| 30 Jan 2024 |
 hexa | or just aborted the bisect | 00:01:23 |
 adamcstephens | Literally walked away from the computer | 00:02:39 |
| adamcstephens | It’s running in a tmux session | 00:02:58 |
| adamcstephens | i can see how that was confusing | 00:05:07 |
| adamcstephens | How’s the performance on the aarch64 build box? My VM on an M1 mini works well for a lot of stuff, but is kinda slow on these large rebuilds | 03:27:50 |
| hexa | ❯ git bisect log
git bisect start
# status: waiting for both good and bad commits
# good: [acd19036aca8b793f7f0099881d498c50b5ee147] sirius: fixup build with gcc13
git bisect good acd19036aca8b793f7f0099881d498c50b5ee147
# status: waiting for bad commit, 1 good commit known
# bad: [6876a3e7ac5b3d42d8061ca85fcd0723c3584298] Merge pull request #282045 from K900/cursed-qml-paths
git bisect bad 6876a3e7ac5b3d42d8061ca85fcd0723c3584298
# good: [8298d445bc8fca11a7d4ca0a2566bea600c0e28f] Merge pull request #280817 from NixOS/pkg/pythran
git bisect good 8298d445bc8fca11a7d4ca0a2566bea600c0e28f
# good: [9504a31fcd22ec75016efed8e8abe73fc0a4f065] Merge pull request #273494 from katexochen/grmon/init
git bisect good 9504a31fcd22ec75016efed8e8abe73fc0a4f065
# good: [f567f4ba89ee93c612f275bbbe50b12739a7e625] Merge pull request #281571 from michaelgrahamevans/forge-sparks
git bisect good f567f4ba89ee93c612f275bbbe50b12739a7e625
# good: [5da20659ca26d054b088cd11eac98f472ec061c9] Merge pull request #281401 from fabaff/uchecker
git bisect good 5da20659ca26d054b088cd11eac98f472ec061c9
# bad: [1fea4d70658b9b1ff831981b01f753741d932a35] Merge pull request #281945 from marsam/update-bundler
git bisect bad 1fea4d70658b9b1ff831981b01f753741d932a35
# good: [34227d1e30baf6e873ab18b0383d702e584ae583] Merge pull request #281591 from K900/vulkan-bumps
git bisect good 34227d1e30baf6e873ab18b0383d702e584ae583
| 03:28:33 |
| hexa | you tell me | 03:28:39 |
| adamcstephens | I can’t tell. We’re running different things apparently | 03:48:47 |
| adamcstephens | 🤔 | 03:49:06 |
| hexa | In reply to @hexa:lossy.network
https://hydra.nixos.org/build/247058670 from here | 03:49:09 |
| hexa | 
Download image.png | 03:49:18 |
| hexa | it says that during this eval it moved from commit a to commit b | 03:49:29 |
| hexa | tested A as good, B as bad | 03:49:37 |
| hexa | then git bisect run | 03:49:42 |
| hexa | * it says that during this eval it moved from commit A to commit B | 03:49:54 |
| adamcstephens | Thanks | 03:57:32 |
| hexa | beware of git bisect run though | 03:58:04 |
| hexa | lots of stuff on staging is broken and you might need to apply fixes or skip certain commits | 03:58:20 |
| adamcstephens | Yeah I just saw one fail on libuv | 03:58:38 |
| hexa | that one is often flaky 😄 | 03:58:53 |
| hexa | in case it fails on tests | 03:59:12 |
| hexa | 1cf2d7357c61b9c9ca5c92cb0e810571edee948a is the first bad commit
commit 1cf2d7357c61b9c9ca5c92cb0e810571edee948a
Author: Uri Baghin <uri@canva.com>
Date: Thu Jun 29 09:11:55 2023 +1000
gcc: fix c++ headers when same triplet cross compiling
When build platform and host platform differ, but have the same
triple, the code in nixpkgs will consider it a cross compilation,
but gcc won't. This will lead some derivations to look for c++
headers in the wrong place. To solve this always output the headers
in the non-cross location, like we do for the other gcc headers
already.
pkgs/development/compilers/gcc/common/configure-flags.nix | 2 ++
1 file changed, 2 insertions(+)
| 12:12:51 |
| adamcstephens | In reply to @hexa:lossy.network
1cf2d7357c61b9c9ca5c92cb0e810571edee948a is the first bad commit
commit 1cf2d7357c61b9c9ca5c92cb0e810571edee948a
Author: Uri Baghin <uri@canva.com>
Date: Thu Jun 29 09:11:55 2023 +1000
gcc: fix c++ headers when same triplet cross compiling
When build platform and host platform differ, but have the same
triple, the code in nixpkgs will consider it a cross compilation,
but gcc won't. This will lead some derivations to look for c++
headers in the wrong place. To solve this always output the headers
in the non-cross location, like we do for the other gcc headers
already.
pkgs/development/compilers/gcc/common/configure-flags.nix | 2 ++
1 file changed, 2 insertions(+)
thanks, mine still isn't done | 13:27:51 |
| hexa | I did the last few steps manually, and went to bed before finishing them | 13:28:21 |
| adamcstephens | so i think the answer to the original question is that yes the build box is faster :) | 13:35:29 |
| adamcstephens | time to open a PR | 13:35:36 |
| adamcstephens | * time to open a PR to add myself | 13:35:47 |
| hexa | 80 cores and different thermal limits will do that 😄 | 13:40:10 |
| adamcstephens | and more memory too | 13:42:37 |
| adamcstephens | 
Download image.png | 13:43:07 |
