[Bug middle-end/17066] New: ICE in do_jump, at dojump.c:275

[Bug middle-end/17066] New: ICE in do_jump, at dojump.c:275

[green at redhat dot com]

Compiling the following with any level of optimization...

void foo(char *ptr, double d, double div)
{
  while (d >= div && ptr[1] != '\0');
}

results in...

bug.i: In function `foo':
bug.i:2: internal compiler error: in do_jump, at dojump.c:275

I believe this is a new regression (within 24hrs or so)

-- 
           Summary: ICE in do_jump, at dojump.c:275
           Product: gcc
           Version: 3.5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: middle-end
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: green at redhat dot com
                CC: gcc-bugs at gcc dot gnu dot org
GCC target triplet: frv-elf


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17066

[Bug middle-end/17066] ICE in do_jump, at dojump.c:275

[green at redhat dot com]

------- Additional Comments From green at redhat dot com  2004-08-17 17:12 -------
The abort() was introduced by Paolo's recent dojump.c patch.

        * dojump.c (do_jump) <COND_EXPR, EQ_EXPR, NE_EXPR,
        TRUTH_ANDIF_EXPR, TRUTH_ORIF_EXPR, COMPOUND_EXPR>:
        Abort on gimplified cases.

Is this just catching a preexisting problem?


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |bonzini at gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17066

[Bug middle-end/17066] [3.5 Regression] ICE in do_jump, at dojump.c:275

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-08-17 18:54 -------
Confirmed, I will note that we buildTRUTH_ORIF_EXPR  in dojump:
            /* If the target doesn't support combined unordered
               compares, decompose into two comparisons.  */
            cmp0 = fold (build2 (tcode1, TREE_TYPE (exp), op0, op1));
            cmp1 = fold (build2 (tcode2, TREE_TYPE (exp), op0, op1));
            exp = build2 (TRUTH_ORIF_EXPR, TREE_TYPE (exp), cmp0, cmp1);


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |critical
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|0000-00-00 00:00:00         |2004-08-17 18:54:44
               date|                            |
            Summary|ICE in do_jump, at          |[3.5 Regression] ICE in
                   |dojump.c:275                |do_jump, at dojump.c:275
   Target Milestone|---                         |3.5.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17066

[Bug middle-end/17066] [3.5 Regression] ICE in do_jump, at dojump.c:275

[zack at gcc dot gnu dot org]

------- Additional Comments From zack at gcc dot gnu dot org  2004-08-18 16:41 -------
Fixed with this patch:

http://gcc.gnu.org/ml/gcc-patches/2004-08/msg01288.html

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17066