package Benchmark;
+use strict;
+
+
=head1 NAME
Benchmark - benchmark running times of Perl code
The COUNT can be zero or negative, see timethis().
-Returns a hash of Benchmark objects, keyed by name.
+Returns a hash reference of Benchmark objects, keyed by name.
=item timediff ( T1, T2 )
cmpthese( $results );
in case you want to see both sets of results.
+If the first argument is an unblessed hash reference,
+that is RESULTSHASHREF; otherwise that is COUNT.
Returns a reference to an ARRAY of rows, each row is an ARRAY of cells from the
above chart, including labels. This:
September, 2002; by Jarkko Hietaniemi: add ':hireswallclock' special tag.
+February, 2004; by Chia-liang Kao: make cmpthese and timestr use time
+statistics for children instead of parent when the style is 'nop'.
+
=cut
# evaluate something in a clean lexical environment
-sub _doeval { eval shift }
+sub _doeval { no strict; eval shift }
#
# put any lexicals at file scope AFTER here
use Carp;
use Exporter;
-@ISA=(Exporter);
+
+our(@ISA, @EXPORT, @EXPORT_OK, %EXPORT_TAGS, $VERSION);
+
+@ISA=qw(Exporter);
@EXPORT=qw(timeit timethis timethese timediff timestr);
@EXPORT_OK=qw(timesum cmpthese countit
clearcache clearallcache disablecache enablecache);
%EXPORT_TAGS=( all => [ @EXPORT, @EXPORT_OK ] ) ;
-$VERSION = 1.0501;
+$VERSION = 1.09;
# --- ':hireswallclock' special handling
sub mytime () { time }
-&init;
+init();
sub BEGIN {
if (eval 'require Time::HiRes') {
my $class = shift;
if (grep { $_ eq ":hireswallclock" } @_) {
@_ = grep { $_ ne ":hireswallclock" } @_;
+ local $^W=0;
*mytime = $hirestime if defined $hirestime;
}
Benchmark->export_to_level(1, $class, @_);
}
+our($Debug, $Min_Count, $Min_CPU, $Default_Format, $Default_Style,
+ %_Usage, %Cache, $Do_Cache);
+
sub init {
- $debug = 0;
- $min_count = 4;
- $min_cpu = 0.4;
- $defaultfmt = '5.2f';
- $defaultstyle = 'auto';
+ $Debug = 0;
+ $Min_Count = 4;
+ $Min_CPU = 0.4;
+ $Default_Format = '5.2f';
+ $Default_Style = 'auto';
# The cache can cause a slight loss of sys time accuracy. If a
# user does many tests (>10) with *very* large counts (>10000)
# or works on a very slow machine the cache may be useful.
- &disablecache;
- &clearallcache;
+ disablecache();
+ clearallcache();
}
-sub debug { $debug = ($_[1] != 0); }
+sub debug { $Debug = ($_[1] != 0); }
+
+sub usage {
+ my $calling_sub = (caller(1))[3];
+ $calling_sub =~ s/^Benchmark:://;
+ return $_Usage{$calling_sub} || '';
+}
# The cache needs two branches: 's' for strings and 'c' for code. The
-# emtpy loop is different in these two cases.
-sub clearcache { delete $cache{"$_[0]c"}; delete $cache{"$_[0]s"}; }
-sub clearallcache { %cache = (); }
-sub enablecache { $cache = 1; }
-sub disablecache { $cache = 0; }
+# empty loop is different in these two cases.
