[moggers123]

SQLx is great, but I really wish they had a non-async interface. I had to switch a project from sqlx to rusqlite seemingly just due to the overhead of the async machinery. Saw a 20x latency reduction that I narrowed down to "probably async" (sort of hard to tell, I find it very difficult to do perf analysis of async code). I try to avoid discussing async so as to not come off as a frothing-at-the-mouth-chest-thumping-luddite but honestly, if sqlx had a non-async interface I'd be very happy to accept the "you don't need to use it" argument. its the only place where I don't feel like I really have a choice.

[pie_flavor]

Async does not incur 20x slowdowns when you're I/O bound. It would be ridiculous for copying a few bytes to be slower than a syscall. This sounds like mutex issues, or WAL config, or something like that.

[moggers123]

I just chucked something together to try and demonstrate. I don't see the massive 20x slowdown, only about a 3x slowdown (5x on release build). Still enough to be painful for the use case in question.

https://github.com/Moggers/rusqlitebenchmark

Do you think you could look through it and point out what you think the reason is? I think they've both got the same WAL and mutex settings. Its a very contrived and synthetic example but actually somewhat representative of what the original code wanted to do.