Miscellaneous tidying up

[koushiro]

• Replace chashmap with dashmap
• Replace fnv with fxhash Replace fnv with ahash
• Update some outdated dependencies
• Fix some clippy warnings
• Upgrade MSRV to 1.40.0 (required by dashmap)

Signed-off-by: koushiro koushiro.cqx@gmail.com

$ cargo tree --no-indent | sed 's/ (\*)//' | sort -u | wc -l
110 => 108

[jonhoo]

Updating outdated dependencies + fixing clippy issues is great, but I'm not sure I follow the reasons for moving from chashmap + fnv to dashmap + fxhash? Could you run the benchmarks and see how much of a difference it makes?

[koushiro]

I run the benchmarks several times, the main difference is the sample.

fnv vs fxhash
chashmap vs dashmap (In addition, chashmap has not been actively maintained for a long time)

The following are the results of the benchmarks :

Benchmarking collapse/dtrace/1: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 36.0s or reduce sample count to 20
collapse/dtrace/1       time:   [7.1749 ms 7.1834 ms 7.1936 ms]                               
                        thrpt:  [183.28 MiB/s 183.54 MiB/s 183.75 MiB/s]
                 change:
                        time:   [-10.591% -10.208% -9.8182%] (p = 0.00 < 0.05)
                        thrpt:  [+10.887% +11.369% +11.846%]
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) high mild
  1 (1.00%) high severe
Benchmarking collapse/dtrace/8: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 15.6s or reduce sample count to 30
collapse/dtrace/8       time:   [3.7003 ms 3.7200 ms 3.7415 ms]                               
                        thrpt:  [352.38 MiB/s 354.41 MiB/s 356.30 MiB/s]
                 change:
                        time:   [+3.5867% +6.3238% +8.9067%] (p = 0.00 < 0.05)
                        thrpt:  [-8.1783% -5.9477% -3.4625%]
                        Performance has regressed.
Found 19 outliers among 100 measurements (19.00%)
  12 (12.00%) low severe
  4 (4.00%) low mild
  2 (2.00%) high mild
  1 (1.00%) high severe

Benchmarking collapse/perf/1: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 75.7s or reduce sample count to 10
collapse/perf/1         time:   [14.517 ms 14.579 ms 14.656 ms]                             
                        thrpt:  [204.27 MiB/s 205.35 MiB/s 206.22 MiB/s]
                 change:
                        time:   [-4.5712% -4.0204% -3.3481%] (p = 0.00 < 0.05)
                        thrpt:  [+3.4640% +4.1888% +4.7902%]
                        Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
  1 (1.00%) high mild
  2 (2.00%) high severe
Benchmarking collapse/perf/8: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 30.2s or reduce sample count to 20
collapse/perf/8         time:   [6.9734 ms 7.0537 ms 7.1317 ms]                             
                        thrpt:  [419.78 MiB/s 424.42 MiB/s 429.31 MiB/s]
                 change:
                        time:   [-0.8887% +1.1977% +3.4207%] (p = 0.29 > 0.05)
                        thrpt:  [-3.3076% -1.1835% +0.8967%]
                        No change in performance detected.

Benchmarking collapse/sample: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 31.2s or reduce sample count to 20
collapse/sample         time:   [5.9941 ms 6.0039 ms 6.0157 ms]                             
                        thrpt:  [230.73 MiB/s 231.18 MiB/s 231.56 MiB/s]
                 change:
                        time:   [-32.651% -32.275% -31.850%] (p = 0.00 < 0.05)
                        thrpt:  [+46.735% +47.657% +48.480%]
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) high mild
  1 (1.00%) high severe

     Running target/release/deps/flamegraph-06895b51f7b48927
Benchmarking flamegraph: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 93.9s or reduce sample count to 10
flamegraph              time:   [18.137 ms 18.169 ms 18.207 ms]                        
                        thrpt:  [33.862 MiB/s 33.932 MiB/s 33.992 MiB/s]
                 change:
                        time:   [+0.0002% +0.4953% +0.9201%] (p = 0.03 < 0.05)
                        thrpt:  [-0.9117% -0.4929% -0.0002%]
                        Change within noise threshold.
Found 3 outliers among 100 measurements (3.00%)
  2 (2.00%) high mild
  1 (1.00%) high severe

