The idea is to build ripgrep with as much optimization as possible. This makes compilation times absolutely obscene. They jump from <10 seconds to 30+ seconds on my i9-12900K. I don't even want to know how long CI would take with these. I tried some ad hoc benchmarks and could not notice any meaningful improvement with the LTO binary versus the normal release profile. Because of that, I still don't think it's worth bloating the release cycle times. Ref #1225
3.3 KiB
3.3 KiB