46 |
| Tristan Ross | In reply to @reckenrode:matrix.org
I can’t see x86_64-linux users accepting they have to wait on it. Well, either way I don't think it'd be fair. | 22:58:15 |
| emily | Tier 1 says "Developer/user base: most of the Nix developers/users", which is just not true of aarch64-linux | 23:15:08 |
| emily | admittedly that's in the non-normative section | 23:15:27 |
| emily | "full ofBorg support" is not true of aarch64-linux right now either, though. | 23:15:41 |
| Tristan Ross | In reply to @emilazy:matrix.org
"full ofBorg support" is not true of aarch64-linux right now either, though. What's lacking for Ofborg? | 23:33:25 |
| Tristan Ross | In reply to @emilazy:matrix.org
Tier 1 says "Developer/user base: most of the Nix developers/users", which is just not true of aarch64-linux Do we have an idea of what the x86 vs aarch64 split looks like? | 23:33:49 |
| emily | probably >90% / <10% | 23:34:08 |
| emily | I don't think ofborg has any AArch64 machines (because they are very expensive) | 23:34:23 |
| emily | well, other than Macs of course | 23:34:34 |
