David Himmelstrup
David Himmelstrup
Could you elaborate on your use-case? Why exactly do you want the minimum sample size? And what statistical properties are you looking for in the results?
Ah, so something like #110.
I doubt this would give more stable results. What's your end goal? Stable results even when the CPU is under load? I'm not sure that is possible.
`black_box` doesn't do what you think it does. In your code you have: ```rust let _ = black_box(|| copy += 1u64); ``` But that doesn't make sense. It is identical...
Is this still true? Looking at crates.io, `serde-json` only pulls in `iota` and `ryu`. `miniserde` also depends on those libraries.
Here's a tree of all the `serde` related dependencies required by criterion: ``` ├── serde v1.0.126 ├── serde_cbor v0.11.1 │ ├── half v1.7.1 │ └── serde v1.0.126 ├── serde_derive v1.0.126...
Status update. Criterion contains 10 `unsafe` blocks and 1 `unsafe` function: 1. https://github.com/bheisler/criterion.rs/blob/cee63459dc58f86a9c319bd4fa7e762f7c333bf6/src/stats/univariate/percentiles.rs#L20 2. https://github.com/bheisler/criterion.rs/blob/cee63459dc58f86a9c319bd4fa7e762f7c333bf6/src/stats/univariate/percentiles.rs#L52 2. https://github.com/bheisler/criterion.rs/blob/cee63459dc58f86a9c319bd4fa7e762f7c333bf6/src/stats/univariate/percentiles.rs#L57 3. https://github.com/bheisler/criterion.rs/blob/cee63459dc58f86a9c319bd4fa7e762f7c333bf6/src/stats/univariate/percentiles.rs#L67 4. https://github.com/bheisler/criterion.rs/blob/cee63459dc58f86a9c319bd4fa7e762f7c333bf6/src/stats/univariate/percentiles.rs#L72 2. https://github.com/bheisler/criterion.rs/blob/cee63459dc58f86a9c319bd4fa7e762f7c333bf6/src/stats/univariate/sample.rs#L31 3. https://github.com/bheisler/criterion.rs/blob/cee63459dc58f86a9c319bd4fa7e762f7c333bf6/src/stats/univariate/sample.rs#L133 4. https://github.com/bheisler/criterion.rs/blob/cee63459dc58f86a9c319bd4fa7e762f7c333bf6/src/stats/mod.rs#L84 5. https://github.com/bheisler/criterion.rs/blob/976d51730ce479d3d032d2dbb176022e45b22b57/src/stats/univariate/resamples.rs#L55 6....
> My use case for wanting to disable as much is possible is that I run `cargo test --benches` in CI just to make sure my benchmarks compile and run...
Done in #496
A basic version of this is now operational. I'd like to see it included in the upcoming 0.4 release. ``` $ cargo wasi bench --no-default-features -- Fibonacci/Recursive --discard-baseline --noplot Finished...