averello/swift-mirror
1
0
Fork 0
mirror of https://github.com/apple/swift.git synced 2025-12-21 12:14:44 +01:00
Code Issues Packages Projects Releases Wiki Activity
87,123 Commits 623 Branches 2,949 Tags
22d37eccd818b51fbbefda5b6f13d2bc8f44f757
Commit Graph

383 Commits

Author SHA1 Message Date
Johannes Weiss
7226dca81f benchmark: NIOChannelPipeline imitation 
adds a benchmark that imitates SwiftNIO's ChannelPipeline. Recently,
Swift master regressed on this benchmark, therefore I believe adding it
to Swift provides value as it seems to be different enough to the existing
benchmarks.
 2019-05-14 21:03:03 +01:00
Patrick Pijnappel
150bcb0503 [benchmark] Add integer parsing benchmarks 2019-05-04 09:17:42 +10:00
Pavol Vaskovic
ea80003b0c Merge pull request #24307 from palimondo/legacy-factor-doc 
[benchmark] Documentation for legacyFactor
 2019-04-29 08:02:53 +02:00
Pavol Vaskovic
6ca706eb1d Merge pull request #24357 from palimondo/prng-splitmix64 
[benchmark] Add SplitMix64 PRNG
 2019-04-29 06:53:21 +02:00
Pavol Vaskovic
d2a47db71a [benchmark] Add SplitMix64 PRNG 
Very fast pseudorandom number generator with 64 bits of state, conforming to `RandomNumberGenerator` protocol, passing BigCrush.
 2019-04-28 19:23:42 +02:00
Pavol Vaskovic
e779e06ba2 [benchmark] Documentation for legacyFactor 2019-04-28 18:23:01 +02:00
Raj Barik
34fbbcc842 Merge pull request #24206 from rajbarik/raj-bench 
Performance Benchmarking of ExistentialSpecializer
 2019-04-26 08:55:29 -07:00
Michael Ilseman
2022b5c566 Merge pull request #24210 from troughton/floating-point-parsing-benchmark 
[benchmark] Add a benchmark for floating point parsing performance
 2019-04-25 15:13:38 -07:00
Raj Barik
2f32778de4 Performance Benchmarking of ExistentialSpecializer 2019-04-25 10:38:36 -07:00
Thomas Roughton
4b50c3ef3e [benchmark] Add a benchmark for floating point parsing performance 2019-04-24 13:31:23 +12:00
Michael Ilseman
32a2031a19 [benchmark] View variants for CSVParsing 
Generalize CSVParsing so that it can be ran across all of String's
views, and add corresponding benchmarks.

Note: This does change the profile of the original benchmark, which
was an odd mixture of String and UTF16View.
 2019-04-22 15:37:53 -07:00
Michael Gottesman
115f7a43e0 Move build_script_helper from ./benchmarks/utils => ./benchmarks/scripts. 2019-04-10 22:18:50 -07:00
Michael Gottesman
df539948d5 [benchmarks] Add support for building the benchmarks via build-script against the just created toolchain. 
Now one can on Darwin/Linux build the benchmarks via swiftpm from build-script by passing in:

```
build-script $NORMAL_ARGS --install-swift --install-swiftpm --install-llbuild --toolchain-benchmarks --swiftpm --llbuild
```

This is done using the infrastructure that BenL added for sourcekit-lsp.
 2019-04-07 13:25:31 -07:00
David Smith
d2a59e0edf Add new benchmarks for NSString bridging to cover non-tagged cases 2019-03-21 14:17:39 -07:00
Patrick Pijnappel
06a2e6d5d2 [test] Add Range overlaps benchmark 2019-03-14 20:26:28 +11:00
Pavol Vaskovic
539f2cdcdf [benchmark] Remove DoubleWidthDivision (dead code) 2019-03-01 20:40:18 +01:00
Pavol Vaskovic
40cce2da3d Merge pull request #22707 from palimondo/within-one-stem 
[benchmark] Move fake autoreleasepool to TestUtils
 2019-02-20 12:35:26 +01:00
Pavol Vaskovic
7896909d23 [benchmark] Move fake autoreleasepool to TestUtils 
This workaround for writing cross platform benchmarks is required in multiple performance tests.
 2019-02-19 07:50:12 +01:00
Pavol Vaskovic
16af31a39b [benchmark] Remove HashQuadratic 
This was obsoleted by DictionaryCopy.
 2019-02-17 07:09:14 +01:00
Pavol Vaskovic
0ac591e27b [benchmark] Add .existential BenchmarkCategory 2019-02-06 12:38:50 +01:00
Ben Cohen
779ea19a6a Revert count(where:) 2019-01-31 18:57:17 -08:00
Patrick Balestra
fecff67e3e [benchmark] Sort registerBenchmark statements alphabetically 2019-01-20 22:00:26 +01:00
Lance Parker
15aaa1e777 [stdlib]String normalization functions (#21026) 
* fast/foreignNormalize functions
 2019-01-08 13:55:29 -08:00
Pavol Vaskovic
ec836bd04b Merge pull request #20861 from palimondo/a-tall-white-fountain-played 
[benchmark] Janitor Duty, Legacy Factor: A-C
 2018-12-07 19:49:06 +01:00
Pavol Vaskovic
9d6f7ad160 [benchmark] Driver & Doctor: Lower the sample cap 
Lowered the default sample cap from 2k to 200. (This doesn’t effect manually specified `--num-samples` argument in the driver.)

