| 30 Nov 2025 |
 Grimmauld (any/all) | I mean, it might well be voodoo, git can be weird sometimes. My understanding is probably a bit off here. | 00:36:04 |
 dotlambda | Actually, it looks like the job didn't create new commits cause no commits were made to master in the meantime | 00:39:37 |
| 1 Dec 2025 |
 Fabián Heredia | thanks <3 | 01:59:21 |
| Fabián Heredia | 
Download image.png | 02:04:51 |
| Fabián Heredia | no rush but if someone with eval priviledges could start another one for gcc-15 I would appreciate it (cherry-picked a fix that should address about 40% of breakage)
| 02:04:52 |
 Sergei Zimmerman (xokdvium) | Kicked it | 02:47:53 |
| Fabián Heredia | Thanks <3 | 04:51:12 |
 ghpzin | @Fabián Heredia
I kind of missed it, but are the current plans to merged gcc15 into current staging ?
And if so where is the better place to push new fixes to, seeing as you cherry-pick them for jobset purposes ? | 04:51:32 |
| Fabián Heredia | Those that were cherry-picked are already in staging, the idea is to catch any big breakages before merging to staging (though I'm not sure if it is going to be before the next staging-next cycle, staging is rolling though an only gets restricted near releases) | 04:53:01 |
| Fabián Heredia | Oh, btw thanks for all the PRs/fixes for different packages ghpzin , really appreciate it I think this is mostly ready to be merged into staging. Was going over the list that Alyssa Ross and Yureka (she/her) | 05:17:18 |
| Fabián Heredia | * Oh, btw thanks for all the PRs/fixes for different packages ghpzin , really appreciate it I think this is mostly ready to be merged into staging. Was going over the list that Alyssa Ross and Yureka (she/her) made on the PR about breakages and only big-ish thing is berkleyDB | 05:19:01 |
| Fabián Heredia | * Oh, btw thanks for all the PRs/fixes for different packages ghpzin , really appreciate it I think this is mostly ready to be merged into staging. Was going over the list that Alyssa Ross and Yureka (she/her) made on the PR about breakages and only big-ish pending thing is berkleyDB | 05:19:14 |
 uep | classic, didn't that break on last gcc too? | 05:19:35 |
| Fabián Heredia | * Oh, btw thanks for all the PRs/fixes for different packages ghpzin , really appreciate it I think this is mostly ready to be merged into staging. Was going over the list that Alyssa Ross and Yureka (she/her) made on the PR about breakages and only big-ish pending thing is berkleyDB 5.3.z | 05:19:37 |
| Fabián Heredia | yeah, probably | 05:19:48 |
| ghpzin | Okay, I will try to create PRs for leftover fixes I have this week, they should cover most things.
Would be nice to have somebody who uses pkgsMusl to get involved, seeing as a lot of things break there. I know some required things, just not whether they are "correct to do".
Also there is a pretty scary thing printed in hydra tests and linux_5_10 breaks again with unresolved symbol filp_close after recent kernel update. | 05:22:26 |
| ghpzin | Okay, I will try to create PRs for leftover fixes I have this week, they should cover most things.
Would be nice to have somebody who uses pkgsMusl to get involved, seeing as a lot of things break there. I know some required things, just not whether they are "correct to do".
Also there is a pretty scary thing printed in hydra tests and linux_5_10 breaks again with unresolved symbol filp_close after recent kernel update (but that one was probably just a matter of time, seeing as it pops up every other kernel/gcc update). | 05:25:53 |
| Grimmauld (any/all) | is there a PR against which i should build? | 09:23:30 |
| Grimmauld (any/all) | i was building the fancy llvm-native stuff, but a lot of things are broken there. | 09:23:55 |
| ghpzin | You mean musl with gcc15 ? | 09:26:00 |
| Grimmauld (any/all) | yes | 09:26:08 |
| ghpzin | No PR, it starts pretty much breaking with just (changing default to 15):
https://github.com/NixOS/nixpkgs/pull/440456/commits/5264e09b60c020160780793fa0459b5448646efc | 09:26:51 |
| Grimmauld (any/all) | how far does it get on glibc? | 09:28:07 |
| ghpzin | I mean, at least it builds stdenv, which you can't say with pkgsMusl.stdenv | 09:28:56 |
| Grimmauld (any/all) | OH GOD | 09:29:04 |
| Grimmauld (any/all) | okay fine, i started an stdenv build | 09:30:20 |
| Grimmauld (any/all) | i have a suspicion this will suck | 09:31:04 |
| ghpzin | I can write up what I had fixes for including stdenv, gnumake, which and groff.
Iirc I stopped at patchutils that required update to one of versions in nixpkgs and backporting some huge commit to the other 2.
And generally I am not sure what even worked before on musl. | 09:38:55 |
| Grimmauld (any/all) | i got a full interactive graphical nixos session on musl | 09:39:25 |
| Grimmauld (any/all) | admittedly with a lot of patches and overlays | 09:39:41 |
