[Bug middle-end/57859] New: -ftrapv does not trap on signed overflows for struct fields (32-bit mode)

[Bug middle-end/57859] New: -ftrapv does not trap on signed overflows for struct fields (32-bit mode)

[su at cs dot ucdavis.edu]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57859

            Bug ID: 57859
           Summary: -ftrapv does not trap on signed overflows for struct
                    fields (32-bit mode)
           Product: gcc
           Version: 4.9.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
          Assignee: unassigned at gcc dot gnu.org
          Reporter: su at cs dot ucdavis.edu

The following code (which has a signed integer overflow) doesn't trap when
compiled at -O1 or above with -ftrapv on x86_64-linux. This applies to the
current gcc trunk, as well as gcc-4.6, gcc-4.7, and gcc-4.8. 

$ gcc-trunk -v
gcc version 4.9.0 20130708 (experimental) [trunk revision 200751] (GCC) 
$ gcc-trunk -m32 -O0 -ftrapv small.c
$ a.out
Aborted (core dumped)
$ gcc-trunk -m32 -O1 -ftrapv small.c      
$ a.out
$ gcc-4.8 -m32 -O1 -ftrapv small.c
$ a.out
$ gcc-4.7 -m32 -O1 -ftrapv small.c
$ a.out
$ gcc-4.6 -m32 -O1 -ftrapv small.c
$ a.out
$ 


-------------------------------------------


struct S
{
  int f;
}; 

int
main ()
{
  struct S s; 
  for (s.f = 1; s.f > 0; ++s.f)
    ; 
  return 0;
}


-------------------------------------------

The following code does trap in 32-bit mode (but still not in 64-bit): 

int
main ()
{
  int i; 
  for (i = 1; i > 0; ++i)
    ; 
  return 0;
}

[Bug middle-end/57859] -ftrapv does not trap on signed overflows for struct fields (32-bit mode)

[mikpe at it dot uu.se]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57859

--- Comment #1 from Mikael Pettersson <mikpe at it dot uu.se> ---
With -m64: both test cases work with gcc-3.2.3, and fail with every release
from 3.3.6 up to current trunk.

With -m32: the first test case doesn't trap with any release since 3.2.3, the
second test case traps with current trunk and 4.8, but not with any release
from 3.2.3 to 4.7.3.

[Bug middle-end/57859] -ftrapv does not trap on signed overflows for struct fields (32-bit mode)

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=57859

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2013-11-09
     Ever confirmed|0                           |1

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
-ftrapv is very broken right now.  There are some changes going to happen to
add -fundefined=overflow soon which should replace fully -ftrapv to be more
correct implementation.

[Bug middle-end/57859] -ftrapv does not trap on signed overflows for struct fields (32-bit mode)

[mikpelinux at gmail dot com]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=57859

--- Comment #3 from Mikael Pettersson <mikpelinux at gmail dot com> ---
-ftrapv seems to be working a bit better in gcc-6: only the first test case
fails (doesn't report the overflow) at -O1/2 for both -m32/-m64, but all other
combinations of test case, -O0 vs -O1/2, and -m32 vs -m64, seem to work.

However, since -fsanitize=undefined works since 4.9, should -ftrapv be
deprecated and this bug be closed as WONTFIX?