Logo Questions Linux Laravel Mysql Ubuntu Git Menu
 

perl integer arithmetic giving floating point answer

The below code mimics actual production code. double quotes are used as the actual data comes from an XML file, parsed using XML:Twig:

#!/usr/bin/perl

use strict;
use warnings;
use diagnostics;
use Devel::Peek;

my $linetotalinclusive = "8458.80" * 1_000_000;

$linetotalinclusive = $linetotalinclusive;

my $c = "7980.00" * 1_000_000;

my $data = $linetotalinclusive - $c;

print Dump $c;

print Dump $linetotalinclusive;

print "$linetotalinclusive - $c = $data \n";

Gives the following result:

SV = PVNV(0x22885f0) at 0x21984f8
  REFCNT = 1
  FLAGS = (PADMY,IOK,NOK,pIOK,pNOK)
  IV = 7980000000
  NV = 7980000000
  PV = 0
SV = PVNV(0x2288650) at 0x21984c8
  REFCNT = 1
  FLAGS = (PADMY,NOK,pIOK,pNOK)
  IV = 8458799999
  NV = 8458800000
  PV = 0

8458800000 - 7980000000 = 478799999.999999

When run on my laptop and on our production server. (the above is from my laptop) However, when I run it on another production machine, it works fine. use integer; on the above code makes it work. But I am unable to do that to production code (easily). So, I was would like to know...

  1. Why is the above happening.
  2. What compile option is missing from the perl interpreter that would fix it.

Further info: This is from the broken machine:

This is perl 5, version 18, subversion 1 (v5.18.1) built for x86_64-linux-thread-multi
perl -MPOSIX -le 'print LONG_MAX'
9223372036854775807

perl -V:[in]vsize
ivsize='8';
nvsize='8';

This is from the machine that works:

This is perl, v5.8.9 built for x86_64-linux-ld

perl -MPOSIX -le 'print LONG_MAX'
9223372036854775807

perl -V:[in]vsize
ivsize='8';
nvsize='16';

This gives the expected answer:

Summary of my perl5 (revision 5 version 8 subversion 9) configuration:
  Platform:
    osname=linux, osvers=2.6.32-431.3.1.el6.x86_64, archname=x86_64-linux-ld
    uname='linux 553291-amon-sul2.firstb2b.net 2.6.32-431.3.1.el6.x86_64 #1 smp sat jan 4 02:04:49 est 2014 x86_64 x86_64 x86_64 gnulinux '
    config_args=''
    hint=recommended, useposix=true, d_sigaction=define
    usethreads=undef use5005threads=undef useithreads=undef usemultiplicity=undef
    useperlio=define d_sfio=undef uselargefiles=define usesocks=undef
    use64bitint=define use64bitall=define uselongdouble=define
    usemymalloc=n, bincompat5005=undef
  Compiler:
    cc='cc', ccflags ='-fno-strict-aliasing -pipe -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -I/usr/include/gdbm',
    optimize='-O2',
    cppflags='-fno-strict-aliasing -pipe -I/usr/local/include -I/usr/include/gdbm'
    ccversion='', gccversion='4.4.7 20120313 (Red Hat 4.4.7-4)', gccosandvers=''
    intsize=4, longsize=8, ptrsize=8, doublesize=8, byteorder=12345678
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
    ivtype='long', ivsize=8, nvtype='long double', nvsize=16, Off_t='off_t', lseeksize=8
    alignbytes=16, prototype=define
  Linker and Libraries:
    ld='cc', ldflags =' -L/usr/local/lib'
    libpth=/usr/local/lib /lib /usr/lib /lib64 /usr/lib64 /usr/local/lib64
    libs=-lnsl -lgdbm -ldb -ldl -lm -lcrypt -lutil -lc
    perllibs=-lnsl -ldl -lm -lcrypt -lutil -lc
    libc=, so=so, useshrplib=false, libperl=libperl.a
    gnulibc_version='2.12'
  Dynamic Linking:
    dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-Wl,-E'
    cccdlflags='-fPIC', lddlflags='-shared -O2 -L/usr/local/lib'


Characteristics of this binary (from libperl): 
  Compile-time options: PERL_MALLOC_WRAP USE_64_BIT_ALL USE_64_BIT_INT
                        USE_FAST_STDIO USE_LARGE_FILES USE_LONG_DOUBLE
                        USE_PERLIO

This one doesn't:

