Crate wgpu_benchmark 

Benchmarking framework for wgpu.

This crate is a basic framework for benchmarking. Its design is guided by a few goals:

• Enumerating tests should be extremely cheap. criterion needs to run all of your benchmark functions to enumerate them during testing. This requires your code to contort itself to avoid doing any work until you enter a benchmark callback. This framework avoids that by having an explicit list of benchmark function.
• It must be compatible with cargo-nextest and have a compatible “test” mode that runs each benchmark exactly once.
• It should be able to have intuitive test grouping, allowing for allowing for quick execution of a reasonable baseline set of benchmarks during development, while still allowing for a more exhaustive benchmark suite to be run if desired.

By default all tests run for 2 seconds, but this can be overridden by individual tests.

Modules

context 🔒

file 🔒

iter 🔒

print 🔒

Structs

Benchmark

BenchmarkContext

BenchmarkFile

SubBenchResult

Enums

LoopControl

Constants

HELP 🔒

Functions

iter

iter_auto

iter_many

main