| 9 Dec 2021 |
 cleverca22 | perhaps have a dedicated binutils-vc4.nix, that doesnt get updated with new patches | 18:03:03 |
| cleverca22 | how are other targets with binutils forks handling this? | 18:03:34 |
| cleverca22 | Mic92: i can either do an ugly [ constants ] ++ optional ++ [ constants ] change that will avoid a mass-rebuild of x86
or i can do a pretty [ constants ] ++ optional that will change the order of x86 patches and cause a rebuild | 18:16:06 |
| cleverca22 | ah, every patch after the conflicting one also conflicts | 18:22:42 |
 Mic92 | cleverca22: I am not aware of forks | 18:22:45 |
| cleverca22 | so they are all optional | 18:22:46 |
| cleverca22 | testing the fix.... | 18:23:13 |
| cleverca22 | gcc has the same issues | 18:23:35 |
| cleverca22 | ../../source/gcc/graphite-dependences.c: In function 'isl_map* constrain_domain(isl_map*, isl_set*)':
../../source/gcc/graphite-dependences.c:48:16: error: 'isl_space_get_tuple_id' was not declared in this scope; did you mean 'isl_map_get_tuple_id'?
48 | isl_id *id = isl_space_get_tuple_id (d, isl_dim_in);
| 18:29:47 |
| cleverca22 | something got renamed, causing build failures | 18:29:55 |
| cleverca22 | but no patches touched that file, and the symbol is from outside the file | 18:47:33 |
| cleverca22 | * but no patches touched that file, and the symbol is from outside gcc | 18:47:37 |
| Mic92 | cleverca22: you might need a different version of isl | 20:31:07 |
| Mic92 | isl is a compiler dependency | 20:31:15 |
| cleverca22 | ah, i was trying to figure out what pkg it came from | 20:31:24 |
| cleverca22 | then got distracted | 20:31:27 |
| Mic92 | Alyssa Ross: is pkgsCross.netbsd-llvm broken an known issue? https://hercules-ci.com/github/nix-community/cross-toolchains.nix/jobs/8 | 21:40:15 |
 Alyssa Ross | mic92: not known by me, but I don't think I've ever used netbsd-llvm — I think John added it | 21:43:33 |
| Mic92 | John Ericson: ^ | 21:45:26 |
 John Ericson | yeah i didn't quite finish getting it it work | 21:46:27 |
| John Ericson | there is some -lgcc I couldn't find the origin of | 21:46:40 |
 sterni | this isn't the proper thing anyways as NetBSD upstream use gcc | 21:55:12 |
| Mic92 | well. Let's drop stuff that never really worked, so we can keep an eye on regressions in other areas | 21:57:51 |
| John Ericson | Well I would rather ensure it still evals | 21:59:17 |
| John Ericson | netbsd is supposed to have support for both | 21:59:29 |
| John Ericson | it has clang things in it's makesfiles a plenty | 21:59:38 |
| John Ericson | I just don't think anyone might be building it entirely from scratch with clang yet | 21:59:54 |
| Mic92 | I don't think anyone checks if platforms in this file evals | 22:02:13 |
| Mic92 | I fixedd a couple of platforms in their today | 22:02:23 |
|  ris_ joined the room. | 23:04:22 |
