| 24 May 2024 |
 Lily Foster | but | 11:13:52 |
| Lily Foster | when tests were moved from meta.tests to passthru.tests before that PR was merged | 11:14:14 |
| Lily Foster | the check-meta check was not updated | 11:14:20 |
| Lily Foster | so it's not useful right now | 11:14:27 |
| Lily Foster | but checking it there-ish would be ideal | 11:14:49 |
| Lily Foster | (and then abort-ish eval failures (e.g. syntax errors) vs throw-ish eval failures would also be differentiated, as intended by most eval checks) | 11:15:52 |
| Lily Foster | * (and then abort-ish eval failures (e.g. syntax errors) vs throw-ish eval failures (e.g. not available on platform) would also be differentiated, as intended by most eval checks) | 11:16:18 |
| 27 May 2024 |
 vcunat | Any idea if the eval label mechanism is broken? It got completely wrong in this case: https://github.com/NixOS/nixpkgs/pull/314255#issuecomment-2132781439 | 07:06:12 |
| vcunat | (also in the original PR from which this backport was created) | 07:07:28 |
| vcunat | Or do the linux rebuild labels not take aarch64-linux into account? | 07:07:58 |
 hexa | vcunat: https://github.com/NixOS/ofborg/blob/1d74f577a10ca1adbc933bdfb966d380aa343eb1/ofborg/src/tagger.rs#L327 🙂 | 12:47:59 |
| hexa | aarch64 is not represented in there | 12:48:53 |
| hexa | * aarch64 is not represented in there, not for linux, not for darwin | 12:49:06 |
| vcunat | Maybe we could do an average over the pair of platforms? | 12:49:21 |
| hexa | or create two more sets of labels 🤷 | 12:49:44 |
| vcunat | For darwin we even build them on the same set of machines. | 12:50:44 |
| vcunat | So it probably doesn't make much sense to differentiate if the work will be for x86_64-darwin or aarch64-darwin. | 12:51:19 |
| hexa | if we information is available we might a well expose it | 12:52:15 |
| hexa | * if the information is available we might a well expose it | 12:52:18 |
| hexa | the downside is probably that the code does not look too maintainble 😄 | 12:52:40 |
| hexa | * the downside is probably that the code does not look too maintainable 😄 | 12:52:49 |
| vcunat | I certainly see the rebuild list for all platforms in the "eval" link. | 13:23:24 |
| vcunat | * I certainly see the rebuild list for all four platforms in the "eval" link. | 13:23:32 |
| Lily Foster | if y'all decide on something, i can make a quick PR to update labelling | 13:23:47 |
| Lily Foster | * if y'all decide on something, i can make a quick PR to ofborg to update labelling | 13:25:04 |
| hexa | In reply to @vcunat:matrix.org
Maybe we could do an average over the pair of platforms? the average between two platforms might very well be misleading, e.g. 0 and 1000 -> 500, 0 and 3000 -> 1500 | 13:29:18 |
| hexa | and you'd always have to dig into the list to get the accurate picture, and the list does not necessarily load/render very fast 🙂 | 13:30:00 |
| vcunat | Intuitively I'd want just a single number that estimates the amount of work for Hydra (say, even hours needed if doing nothing else). | 13:30:42 |
| vcunat | But perhaps there are more use cases for this. | 13:31:10 |
| hexa | so what would be a good number you could derive from two linux platforms for example? | 13:32:36 |
