[Bug tree-optimization/56928] [4.6/4.7/4.8/4.9 Regression] false positive when using -Wstrict-overflow

[pinskia at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=56928



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



           What    |Removed                     |Added

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

           Keywords|                            |diagnostic

             Status|UNCONFIRMED                 |NEW

   Last reconfirmed|                            |2013-04-12

          Component|c                           |tree-optimization

   Target Milestone|---                         |4.6.4

            Summary|false positive when using   |[4.6/4.7/4.8/4.9

                   |-Wstrict-overflow           |Regression] false positive

                   |                            |when using

                   |                            |-Wstrict-overflow

     Ever Confirmed|0                           |1



--- Comment #1 from Andrew Pinski <pinskia at gcc dot gnu.org> 2013-04-12 
05:06:26 UTC ---

Visiting PHI node: dynamic_3 = PHI <0(2), dynamic_2(11)>



    Argument #0 (2 -> 5 executable)

    0

    Value: [0, 0]



    Argument #1 (11 -> 5 executable)

    dynamic_2

    Value: [0, 1]

Meeting

  [0, 0]

and

  [0, 1]

to

  [0, 1]

Found new range for dynamic_3: [0, +INF(OVF)]





I don't see any overflow there or even a way to cause the overflow.



Looks like there is a bad type issue, dynamic is char but it is being treated

as a boolean type internally.