Swift benchmarks have pretty constant performance profile over time. It’s more beneficial to get multiple independent measurements faster, than more samples from the same run.
 2018-12-07 15:06:43 +01:00
David Smith
f85e1fee22 Generalize the JSON perf tests to Codable 2018-12-03 11:16:47 -08:00
David Smith
c129066b39 Add simple benchmarks for JSON coding 2018-11-29 17:41:50 -08:00
Karoy Lorentey
74eaea2a6a [benchmark] Add some benchmarks for String breadcrumbs (#20769) 
[benchmark] Add some benchmarks for String breadcrumbs
 2018-11-28 16:24:29 -08:00
Pavol Vaskovic
0bdd3ef275 [benchmark] Equalize memory usage (w&w/o verbose) 
The use of `--verbose` parameter was affecting the reported memory usage (`--memory`), because it front-loads initialization of string interpolation and printing.

By always computing the configuration string and always calling print, the baseline memory measurement no longer includes this constant overhead.
 2018-11-28 21:34:12 +01:00
Andreas Jönsson
09768f29c5 [benchmark] fixes index(of:) deprecation warnings in ArgParse 2018-11-27 22:57:22 +01:00
swift-ci
f093925036 Merge pull request #20452 from Catfish-Man/stringtrampolines 2018-11-15 17:55:34 -08:00
David Smith
8bb6b78e36 Use super trampolines to let us override several more NSString methods (-UTF8String, -cStringUsingEncoding:, and -getCString:maxLength:encoding:) for performance 2018-11-15 17:10:51 -08:00
Maxim Moiseev
cbf83ac04f [NFC][stdlib] Add FIXME markers to simplify audit 2018-11-14 11:58:42 -08:00
Slava Pestov
f6c2caf64b stdlib: Add @inlinable to @inline(__always) declarations 
These should be audited since some might not actually need to be
@inlinable, but for now:

- Anything public and @inline(__always) is now also @inlinable
- Anything @usableFromInline and @inline(__always) is now @inlinable
 2018-11-13 15:15:07 -05:00
David Smith
634e6bc584 Add benchmarks for various ObjC NSString operations on bridged Swift Strings 2018-11-06 13:43:24 -08:00
eeckstein
f64f02bfde Merge pull request #20212 from palimondo/fluctuation-of-the-pupil 
[benchmark] Legacy Factor
 2018-11-06 11:31:08 -08:00
Lance Parker
7376009ccc Add benchmarks and tests for the normalized iterator (#32) 
Add benchmarks and tests for the normalized iterator
 2018-11-04 10:42:41 -08:00
Pavol Vaskovic
a7f832fb57 [benchmark] Legacy factor 
This adds optional `legacyFactor` to the `BenchmarkInfo`, which allows for linear modification of constants that unnecesarily inflate the base workload of benchmarks, while maintaining the continuity of log-term benchmark tracking.

For example, if a benchmark uses `for _ in N*10_000` in its run function, we could lower this to `for _ in N*1_000` and adding a `legacyFactor: 10` to its `BenchmarkInfo`.

Note that this doesn’t affect the real measurements gathered from the `--verbose` output. The `BenchmarkDoctor` has been slightly adjusted to work with these real samples, therefore `Benchmark_Driver check` will not flag these benchmarks for slow run time reported in the summary, if their real runtimes fall into the recommended range.
 2018-11-01 06:24:27 +01:00
eeckstein
cd920b69f4 Merge pull request #19910 from palimondo/fluctuation-of-the-pupil 
[benchmark] More Robust Benchmark_Driver
 2018-10-29 15:02:07 -07:00
Michael Gottesman
ba7815b663 [benchmark] Fix swiftpm based benchmark build on Linux. 2018-10-29 12:15:20 -07:00
Michael Gottesman
4d76ff9681 [benchmark] Add two benchmarks that show performance of flattening an array. 
The first is a naive imperative approach using appends in a loop. The second
uses flatMap. We would like both of these to have equivalent performance.
 2018-10-28 15:55:26 -07:00
Karoy Lorentey
f93dcf3dfa [benchmark] Add benchmark for [AnyHashable: Any] with String keys 2018-10-26 11:56:41 +01:00
Patrick Balestra
1c0778bb5b [benchmark] Add insert(_:Character) benchmark with ASCII and non-ASCII characters 
Adds insert character benchmark with ASCII and non-ASCII characters
 2018-10-24 18:55:20 -07:00
Pavol Vaskovic
67b489dcb1 [benchmark] Auto-determine number of samples 
When measuring with specified number of iterations (generally, `--num-iters=1` makes sense), automaticially determine the number of samples to take, so that the overall measurement duration comes close to `sample-time`.

