Fast single-threaded logging framework, almost 200x faster than tracing and delog for large structs.
Supports standard logging macros like trace!, debug!, info!, warn! and error!.
Flushing is deferred until flush!() macro is called.
- Deferred formatting
- Deferred I/O
- Low call site latency
While tracing is a popular library for event tracing and are really good at what they do, quicklog is optimized for low callsite latency, paying the cost of formatting and I/O on a separate thread, away from the hot path.
cargo add quicklog# Cargo.toml
[dependencies]
quicklog = "0.2"
# ...use quicklog::{info, init, flush};
fn main() {
// initialise required resources, called near application entry point
init!();
// adds item to logging queue
info!("hello world");
let some_var = 10;
// variables are passed by copy
info!("value of some_var: {}", some_var);
// flushes everything in queue
flush!();
}In order to avoid cloning a large struct, you can implement the Serialize trait.
This allows you to copy specific parts of your struct onto a circular byte buffer and avoid copying the rest by encoding providing a function to decode your struct from a byte buffer.
For a complete example, refer to ~/quicklog/benches/logger_benchmark.rs.
use quicklog::serialize::{Serialize, Store};
struct SomeStruct {
num: i64
}
impl Serialize for SomeStruct {
fn encode(&self, write_buf: &'static mut [u8]) -> Store { /* some impl */ }
fn buffer_size_required(&self) -> usize { /* some impl */ }
}
fn main() {
let s = SomeStruct { num: 1_000_000 };
info!("some struct: {}", ^s);
}use quicklog_flush::stdout_flusher::StdoutFlusher;
use quicklog::{info, init, flush, with_flush_into_file, with_flush};
fn main() {
init!();
// flush into stdout
with_flush!(StdoutFlusher);
// item goes into logging queue
info!("hello world");
// flushed into stdout
flush!()
// flush into a file path specified
with_flush_into_file!("logs/my_log.log");
info!("shave yaks");
// flushed into file
flush!();
}More usage examples are available here.
Measurements are made on a 2020 16 core M1 Macbook Air with 16 GB RAM.
| Logger | Lower Bound | Estimate | Upper Bound |
|---|---|---|---|
| quicklog | 103.76 ns | 104.14 ns | 104.53 ns |
| tracing | 22.336 µs | 22.423 µs | 22.506 µs |
| delog | 21.528 µs | 21.589 µs | 21.646 µs |
| Logger | Lower Bound | Estimate | Upper Bound |
|---|---|---|---|
| quicklog | 61.399 ns | 62.436 ns | 63.507 ns |
| tracing | 2.6501 µs | 2.6572 µs | 2.6646 µs |
| delog | 2.7610 µs | 2.7683 µs | 2.7761 µs |
| Logger | Lower Bound | Estimate | Upper Bound |
|---|---|---|---|
| quicklog | 28.561 ns | 28.619 ns | 28.680 ns |
| tracing | 627.79 µs | 629.91 µs | 632.06 µs |
| delog | 719.54 µs | 721.19 µs | 722.96 µs |
We are open to contributions and requests!
Please post your bug reports or feature requests on Github Issues.
- [] add single-threaded and multi-threaded variants
- [] Try to remove nested
lazy_formatin recursion - [] Check number of copies of data made in each log line and possibly reduce it
- [] Review uses of unsafe code
- [] Benchmark multi-threaded performance
- [] Statically assert that strings inside Level and LevelFilter are the same size
Zack Ng, Tien Dat Nguyen, Michiel van Slobbe, Dheeraj Oswal
Copyright 2023 Grasshopper Asia
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.