+
+$_Usage{clearcache} = <<'USAGE';
+usage: clearcache($count);
+USAGE
+
+sub clearcache {
+ die usage unless @_ == 1;
+ delete $Cache{"$_[0]c"}; delete $Cache{"$_[0]s"};
+}
+
+$_Usage{clearallcache} = <<'USAGE';
+usage: clearallcache();
+USAGE
+
+sub clearallcache {
+ die usage if @_;
+ %Cache = ();
+}
+
+$_Usage{enablecache} = <<'USAGE';
+usage: enablecache();
+USAGE
+
+sub enablecache {
+ die usage if @_;
+ $Do_Cache = 1;
+}
+
+$_Usage{disablecache} = <<'USAGE';
+usage: disablecache();
+USAGE
+
+sub disablecache {
+ die usage if @_;
+ $Do_Cache = 0;
+}
+
# --- Functions to process the 'time' data type
sub new { my @t = (mytime, times, @_ == 2 ? $_[1] : 0);
- print STDERR "new=@t\n" if $debug;
+ print STDERR "new=@t\n" if $Debug;
bless \@t; }
sub cpu_p { my($r,$pu,$ps,$cu,$cs) = @{$_[0]}; $pu+$ps ; }
sub real { my($r,$pu,$ps,$cu,$cs) = @{$_[0]}; $r ; }
sub iters { $_[0]->[5] ; }
+
+$_Usage{timediff} = <<'USAGE';
+usage: $result_diff = timediff($result1, $result2);
+USAGE
+
sub timediff {
my($a, $b) = @_;
+
+ die usage unless ref $a and ref $b;
+
my @r;
for (my $i=0; $i < @$a; ++$i) {
push(@r, $a->[$i] - $b->[$i]);
}
+ #die "Bad timediff(): ($r[1] + $r[2]) <= 0 (@$a[1,2]|@$b[1,2])\n"
+ # if ($r[1] + $r[2]) < 0;
bless \@r;
}
+$_Usage{timesum} = <<'USAGE';
+usage: $sum = timesum($result1, $result2);
+USAGE
+
sub timesum {
- my($a, $b) = @_;
- my @r;
- for (my $i=0; $i < @$a; ++$i) {
+ my($a, $b) = @_;
+
+ die usage unless ref $a and ref $b;
+
+ my @r;
+ for (my $i=0; $i < @$a; ++$i) {
push(@r, $a->[$i] + $b->[$i]);
- }
- bless \@r;
+ }
+ bless \@r;
}
+
+$_Usage{timestr} = <<'USAGE';
+usage: $formatted_result = timestr($result1);
+USAGE
+
sub timestr {
my($tr, $style, $f) = @_;
+
+ die usage unless ref $tr;
+
my @t = @$tr;
warn "bad time value (@t)" unless @t==6;
my($r, $pu, $ps, $cu, $cs, $n) = @t;
my($pt, $ct, $tt) = ($tr->cpu_p, $tr->cpu_c, $tr->cpu_a);
- $f = $defaultfmt unless defined $f;
+ $f = $Default_Format unless defined $f;
# format a time in the required style, other formats may be added here
- $style ||= $defaultstyle;
+ $style ||= $Default_Style;
return '' if $style eq 'none';
$style = ($ct>0) ? 'all' : 'noc' if $style eq 'auto';
my $s = "@t $style"; # default for unknown style
$r,$pu,$ps,$pt) if $style eq 'noc';
$s=sprintf("$w wallclock secs (%$f cusr + %$f csys = %$f CPU)",
$r,$cu,$cs,$ct) if $style eq 'nop';
- $s .= sprintf(" @ %$f/s (n=$n)", $n / ( $pu + $ps )) if $n && $pu+$ps;
+ $s .= sprintf(" @ %$f/s (n=$n)", $n / ( $style eq 'nop' ? $cu + $cs : $pu + $ps ))
+ if $n && ($style eq 'nop' ? $cu+$cs : $pu+$ps);
$s;
}
sub timedebug {
my($msg, $t) = @_;
- print STDERR "$msg",timestr($t),"\n" if $debug;
+ print STDERR "$msg",timestr($t),"\n" if $Debug;
}
# --- Functions implementing low-level support for timing loops
+$_Usage{runloop} = <<'USAGE';
+usage: runloop($number, [$string | $coderef])
+USAGE
+
sub runloop {
my($n, $c) = @_;
$n+=0; # force numeric now, so garbage won't creep into the eval
croak "negative loopcount $n" if $n<0;
- confess "Usage: runloop(number, [string | coderef])" unless defined $c;
+ confess usage unless defined $c;
my($t0, $t1, $td); # before, after, difference
# find package of caller so we can execute code there
$subref = _doeval($subcode);
}
croak "runloop unable to compile '$c': $@\ncode: $subcode\n" if $@;
- print STDERR "runloop $n '$subcode'\n" if $debug;
+ print STDERR "runloop $n '$subcode'\n" if $Debug;