Summary of my perl5 (revision 5 version 8 subversion 8) configuration:
  Platform:
    osname=linux, osvers=2.6.18-194.26.1.el5, archname=x86_64-linux-thread-multi
    uname='linux x86-002.build.bos.redhat.com 2.6.18-194.26.1.el5 #1 smp fri oct 29 14:21:16 edt 2010 x86_64 x86_64 x86_64 gnulinux '
    config_args='-des -Doptimize=-O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic -Dversion=5.8.8 -Dmyhostname=localhost -Dperladmin=root@localhost -Dcc=gcc -Dcf_by=Red Hat, Inc. -Dinstallprefix=/usr -Dprefix=/usr -Dlibpth=/usr/local/lib64 /lib64 /usr/lib64 -Dprivlib=/usr/lib/perl5/5.8.8 -Dsitelib=/usr/lib/perl5/site_perl/5.8.8 -Dvendorlib=/usr/lib/perl5/vendor_perl/5.8.8 -Darchlib=/usr/lib64/perl5/5.8.8/x86_64-linux-thread-multi -Dsitearch=/usr/lib64/perl5/site_perl/5.8.8/x86_64-linux-thread-multi -Dvendorarch=/usr/lib64/perl5/vendor_perl/5.8.8/x86_64-linux-thread-multi -Darchname=x86_64-linux-thread-multi -Dvendorprefix=/usr -Dsiteprefix=/usr -Duseshrplib -Dusethreads -Duseithreads -Duselargefiles -Dd_dosuid -Dd_semctl_semun -Di_db -Ui_ndbm -Di_gdbm -Di_shadow -Di_syslog -Dman3ext=3pm -Duseperlio -Dinstallusrbinperl=n -Ubincompat5005 -Uversiononly -Dpager=/usr/bin/less -isr -Dd_gethostent_r_proto -Ud_endhostent_r_proto -Ud_sethostent_r_proto -Ud_endprotoent_r_proto -Ud_setprotoent_r_proto -Ud_endservent_r_proto -Ud_setservent_r_proto -Dinc_version_list=5.8.7 5.8.6 5.8.5 -Dscriptdir=/usr/bin'
    hint=recommended, useposix=true, d_sigaction=define
    usethreads=define use5005threads=undef useithreads=define usemultiplicity=define
    useperlio=define d_sfio=undef uselargefiles=define usesocks=undef
    use64bitint=define use64bitall=define uselongdouble=undef
    usemymalloc=n, bincompat5005=undef
  Compiler:
    cc='gcc', ccflags ='-D_REENTRANT -D_GNU_SOURCE -fno-strict-aliasing -pipe -Wdeclaration-after-statement -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 -I/usr/include/gdbm',
    optimize='-O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic',
    cppflags='-D_REENTRANT -D_GNU_SOURCE -fno-strict-aliasing -pipe -Wdeclaration-after-statement -I/usr/local/include -I/usr/include/gdbm'
    ccversion='', gccversion='4.1.2 20080704 (Red Hat 4.1.2-50)', gccosandvers=''
    intsize=4, longsize=8, ptrsize=8, doublesize=8, byteorder=12345678
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
    ivtype='long', ivsize=8, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8
    alignbytes=8, prototype=define
  Linker and Libraries:
    ld='gcc', ldflags =''
    libpth=/usr/local/lib64 /lib64 /usr/lib64
    libs=-lresolv -lnsl -lgdbm -ldb -ldl -lm -lcrypt -lutil -lpthread -lc
    perllibs=-lresolv -lnsl -ldl -lm -lcrypt -lutil -lpthread -lc
    libc=, so=so, useshrplib=true, libperl=libperl.so
    gnulibc_version='2.5'
  Dynamic Linking:
    dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-Wl,-E -Wl,-rpath,/usr/lib64/perl5/5.8.8/x86_64-linux-thread-multi/CORE'
    cccdlflags='-fPIC', lddlflags='-shared -O2 -g -pipe -Wall -Wp,-D_FORTIFY_SOURCE=2 -fexceptions -fstack-protector --param=ssp-buffer-size=4 -m64 -mtune=generic'


Characteristics of this binary (from libperl): 
  Compile-time options: MULTIPLICITY PERL_IMPLICIT_CONTEXT
                        PERL_MALLOC_WRAP USE_64_BIT_ALL USE_64_BIT_INT
                        USE_ITHREADS USE_LARGE_FILES USE_PERLIO
                        USE_REENTRANT_API
like image 467
ftumsh Avatar asked Dec 19 '22 09:12

ftumsh


1 Answers

That isn't integer arithmetic. Both equations contain floating point numbers.

my $linetotalinclusive = "8458.80" * 1_000_000;
                         ^^^^^^^^^    
my $c = "7980.00" * 1_000_000;
        ^^^^^^^^^

Sometimes Perl gets smart and notices that the floating point number can be stored as an integer, but it seems to have trouble when there's a string conversion.

Dump 7980.00 + 1_000_000_000_000;
Dump "7980.00" + 1_000_000_000_000;

SV = IV(0x7fd55401c8e0) at 0x7fd55401c8f0
  REFCNT = 1
  FLAGS = (PADTMP,IOK,READONLY,pIOK)
  IV = 1000000007980
SV = NV(0x7fd553831200) at 0x7fd553844990
  REFCNT = 1
  FLAGS = (PADTMP,NOK,READONLY,pNOK)
  NV = 1000000007980

There's other things going on: constant folding. If an expression contains nothing but constants, Perl will often perform the math at compile time. If you run your code through B::Deparse which reconstructs the code from the compiled opcodes, you see that your equations have been turned into constants.

my $linetotalinclusive = 8458799999.9999990463;
my $c = 7980000000;

The first has suffered from floating point error. The second has not.

like image 128
Schwern Avatar answered Jan 06 '23 18:01

Schwern