comparison benchmark/run_and_view.jl @ 1188:d2e73d7b417d tooling/benchmarks

Fix bug in runbenchmark(t,b)
author Jonatan Werpers <jonatan@werpers.com>
date Fri, 27 Jan 2023 12:11:04 +0100
parents 2ece37edb467
children 525015283843
comparison
equal deleted inserted replaced
1187:2ece37edb467 1188:d2e73d7b417d
63 `target` and `baseline` can be any identifier compatible with `hg update`. 63 `target` and `baseline` can be any identifier compatible with `hg update`.
64 64
65 Returns a `PkgBenchmark.BenchmarkJudgement` 65 Returns a `PkgBenchmark.BenchmarkJudgement`
66 """ 66 """
67 function run_benchmark(target, baseline, f=minimum; judgekwargs=Dict()) 67 function run_benchmark(target, baseline, f=minimum; judgekwargs=Dict())
68 t = run_benchmark(target) 68 rev_before = hg_rev()
69 b = run_benchmark(baseline) 69 hg_update(target)
70 t = run_benchmark()
71 hg_update(baseline)
72 b = run_benchmark()
73 hg_update(rev_before)
70 74
71 judged = PkgBenchmark.judge(t,b,f; judgekwargs...) 75 return PkgBenchmark.judge(t,b,f; judgekwargs...)
72
73 return BenchmarkJudgement(t,b,judged)
74 end 76 end
75 77
76 # TBD: How to compare against current working directory? Possible to create a temporary commit? 78 # TBD: How to compare against current working directory? Possible to create a temporary commit?
77 79
78 80