Benchmarking collapse/dtrace/1: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 36.9s or reduce sample count to 20
collapse/dtrace/1       time:   [7.2298 ms 7.2383 ms 7.2487 ms]                               
                        thrpt:  [181.88 MiB/s 182.15 MiB/s 182.36 MiB/s]
                 change:
                        time:   [+3.3492% +3.7599% +4.2092%] (p = 0.00 < 0.05)
                        thrpt:  [-4.0392% -3.6236% -3.2407%]
                        Performance has regressed.
Found 7 outliers among 100 measurements (7.00%)
  3 (3.00%) high mild
  4 (4.00%) high severe
Benchmarking collapse/dtrace/8: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 15.0s or reduce sample count to 40
collapse/dtrace/8       time:   [3.5561 ms 3.6172 ms 3.6662 ms]                               
                        thrpt:  [359.61 MiB/s 364.48 MiB/s 370.75 MiB/s]
                 change:
                        time:   [-13.011% -10.631% -8.3676%] (p = 0.00 < 0.05)
                        thrpt:  [+9.1317% +11.895% +14.957%]
                        Performance has improved.

Benchmarking collapse/perf/1: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 72.1s or reduce sample count to 10
collapse/perf/1         time:   [13.954 ms 13.966 ms 13.980 ms]                             
                        thrpt:  [214.14 MiB/s 214.36 MiB/s 214.54 MiB/s]
                 change:
                        time:   [-2.5281% -2.2371% -1.8959%] (p = 0.00 < 0.05)
                        thrpt:  [+1.9325% +2.2882% +2.5937%]
                        Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
  2 (2.00%) high mild
  1 (1.00%) high severe
Benchmarking collapse/perf/8: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 32.5s or reduce sample count to 20
collapse/perf/8         time:   [6.9738 ms 6.9876 ms 7.0031 ms]                             
                        thrpt:  [427.49 MiB/s 428.44 MiB/s 429.29 MiB/s]
                 change:
                        time:   [+1.8551% +2.4680% +3.0320%] (p = 0.00 < 0.05)
                        thrpt:  [-2.9427% -2.4086% -1.8213%]
                        Performance has regressed.
Found 4 outliers among 100 measurements (4.00%)
  1 (1.00%) low mild
  3 (3.00%) high severe

Benchmarking collapse/sample: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 30.8s or reduce sample count to 20
collapse/sample         time:   [5.9055 ms 5.9179 ms 5.9322 ms]                             
                        thrpt:  [233.98 MiB/s 234.54 MiB/s 235.04 MiB/s]
                 change:
                        time:   [-33.139% -32.558% -31.924%] (p = 0.00 < 0.05)
                        thrpt:  [+46.895% +48.275% +49.564%]
                        Performance has improved.
Found 3 outliers among 100 measurements (3.00%)
  1 (1.00%) high mild
  2 (2.00%) high severe

     Running target/release/deps/flamegraph-06895b51f7b48927
Benchmarking flamegraph: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 95.3s or reduce sample count to 10
flamegraph              time:   [18.249 ms 18.272 ms 18.299 ms]                        
                        thrpt:  [33.691 MiB/s 33.741 MiB/s 33.783 MiB/s]
                 change:
                        time:   [+0.4541% +0.7753% +1.1155%] (p = 0.00 < 0.05)
                        thrpt:  [-1.1031% -0.7694% -0.4521%]
                        Change within noise threshold.
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) high mild
  1 (1.00%) high severe

[jonhoo]

Thanks! I'm inclined to merge this once CI passes :)
cc @jasonrhansen

[koushiro]

Replace fxhash with ahash.

fnv vs fxhash vs ahash

The following are the results of the benchmarks (fnv vs ahash):

Benchmarking collapse/dtrace/1: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 38.4s or reduce sample count to 20.
collapse/dtrace/1       time:   [7.2652 ms 7.2828 ms 7.3032 ms]                               
                        thrpt:  [180.52 MiB/s 181.03 MiB/s 181.47 MiB/s]
                 change:
                        time:   [-2.4640% -2.0866% -1.6656%] (p = 0.00 < 0.05)
                        thrpt:  [+1.6938% +2.1311% +2.5262%]
                        Performance has improved.
Benchmarking collapse/dtrace/8: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 15.4s or reduce sample count to 40.
collapse/dtrace/8       time:   [3.6853 ms 3.7133 ms 3.7389 ms]                               
                        thrpt:  [352.62 MiB/s 355.06 MiB/s 357.75 MiB/s]
                 change:
                        time:   [+0.7949% +3.2707% +5.6091%] (p = 0.01 < 0.05)
                        thrpt:  [-5.3112% -3.1671% -0.7886%]
                        Change within noise threshold.

