v1.6.0

BenchmarkTools v1.6.0

Diff since v1.5.0

Merged pull requests:

• Bump codecov/codecov-action from 3 to 4 (#355) (@dependabot[bot])
• Bump julia-actions/setup-julia from 1 to 2 (#361) (@dependabot[bot])
• Add buildkite configuration stub. (#363) (@fredrikekre)
• Raise minimum Julia version to 1.6 (#366) (@Zentrik)
• Run gcscrub after warmup and remove unnecessary gc run for first sample (#367) (@Zentrik)
• no empty varargs (#373) (@KlausC)
• add docstring to judge (#378) (@bclrk)
• "evaluations" -> "evaluations each" when printing Trial (#380) (@asinghvi17)
• Update Profile output test for v1.12 (#381) (@vtjnash)
• Add @btimed and @ballocations macros (#383) (@singularitti)
• rename default branch to main (#384) (@willow-ahrens)
• Fix compilation and interpolation memory leaks (#385) (@willow-ahrens)
• deprecate @benchmarkset (#386) (@willow-ahrens)

Closed issues:

• Pass setup/interpolated variables through as run arguments instead of closing over them in the harness (#127)
• @benchmarkset breaks on ^C (#279)
• Feature for just running the code without benchmarking (#300)
• Memory leak when repeatedly benchmarking (#339)
• is @benchmarkset usable ? (#343)
• judge lacks a docstring (#356)
• timeout measures warmup (eliminating many of the reasons to use timeout) (#368)
• @Benchmarkset fails silently, returns null (#369)