From: Joe Tsai Date: Sun, 1 Oct 2023 19:15:14 +0000 (-0700) Subject: strings: optimize Repeat for common substrings X-Git-Tag: go1.23rc1~847 X-Git-Url: http://www.git.cypherpunks.su/?a=commitdiff_plain;h=3c78ace24f3aa025a72b53be3b83423f9f24ee5d;p=gostls13.git strings: optimize Repeat for common substrings According to static analysis of Go source code known by the module proxy, spaces, dashes, zeros, and tabs are the most commonly repeated string literals. Out of ~69k total calls to Repeat: * ~25k calls are repeats of " " * ~7k calls are repeats of "-" * ~4k calls are repeats of "0" * ~2k calls are repeats of "=" * ~2k calls are repeats of "\t" After this optimization, ~60% of Repeat calls will go through the fast path. These are often used in padding of fixed-width terminal UI or in the presentation of humanly readable text (e.g., indentation made of spaces or tabs). Optimize for this case by handling short repeated sequences of common literals. Performance: name old time/op new time/op delta RepeatSpaces-24 19.3ns ± 1% 5.0ns ± 1% -74.27% (p=0.000 n=8+9) name old alloc/op new alloc/op delta RepeatSpaces-24 2.00B ± 0% 0.00B -100.00% (p=0.000 n=10+10) name old allocs/op new allocs/op delta RepeatSpaces-24 1.00 ± 0% 0.00 -100.00% (p=0.000 n=10+10) Change-Id: Id1cafd0cc509e835c8241a626489eb206e0adc3c Reviewed-on: https://go-review.googlesource.com/c/go/+/536615 Reviewed-by: Ian Lance Taylor Reviewed-by: Emmanuel Odeke Reviewed-by: Than McIntosh LUCI-TryBot-Result: Go LUCI --- diff --git a/src/strings/strings.go b/src/strings/strings.go index f3f0723721..f53ae1f9a7 100644 --- a/src/strings/strings.go +++ b/src/strings/strings.go @@ -530,6 +530,27 @@ func Map(mapping func(rune) rune, s string) string { return b.String() } +// According to static analysis, spaces, dashes, zeros, equals, and tabs +// are the most commonly repeated string literal, +// often used for display on fixed-width terminal windows. +// Pre-declare constants for these for O(1) repetition in the common-case. +const ( + repeatedSpaces = "" + + " " + + " " + repeatedDashes = "" + + "----------------------------------------------------------------" + + "----------------------------------------------------------------" + repeatedZeroes = "" + + "0000000000000000000000000000000000000000000000000000000000000000" + repeatedEquals = "" + + "================================================================" + + "================================================================" + repeatedTabs = "" + + "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t" + + "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t" +) + // Repeat returns a new string consisting of count copies of the string s. // // It panics if count is negative or if the result of (len(s) * count) @@ -557,6 +578,23 @@ func Repeat(s string, count int) string { return "" } + // Optimize for commonly repeated strings of relatively short length. + switch s[0] { + case ' ', '-', '0', '=', '\t': + switch { + case n <= len(repeatedSpaces) && HasPrefix(repeatedSpaces, s): + return repeatedSpaces[:n] + case n <= len(repeatedDashes) && HasPrefix(repeatedDashes, s): + return repeatedDashes[:n] + case n <= len(repeatedZeroes) && HasPrefix(repeatedZeroes, s): + return repeatedZeroes[:n] + case n <= len(repeatedEquals) && HasPrefix(repeatedEquals, s): + return repeatedEquals[:n] + case n <= len(repeatedTabs) && HasPrefix(repeatedTabs, s): + return repeatedTabs[:n] + } + } + // Past a certain chunk size it is counterproductive to use // larger chunks as the source of the write, as when the source // is too large we are basically just thrashing the CPU D-cache. diff --git a/src/strings/strings_test.go b/src/strings/strings_test.go index f93cf6842f..ac493c7dcd 100644 --- a/src/strings/strings_test.go +++ b/src/strings/strings_test.go @@ -1111,6 +1111,13 @@ func TestCaseConsistency(t *testing.T) { } var longString = "a" + string(make([]byte, 1<<16)) + "z" +var longSpaces = func() string { + b := make([]byte, 200) + for i := range b { + b[i] = ' ' + } + return string(b) +}() var RepeatTests = []struct { in, out string @@ -1123,6 +1130,12 @@ var RepeatTests = []struct { {"-", "-", 1}, {"-", "----------", 10}, {"abc ", "abc abc abc ", 3}, + {" ", " ", 1}, + {"--", "----", 2}, + {"===", "======", 2}, + {"000", "000000000", 3}, + {"\t\t\t\t", "\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t", 4}, + {" ", longSpaces, len(longSpaces)}, // Tests for results over the chunkLimit {string(rune(0)), string(make([]byte, 1<<16)), 1 << 16}, {longString, longString + longString, 2}, @@ -1925,6 +1938,13 @@ func BenchmarkRepeatLarge(b *testing.B) { } } +func BenchmarkRepeatSpaces(b *testing.B) { + b.ReportAllocs() + for i := 0; i < b.N; i++ { + Repeat(" ", 2) + } +} + func BenchmarkIndexAnyASCII(b *testing.B) { x := Repeat("#", 2048) // Never matches set cs := "0123456789abcdefghijklmnopqrstuvwxyz0123456789abcdefghijklmnopqrstuvwxyz"