[Bug tree-optimization/50188] New: Optimizer doesn't take into account, that longjmp could lead to loops, which causes illegal code transformations.

[Bug tree-optimization/50188] New: Optimizer doesn't take into account, that longjmp could lead to loops, which causes illegal code transformations.

[michael.v.zolotukhin at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50188

             Bug #: 50188
           Summary: Optimizer doesn't take into account, that longjmp
                    could lead to loops, which causes illegal code
                    transformations.
    Classification: Unclassified
           Product: gcc
           Version: 4.7.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned@gcc.gnu.org
        ReportedBy: michael.v.zolotukhin@gmail.com


Created attachment 25104
  --> http://gcc.gnu.org/bugzilla/attachment.cgi?id=25104
Bug reproducer.

Setjmp/longjmp could form loops, like following:
  int i = 0;
  (void) setjmp (env);
  i++;
  if (i < 10)
    longjmp (env, 0);

Optimizer removes check 'if(i<10)' and 'i++' (seemingly, as a dead code). In
this example after such transformations the loop becomes infinite.

[Bug tree-optimization/50188] Optimizer doesn't take into account, that longjmp could lead to loops, which causes illegal code transformations.

[hjl.tools at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50188

H.J. Lu <hjl.tools at gmail dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2011-08-25
     Ever Confirmed|0                           |1
              Build|4.7.0 20110801, 4.5.1       |
                   |20100924 (Red Hat 4.5.1-4)  |

--- Comment #1 from H.J. Lu <hjl.tools at gmail dot com> 2011-08-25 17:00:07 UTC ---
It also fails with GCC 4.1.2.

[Bug tree-optimization/50188] Optimizer doesn't take into account, that longjmp could lead to loops, which causes illegal code transformations.

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50188

--- Comment #2 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-08-25 18:33:44 UTC ---
IIRC the variables need to be marked as volatile if you want them to be correct
over setjmp/longjmp.

[Bug tree-optimization/50188] Optimizer doesn't take into account, that longjmp could lead to loops, which causes illegal code transformations.

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50188

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |INVALID

--- Comment #3 from Andrew Pinski <pinskia at gcc dot gnu.org> 2011-08-25 18:43:54 UTC ---
POSIX says this is correct behavior:
http://pubs.opengroup.org/onlinepubs/7908799/xsh/longjmp.html

[Bug tree-optimization/50188] Optimizer doesn't take into account, that longjmp could lead to loops, which causes illegal code transformations.

[michael.v.zolotukhin at gmail dot com]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50188

--- Comment #4 from Michael Zolotukhin <michael.v.zolotukhin at gmail dot com> 2011-08-25 20:21:10 UTC ---
If I understand standard correctly, in this case behavior isn't undefined. Am I
right?
If so, then if behavior of optimized code (loop is infinite) is correct,
behavior of not optimized code isn't.

[Bug tree-optimization/50188] Optimizer doesn't take into account, that longjmp could lead to loops, which causes illegal code transformations.

[jakub at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50188

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jakub at gcc dot gnu.org

--- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-08-26 11:28:01 UTC ---
No, you misunderstood it.  The values of non-volatile variables (meeting the
other conditions in POSIX) are indeterminate, so it is the testcase that is
wrong.