# Wait for the user timer to tick. This makes the error range more like
# -0.01, +0. If we don't wait, then it's more like -0.01, +0.01. This
# &runloop a lot, and thus reduce additive errors.
my $tbase = Benchmark->new(0)->[1];
while ( ( $t0 = Benchmark->new(0) )->[1] == $tbase ) {} ;
- &$subref;
+ $subref->();
$t1 = Benchmark->new($n);
$td = &timediff($t1, $t0);
timedebug("runloop:",$td);
$td;
}
+$_Usage{timeit} = <<'USAGE';
+usage: $result = timeit($count, 'code' ); or
+ $result = timeit($count, sub { code } );
+USAGE
sub timeit {
my($n, $code) = @_;
my($wn, $wc, $wd);
- printf STDERR "timeit $n $code\n" if $debug;
+ die usage unless defined $code and
+ (!ref $code or ref $code eq 'CODE');
+
+ printf STDERR "timeit $n $code\n" if $Debug;
my $cache_key = $n . ( ref( $code ) ? 'c' : 's' );
- if ($cache && exists $cache{$cache_key} ) {
- $wn = $cache{$cache_key};
+ if ($Do_Cache && exists $Cache{$cache_key} ) {
+ $wn = $Cache{$cache_key};
} else {
$wn = &runloop($n, ref( $code ) ? sub { } : '' );
# Can't let our baseline have any iterations, or they get subtracted
# out of the result.
$wn->[5] = 0;
- $cache{$cache_key} = $wn;
+ $Cache{$cache_key} = $wn;
}
$wc = &runloop($n, $code);
my $min_for = 0.1;
+$_Usage{countit} = <<'USAGE';
+usage: $result = countit($time, 'code' ); or
+ $result = countit($time, sub { code } );
+USAGE
+
sub countit {
my ( $tmax, $code ) = @_;
+ die usage unless @_;
+
if ( not defined $tmax or $tmax == 0 ) {
$tmax = $default_for;
} elsif ( $tmax < 0 ) {
my ($n, $tc);
# First find the minimum $n that gives a significant timing.
+ my $zeros=0;
for ($n = 1; ; $n *= 2 ) {
my $td = timeit($n, $code);
$tc = $td->[1] + $td->[2];
+ if ( $tc <= 0 and $n > 1024 ) {
+ ++$zeros > 16
+ and die "Timing is consistently zero in estimation loop, cannot benchmark. N=$n\n";
+ } else {
+ $zeros = 0;
+ }
last if $tc > 0.1;
}
# with stable times and avoiding extra timeit()s is nice for
# accuracy's sake.
$n = int( $n * ( 1.05 * $tmax / $tc ) );
-
+ $zeros=0;
while () {
my $td = timeit($n, $code);
$ntot += $n;
$cstot += $td->[4];
$ttot = $utot + $stot;
last if $ttot >= $tmax;
-
+ if ( $ttot <= 0 ) {
+ ++$zeros > 16
+ and die "Timing is consistently zero, cannot benchmark. N=$n\n";
+ } else {
+ $zeros = 0;
+ }
$ttot = 0.01 if $ttot < 0.01;
my $r = $tmax / $ttot - 1; # Linear approximation.
$n = int( $r * $ntot );
return $n == 0 ? $default_for : $n < 0 ? -$n : undef;
}
+$_Usage{timethis} = <<'USAGE';
+usage: $result = timethis($time, 'code' ); or
+ $result = timethis($time, sub { code } );
+USAGE
+
sub timethis{
my($n, $code, $title, $style) = @_;
- my($t, $for, $forn);
+ my($t, $forn);
+
+ die usage unless defined $code and
+ (!ref $code or ref $code eq 'CODE');
if ( $n > 0 ) {
croak "non-integer loopcount $n, stopped" if int($n)<$n;
$t = timeit($n, $code);
$title = "timethis $n" unless defined $title;
} else {
- $fort = n_to_for( $n );
+ my $fort = n_to_for( $n );
$t = countit( $fort, $code );
$title = "timethis for $fort" unless defined $title;
$forn = $t->[-1];
local $| = 1;
$style = "" unless defined $style;
printf("%10s: ", $title) unless $style eq 'none';
- print timestr($t, $style, $defaultfmt),"\n" unless $style eq 'none';
+ print timestr($t, $style, $Default_Format),"\n" unless $style eq 'none';
$n = $forn if defined $forn;