Benchmarking collapse/perf/1: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 70.2s or reduce sample count to 10.
collapse/perf/1         time:   [13.347 ms 13.369 ms 13.395 ms]                             
                        thrpt:  [223.50 MiB/s 223.94 MiB/s 224.29 MiB/s]
                 change:
                        time:   [-11.943% -11.483% -11.034%] (p = 0.00 < 0.05)
                        thrpt:  [+12.403% +12.973% +13.563%]
                        Performance has improved.
Benchmarking collapse/perf/8: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 29.9s or reduce sample count to 20.
collapse/perf/8         time:   [6.8566 ms 6.8700 ms 6.8830 ms]                             
                        thrpt:  [434.95 MiB/s 435.77 MiB/s 436.62 MiB/s]
                 change:
                        time:   [+1.3629% +2.8959% +4.4474%] (p = 0.00 < 0.05)
                        thrpt:  [-4.2580% -2.8144% -1.3445%]
                        Performance has regressed.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) low mild

Benchmarking collapse/sample: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 28.9s or reduce sample count to 20.
collapse/sample         time:   [5.6617 ms 5.6686 ms 5.6768 ms]                             
                        thrpt:  [244.51 MiB/s 244.86 MiB/s 245.15 MiB/s]
                 change:
                        time:   [-35.032% -34.876% -34.706%] (p = 0.00 < 0.05)
                        thrpt:  [+53.154% +53.553% +53.921%]
                        Performance has improved.
Found 7 outliers among 100 measurements (7.00%)
  2 (2.00%) high mild
  5 (5.00%) high severe

     Running target/release/deps/flamegraph-167a004553da2145
Benchmarking flamegraph: Warming up for 3.0000 s
Warning: Unable to complete 100 samples in 5.0s. You may wish to increase target time to 90.7s or reduce sample count to 10.
flamegraph              time:   [17.696 ms 17.825 ms 17.949 ms]                        
                        thrpt:  [34.349 MiB/s 34.588 MiB/s 34.839 MiB/s]
                 change:
                        time:   [-5.4637% -4.9816% -4.4606%] (p = 0.00 < 0.05)
                        thrpt:  [+4.6688% +5.2428% +5.7795%]
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  1 (1.00%) high mild
  1 (1.00%) high severe

[koushiro]

Azure Pipelines seems to be wrong due to configuration issues. I'm not familiar with the configuration of Azure Pipelines :(
Can you fix the CI? @jonhoo

[jonhoo]

@koushiro CI should now be fixed on master

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (master@db49bd8). Click here to learn what that means.
The diff coverage is 100%.

@@            Coverage Diff            @@
##             master     #155   +/-   ##
=========================================
  Coverage          ?   88.51%           
=========================================
  Files             ?       16           
  Lines             ?     2212           
  Branches          ?        0           
=========================================
  Hits              ?     1958           
  Misses            ?      254           
  Partials          ?        0

Impacted Files	Coverage Δ
src/collapse/perf.rs	97.39% <ø> (ø)
src/collapse/common.rs	61.11% <ø> (ø)
src/collapse/dtrace.rs	89.8% <ø> (ø)
src/flamegraph/color/mod.rs	73.04% <100%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update db49bd8...7d7ed65. Read the comment docs.

[jonhoo]

Great, thanks @koushiro! I forgot to ask you if you could please also update the CHANGELOG to reflect this change? Once you do I can do another release! I think this should just be a minor release?

[koushiro]

@jonhoo Yes, it should just be a minor release.

How about the CHANGELOG below?

## [0.9.2] - 2020-01-30
### Changed
- Replace `chashmap` with `dashmap`
- Replace `fnv` with `ahash`
- Update some outdated dependencies
- Fix some clippy warnings
- Upgrade MSRV to 1.40.0 (required by `dashmap`)

Is there anything else to add?

[jonhoo]

Just add it under the existing Unreleased section and I'll bump things when I do the release. You can probably remove the line about clippy warnings. It'd also be good to link to this PR! Apart from that it looks good :)

[koushiro]

Okay :)