This is the same technique used to scale `num-iters` before, but for `num-samples`.
 2018-10-11 18:56:27 +02:00
Pavol Vaskovic
9d9200e9eb [benchmark] Measure setUpFunction 
Measure the duration of the `setUpFunction` and report it in verbose mode.

This will be used by `BenchmarkDoctor`, to ensure there isn’t unreasonably big imbalance between the time it takes to set up and run the actual benchmark.
 2018-10-02 14:34:43 +02:00
Pavol Vaskovic
a9f0ce4338 [benchmark] Fix quantile estimation type 
The correct quantile estimation type for printing all measurements in the summary report while `quantile == num-samples - 1` is R-1, SAS-3.  It's the inverse of empirical distribution function.

References:
* https://en.wikipedia.org/wiki/Quantile#Estimating_quantiles_from_a_sample
* discussion in https://github.com/apple/swift/pull/19097#issuecomment-421238197
 2018-09-20 09:19:07 +02:00
Pavol Vaskovic
f0e7b8737a [benchmark] Round quantile idx to nearest or even 
Explicitly use round-half-to-even rounding algorithm to match the behavior of numpy's quantile(interpolation='nearest') and quantile estimate type R-3, SAS-2. See:
https://en.wikipedia.org/wiki/Quantile#Estimating_quantiles_from_a_sample
 2018-09-14 23:40:43 +02:00
Pavol Vaskovic
8b3b1f695a [benchmark] Option: delta encoded quantiles format 
Added `--delta` argument to print the quantiles in delta encoded format, that ommits 0s.

This results in machine and human readable output that highlights modes and is easily digestible, giving you the feel for the underlying probability distribution of the samples in the reported results:

````
$ ./Benchmark_O --num-iters=1 --num-samples=20 --quantile=20 --delta 170 171 184 185 198 199 418 419 432 433 619 620
#,TEST,SAMPLES,MIN(μs),𝚫V1,𝚫V2,𝚫V3,𝚫V4,𝚫V5,𝚫V6,𝚫V7,𝚫V8,𝚫V9,𝚫VA,𝚫VB,𝚫VC,𝚫VD,𝚫VE,𝚫VF,𝚫VG,𝚫VH,𝚫VI,𝚫VJ,𝚫MAX
170,DropFirstArray,20,171,,,,,,,,,,,,,,,,,,,2,29
171,DropFirstArrayLazy,20,168,,,,,,,,,,,,,,,,,,,,8
184,DropLastArray,20,55,,,,,,,,,,,,,,,,,,,,26
185,DropLastArrayLazy,20,65,,,,,,,,,,,,,,,,,,,1,90
198,DropWhileArray,20,214,1,,,,,,,,,,,,,,,,,1,27,2
199,DropWhileArrayLazy,20,464,,,,1,,,,,,,,1,1,1,4,9,1,9,113,2903
418,PrefixArray,20,132,,,,,,,,,,,,,,,,,1,1,32,394
419,PrefixArrayLazy,20,168,,,,,,,,,,,,1,,2,9,1,15,8,88,3338
432,PrefixWhileArray,20,252,1,,,,1,,,,,,,,,,,1,,,,30
433,PrefixWhileArrayLazy,20,168,,,,,,,,,,,,,1,,6,6,14,43,28,10200
619,SuffixArray,20,68,,,,,,,,,,,,,1,,,,22,1,1,4
620,SuffixArrayLazy,20,65,,,,,,,,,,,,,,,,,,1,9,340
````
 2018-09-14 23:40:43 +02:00
Pavol Vaskovic
72e960457b [benchmark] Gardening maxRSS as Int? 2018-09-14 23:40:43 +02:00
Pavol Vaskovic
022e1111a9 [benchmark] Report quantiles from samples 
The default benchmark result reports statistics of a normal distribution — mean and standard deviation. Unfortunately the samples from our benchmarks are *not normally distributed*. To get a better picture of the underlying probability distribution, this adds support for reporting quantiles.

See https://en.wikipedia.org/wiki/Quantile

This gives better subsample of the measurements in the summary, without need to resort to the use of a full verbose mode, which might be unnecessarily slow.
 2018-09-14 23:40:43 +02:00