# Don't assume that your benchmark is ok simply because
# you don't get this warning!
print " (warning: too few iterations for a reliable count)\n"
- if $n < $min_count
+ if $n < $Min_Count
|| ($t->real < 1 && $n < 1000)
- || $t->cpu_a < $min_cpu;
+ || $t->cpu_a < $Min_CPU;
$t;
}
+
+$_Usage{timethese} = <<'USAGE';
+usage: timethese($count, { Name1 => 'code1', ... }); or
+ timethese($count, { Name1 => sub { code1 }, ... });
+USAGE
+
sub timethese{
my($n, $alt, $style) = @_;
- die "usage: timethese(count, { 'Name1'=>'code1', ... }\n"
- unless ref $alt eq HASH;
+ die usage unless ref $alt eq 'HASH';
+
my @names = sort keys %$alt;
$style = "" unless defined $style;
print "Benchmark: " unless $style eq 'none';
return \%results;
}
+
+$_Usage{cmpthese} = <<'USAGE';
+usage: cmpthese($count, { Name1 => 'code1', ... }); or
+ cmpthese($count, { Name1 => sub { code1 }, ... }); or
+ cmpthese($result, $style);
+USAGE
+
sub cmpthese{
- my ($results, $style) =
- ref $_ [0] ? @_
- : (timethese (@_ [0, 1], @_ > 2 ? $_ [2] : "none"), $_ [2]);
+ my ($results, $style);
+
+ # $count can be a blessed object.
+ if ( ref $_[0] eq 'HASH' ) {
+ ($results, $style) = @_;
+ }
+ else {
+ my($count, $code) = @_[0,1];
+ $style = $_[2] if defined $_[2];
+
+ die usage unless ref $code eq 'HASH';
+
+ $results = timethese($count, $code, ($style || "none"));
+ }
$style = "" unless defined $style;
for (@vals) {
# The epsilon fudge here is to prevent div by 0. Since clock
# resolutions are much larger, it's below the noise floor.
- my $rate = $_->[6] / ( $_->[2] + $_->[3] + 0.000000000000001 );
+ my $rate = $_->[6] / (( $style eq 'nop' ? $_->[4] + $_->[5]
+ : $_->[2] + $_->[3]) + 0.000000000000001 );
$_->[7] = $rate;
}
@vals = sort { $a->[7] <=> $b->[7] } @vals;
# If more than half of the rates are greater than one...
- my $display_as_rate = $vals[$#vals>>1]->[7] > 1;
+ my $display_as_rate = @vals ? ($vals[$#vals>>1]->[7] > 1) : 0;
my @rows;
my @col_widths;
my $row_rate = $row_val->[7];
# We assume that we'll never get a 0 rate.
- my $a = $display_as_rate ? $row_rate : 1 / $row_rate;
+ my $rate = $display_as_rate ? $row_rate : 1 / $row_rate;
# Only give a few decimal places before switching to sci. notation,
# since the results aren't usually that accurate anyway.
my $format =
- $a >= 100 ?
+ $rate >= 100 ?
"%0.0f" :
- $a >= 10 ?
+ $rate >= 10 ?
"%0.1f" :
- $a >= 1 ?
+ $rate >= 1 ?
"%0.2f" :
- $a >= 0.1 ?
+ $rate >= 0.1 ?
"%0.3f" :
"%0.2e";
$format .= "/s"
if $display_as_rate;
- # Using $b here due to optimizing bug in _58 through _61
- my $b = sprintf( $format, $a );
- push @row, $b;
- $col_widths[1] = length( $b )
- if length( $b ) > $col_widths[1];
+
+ my $formatted_rate = sprintf( $format, $rate );
+ push @row, $formatted_rate;
+ $col_widths[1] = length( $formatted_rate )
+ if length( $formatted_rate ) > $col_widths[1];
# Columns 2..N = performance ratios
my $skip_rest = 0;
https://perl5.git.perl.org / perl5.git / blobdiff