Use performance chart from itoa-benchmark

Author: dtolnay

Cargo.toml

@@ -6,7 +6,7 @@ categories = ["value-formatting", "no-std", "no-std::no-alloc"]
 description = "Fast integer primitive to string conversion"
 documentation = "https://docs.rs/itoa"
 edition = "2021"
-exclude = ["performance.png", "chart/**"]
+exclude = ["*.png", "chart/**"]
 keywords = ["integer"]
 license = "MIT OR Apache-2.0"
 repository = "https://github.com/dtolnay/itoa"

README.md

@@ -35,9 +35,17 @@ fn main() {
 
 <br>
 
-## Performance (lower is better)
+## Performance
 
-![performance](https://raw.githubusercontent.com/dtolnay/itoa/master/performance.png)
+The [itoa-benchmark] compares this library and other Rust integer formatting
+implementations across a range of integer sizes. The vertical axis in this chart
+shows nanoseconds taken by a single execution of
+`itoa::Buffer::new().format(value)` so a lower result indicates a faster
+library.
+
+[itoa-benchmark]: https://github.com/dtolnay/itoa-benchmark
+
+![performance](https://raw.githubusercontent.com/dtolnay/itoa/master/itoa-benchmark.png)
 
 <br>
 

itoa-benchmark.png

@@ -25,9 +25,17 @@
 //! }
 //! ```
 //!
-//! # Performance (lower is better)
+//! # Performance
 //!
-//! ![performance](https://raw.githubusercontent.com/dtolnay/itoa/master/performance.png)
+//! The [itoa-benchmark] compares this library and other Rust integer formatting
+//! implementations across a range of integer sizes. The vertical axis in this
+//! chart shows nanoseconds taken by a single execution of
+//! `itoa::Buffer::new().format(value)` so a lower result indicates a faster
+//! library.
+//!
+//! [itoa-benchmark]: https://github.com/dtolnay/itoa-benchmark
+//!
+//! ![performance](https://raw.githubusercontent.com/dtolnay/itoa/master/itoa-benchmark.png)
 
 #![doc(html_root_url = "https://docs.rs/itoa/1.0.16")]
 #![no_std]