Description:
------------
the problem is doing bitwise operations with large numbers like larger then 28 bit.

when i do 
negative bitwise operation "~" to 4738698913
in binary
11010011100101100111010100001{29bit}
the number becomes binary
10000000000000000000000000000000{32bit} 

expected:
11100101100011010011000101011110



Reproduce code:
---------------
<TEXTAREA STYLE="height:98%;width:98%">
<?php
//

//I think I found out the solution by using GMP function.
//Code:
function gxor($a,$b)
{
 $a=gmp_init("$a",10);
 $b=gmp_init("$b",10);
 return gmp_strval (gmp_xor ($a, $b),10);
}

echo "gxor(-5799680607,167160) = ".gxor(-5799680607,167160)."\n";
echo "                 as expected -5799843495 \n";
echo "     -5799680607^167160  = " .(-5799680607^167160)."\n";
echo "                 wrong result \n\n";
 
echo "var_dump(-4738698913) = ";
var_dump(-4738698913);
echo "so this number is float\n\n";
 
echo "A decbin              43814   SIMPLE       = ".decbin(43814)."\n\n"; //a
echo "B decbin        -4738698913   NOT WORKING  = ".decbin(-4738698913)."\n\n"; //b 
echo "C decbin         4738698913   SIMPLE       = ".decbin(4738698913)."\n\n";
echo "D decbin abs    -4738698913   WORKING      = ".decbin(abs(-4738698913))."\n\n";
echo "E decbin ~abs   -4738698913   NOT WORKING  = ".decbin(~abs(-4738698913))."\n\n";
echo "F decbin ~abs   -1738698913   WORKING      = ".decbin(~abs(-1738698913))."\n\n";
echo "G decbin ~      -4738698913   64bit expected result: 1111111111111111111111111111111011100101100011010011000101011110\n";
echo "H the 64bit bitwise not is from windows calc.exe\n\n";
echo "from the B and G example i can deduce that php takes the left side of 64 bit integer instead of the right and returns it as 32 bit (just a guess)\n\n";
// 1111111111111111111111111111111011100101100011010011000101011110

echo "php version ".PHP_VERSION;
echo "\n";
echo `cat /proc/version`;
echo "\n";
echo `cat /proc/cpuinfo`;
echo "\n";
?>
</TEXTAREA>

Expected result:
----------------
gxor(-5799680607,167160) = -5799843495
                 as expected -5799843495 
     -5799680607^167160  = -2147316488
                 wrong result 

var_dump(-4738698913) = float(-4738698913)
so this number is float

A decbin              43814   SIMPLE       = 1010101100100110

B decbin        -4738698913   NOT WORKING  = 10000000000000000000000000000000

C decbin         4738698913   SIMPLE       = 11010011100101100111010100001

D decbin abs    -4738698913   WORKING      = 11010011100101100111010100001

E decbin ~abs   -4738698913   NOT WORKING  = 1111111111111111111111111111111

F decbin ~abs   -1738698913   WORKING      = 10011000010111011000111101011110

G decbin ~      -4738698913   64bit expected result: 1111111111111111111111111111111011100101100011010011000101011110
H the 64bit bitwise not is from windows calc.exe

from the B and G example i can deduce that php takes the left side of 64 bit integer instead of the right and returns it as 32 bit (just a guess)


Actual result:
--------------
php version 5.1.6
Linux version 2.6.18-92.1.18.el5PAE (mockbuild@builder16.centos.org) (gcc version 4.1.2 20071124 (Red Hat 4.1.2-42)) #1 SMP Wed Nov 12 10:02:30 EST 2008

processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 23
model name	: Intel(R) Xeon(R) CPU           E5405  @ 2.00GHz
stepping	: 6
cpu MHz		: 1995.120
cache size	: 6144 KB
physical id	: 0
siblings	: 4
core id		: 0
cpu cores	: 4
fdiv_bug	: no
hlt_bug		: no
f00f_bug	: no
coma_bug	: no
fpu		: yes
fpu_exception	: yes
cpuid level	: 10
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc pni monitor ds_cpl vmx tm2 cx16 xtpr lahf_lm
bogomips	: 3992.34