| 4 Sep 2021 |
 kranzes | In reply to @kranzes:matrix.org
Once you get the correct repo.json in robotnix, you should remove the redundant device,kernel,vendor trees that i already include as inputs in my flake or you could remove the source.dirs and inputs from the flake and just use the device,kernel,vendor trees from the repo.json | 20:19:13 |
| kranzes | It will take you much less time to build it than what it will take for me | 20:19:37 |
| kranzes | 3970x (not sure how much ram) + ssd (i assume) VS 2700x, 16gb of ram and an HDD | 20:20:24 |
 danielrf | My build times are usually around 30-40 min typically | 20:21:51 |
| danielrf | they just slightly heat up my office room :) | 20:22:20 |
| kranzes | In reply to @danielrf:matrix.org
My build times are usually around 30-40 min typically not including chromium/bromite? | 20:22:48 |
| danielrf | Yep, just android, not including chromium | 20:23:01 |
| kranzes | I see | 20:23:06 |
| danielrf | that's another approx 30-40 min IIRC | 20:23:06 |
| kranzes | so it should be similar because we are using the prebuilt one i tink | 20:23:22 |
| kranzes | expected it to be maybe faster | 20:23:32 |
| kranzes | nah it makes sense nvm | 20:23:43 |
| danielrf | Rebuilds for me typically don't need to build chromium since it's built in another derivation, so we can use a cached one from /nix/store | 20:24:31 |
| kranzes | would it be possible to split the build of the rom into multiple drvs? | 20:26:02 |
| danielrf | Not easily--I've gone through a few attempts to do that | 20:26:35 |
| kranzes | Btw do you use ccache? | 20:26:53 |
| kranzes | i never got around to use it because i couldnt figure out if it actually works or what id does | 20:27:06 |
| danielrf | No, not normally, since my machine is quick enough. It should work, although its use is deprecated by upstream AOSP | 20:27:49 |
| kranzes | Does the mk_repo_file.py script already have the functionality to generate the repo.json from that manifest i sent you? or will you have to create/modify that now to make it work? | 20:32:50 |
| kranzes | * Does the mk_repo_file.py script already have the functionality to generate the repo.json from that manifest i sent you or will you have to create/modify that now to make it work? | 20:33:00 |
| danielrf | Yeah, that's what i'm working through now. I put this manifest file into freshly created local git-repo as default.xml and I'm trying to run mk_repo_file.py on it | 20:34:05 |
| danielrf | The manifest xml file is requiring a couple small changes... For some reason it didn't include the sairam1411 (and a couple others) repos as a <remote> in the xml | 20:35:08 |
| danielrf | even though it had those repos as <project> xml tags | 20:35:34 |
| kranzes | weird that it didnt include their remote | 20:38:40 |
| danielrf | kranzes: There's been quite a few source changes overnight: https://github.com/ForkLineageOS/android/commits/lineage-18.1 | 20:44:29 |
| kranzes | included them already | 20:44:46 |
| kranzes | I think so | 20:45:01 |
| kranzes | Yeah I did | 20:45:54 |
| danielrf | OK, I saw a bunch of differences in URLs between the new repo.json file I'm generating and the previous one we used, and I initially thought we had previously missed a bunch of repos | 20:45:56 |
| kranzes | In reply to @danielrf:matrix.org
kranzes: There's been quite a few source changes overnight: https://github.com/ForkLineageOS/android/commits/lineage-18.1 You can see that they are in fact included in the miatoll.xml i sent earlier | 20:46:30 |
