rustc-perf
rustc-perf copied to clipboard
rust-lang
Website for graphing performance of rustc
This PR changes the absolute (left) graph in benchmark detail to use the "percent from first commit" value instead. This should make it easier to observe smaller changes (more discussion...
Inspired by discussions about my [recent blog post](https://kobzol.github.io/rust/rustc/2024/03/15/rustc-what-takes-so-long.html), I thought that it would be nice if we split the "Backend" section into the actual backend (e.g. LLVM) and the Rust...
Currently, if we want to open a query trace gathered from a self-profile, we need to go through `ui.perfetto.dev`. It takes considerable time to load and redirects to an external...
See https://github.com/rust-lang/rust/pull/117772#issuecomment-1951430297, quoted here to sidestep github's collapsing of comments: > On a closer inspection, the results for ripgrep, regex and html5ever are quite misleading. Those crates actually have redundant...
It would be nice to have some runtime benchmark for `async/await` usage, Futures etc., e.g. using `tokio`.
The runtime benchmark table on the [compage page](https://perf.rust-lang.org/compare.html) currently doesn't show any detail nor historical charts, which makes it difficult to analyze the results. It would be nice to add...
The dashboard currently contains data for `check`/`debug`/`opt`/`doc` builds. It would be nice to also add a historical chart for our runtime benchmarks to it (probably with the wall-time metric, same...
As can be seen on https://perf.rust-lang.org/index.html?kind=percentrelative&stat=size%3Adoc_bytes there some persistent few byte noise on graph, which not exist for other benches, which makes me think that there something interesting going on...
Based on https://github.com/rust-lang/wg-binary-size/issues/3, we would like to see binary size outputs on a few different targets show up as perf results. Just checking output size should be relatively fast compared...
After the recent self-profile refactorings, I thought that we might simplify the self-profile handling even more, by eagerly loading the self-profile file into memory, which avoids the hack with generating...