[Bug testsuite/31290] gcc.c-torture/execute/920612-1.c and gcc.c-torture/execute/920711-1.c depend on signed overflow being defined

[Bug testsuite/31290] gcc.c-torture/execute/920612-1.c and gcc.c-torture/execute/920711-1.c depend on signed overflow being defined

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=31290

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|UNCONFIRMED                 |RESOLVED
   Target Milestone|---                         |4.7.0

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
gcc.c-torture/execute/920612-1.c was fixed with r0-109082.
gcc.c-torture/execute/920711-1.c was fixed with r0-109083.

So closing as fixed.

[Bug testsuite/31290] gcc.c-torture/execute/920612-1.c and gcc.c-torture/execute/920711-1.c depend on signed overflow being defined

[pkk at spth dot de]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=31290

Philipp <pkk at spth dot de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pkk at spth dot de

--- Comment #4 from Philipp <pkk at spth dot de> ---
I disagree about this bug for gcc.c-torture/execute/920711-1.c. I think there
was no bug here, and the "fix" actually introduced a problem, and should be
reverted:

I do not see signed integer overflow here:

f (0x80000000L)

Unless explicitly noted otherwise, all section numbers in the following refer
to the C2X standard draft N3047.

The 0x80000000L might be outside the range of long (i.e. on systems where long
has 32 bits). In that case, it will have type long long, as per the table in
section 6.4.4.1. Since the argument to f is of type long, it will get converted
to type long. That is not undefined, but implementation-defined according to
section 6.3.1.3.
By section 4.5 of the GCC manual
https://gcc.gnu.org/onlinedocs/gcc/Integers-implementation.html#Integers-implementation
the result of the conversion is -1.

[Bug testsuite/31290] gcc.c-torture/execute/920612-1.c and gcc.c-torture/execute/920711-1.c depend on signed overflow being defined

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=31290

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Philipp from comment #4)
> I disagree about this bug for gcc.c-torture/execute/920711-1.c. I think
> there was no bug here, and the "fix" actually introduced a problem, and
> should be reverted:

No there is a bug.
int f(long a){return (--a > 0);}
int main(){if(f(0x80000000L)==0)abort();exit(0);}


0x80000000L is still 0x80000000 bitwise. so we get 0x80000000L - 1 which is
undefined if long is 32bits. If long is 64bits then 0x80000000L is still that
(and no implementation defined casting) and not undefined there.

> By section 4.5 of the GCC manual https://gcc.gnu.org/onlinedocs/gcc/Integers-implementation.html#Integers-implementation the result of the conversion is -1.

No it is INT_MIN (for 32bit long):
For conversion to a type of width N, the value is reduced modulo 2^N to be
within range of the type; no signal is raised.

That means 0x80000000ULL will be 0x80000000L (if long is 32bits and long long
is 64bit).

[Bug testsuite/31290] gcc.c-torture/execute/920612-1.c and gcc.c-torture/execute/920711-1.c depend on signed overflow being defined

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2007-03-21 14:56 -------
gcc.c-torture/execute/920711-1.c has exactly the same issue.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|gcc.c-                      |gcc.c-
                   |torture/execute/920612-1.c  |torture/execute/920612-1.c
                   |depends on overflow being   |and gcc.c-
                   |defined                     |torture/execute/920711-1.c
                   |                            |depend on signed overflow
                   |                            |being defined


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=31290