[Bug c++/27533] wrong result after reinterpret_cast from float* to int*

[pluto at agmk dot net]

--- Comment #1 from pluto at agmk dot net  2006-05-10 14:51 ---
you're violating the aliasing rules, so use -fno-strict-aliasing option.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27533

[Bug c++/27533] wrong result after reinterpret_cast from float* to int*

[ulrich dot lauther at siemens dot com]

--- Comment #2 from ulrich dot lauther at siemens dot com  2006-05-10 14:53 
---
Created an attachment (id=11434)
 -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11434action=view)
Complete tiny example exposing the problem

I submit this version that includes stdio.h for readability, expanded version
will follow


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27533

[Bug c++/27533] wrong result after reinterpret_cast from float* to int*

[ulrich dot lauther at siemens dot com]

--- Comment #3 from ulrich dot lauther at siemens dot com  2006-05-10 14:55 
---
Created an attachment (id=11435)
 -- (http://gcc.gnu.org/bugzilla/attachment.cgi?id=11435action=view)
same as before, but stdio.h expanded


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27533

[Bug c++/27533] wrong result after reinterpret_cast from float* to int*

[reichelt at gcc dot gnu dot org]

--- Comment #4 from reichelt at gcc dot gnu dot org  2006-05-10 14:59 
---
As Pawel already pointed out:
Your code is brokan as you are violating the aliasing rules.

Please read about this in the non-bug section of http://gcc.gnu.org/bugs.html :
Casting does not work as expected when optimization is turned on.


*** This bug has been marked as a duplicate of 21920 ***


-- 

reichelt at gcc dot gnu dot org changed:

   What|Removed |Added

 Status|UNCONFIRMED |RESOLVED
 Resolution||DUPLICATE


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27533

[Bug c++/27533] wrong result after reinterpret_cast from float* to int*

[pluto at agmk dot net]

--- Comment #5 from pluto at agmk dot net  2006-05-10 15:26 ---
(In reply to comment #0)
 The code
 
 inline int almost_equal(float a, float b, int maxUlps = 16) {
   int intDiff = *(reinterpret_castint*(a)) -
 *(reinterpret_castint*(b));

if you really need such casting and still use strict-aliasing
you should rewrite almost_equal().

/*inline*/ int almost_equal( float a, float b )
{
int ia;
memcpy( ia, a, sizeof(int) );
int ib;
memcpy( ib, b, sizeof(int) );
return ( ia - ib );
}

compiler will optimize it nicely ;)

 almost_equal(float, float):
   0:   55  push   %ebp
   1:   89 e5   mov%esp,%ebp
   3:   83 ec 10sub$0x10,%esp
   6:   8b 45 08mov0x8(%ebp),%eax===
   9:   2b 45 0csub0xc(%ebp),%eax===
   c:   c9  leave
   d:   c3  ret


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27533