[Bug tree-optimization/26307] New: load PRE creates type mismatches

[Bug tree-optimization/26307] New: load PRE creates type mismatches

[pinskia at gcc dot gnu dot org]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1419 bytes --]

Source code:
int *insert_automaton_decl (void **entry_ptr, int * automaton_decl)
{
  if (*entry_ptr == ((void *)0))
    *entry_ptr = (void *) automaton_decl;
  return (int*) *entry_ptr;
}
-----
Error message:
t1.c: In function ‘insert_automaton_decl’:
t1.c:2: error: statement types mismatch
#   VUSE <TMT.4D.1900_9>;
storetmp.21D.1918_11 = *entry_ptrD.1886_1;

intD.0 *
voidD.38 *
t1.c:2: internal compiler error: verify_stmts failed

------ the tree dump:
  void * prephitmp.22;
  int * storetmp.21;
  int * D.1891;
  void * D.1890;

<bb 2>:
  D.1890_2 = *entry_ptr_1;
  if (D.1890_2 == 0B) goto <L0>; else goto <L3>;

<L3>:;
  goto <bb 4> (<L1>);

<L0>:;
  *entry_ptr_1 = automaton_decl_6;
  storetmp.21_11 = *entry_ptr_1;

  # prephitmp.22_12 = PHI <D.1890_2(5), storetmp.21_11(3)>;
<L1>:;
  D.1890_3 = prephitmp.22_12;
  D.1891_4 = (int *) D.1890_3;
  return D.1891_4;


Why is storetmp.21 an int* and not a void*?

I will look into this more later today.

Reduced genautomata.c


-- 
           Summary: load PRE creates type mismatches
           Product: gcc
           Version: 4.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: tree-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: pinskia at gcc dot gnu dot org
OtherBugsDependingO 22368
             nThis:


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26307

[Bug tree-optimization/26307] load PRE creates type mismatches

[pinskia at gcc dot gnu dot org]

------- Comment #1 from pinskia at gcc dot gnu dot org  2006-08-15 05:51 -------
Confirmed, this should be resolved by my patch at:
http://gcc.gnu.org/ml/gcc-patches/2006-06/msg00179.html


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2006-08-15 05:51:18
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26307

[Bug tree-optimization/26307] load PRE creates type mismatches

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2006-08-15 05:58 -------
That testcase no longer ICEs, PRE selects void* for the variable.
Here is a testcase that currently does fail:
int *automata_list_finish (void ** entry_ptr, int *current_automata_list)
{
  if (*entry_ptr == ((void *)0))
    *entry_ptr = (void *) current_automata_list;
  else
    free_automata_list ();
  return (int *) *entry_ptr;
}


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26307

[Bug tree-optimization/26307] load PRE creates type mismatches

[steven at gcc dot gnu dot org]

------- Comment #3 from steven at gcc dot gnu dot org  2008-11-29 16:25 -------
Is this still an issue?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26307

[Bug tree-optimization/26307] load PRE creates type mismatches

[rguenth at gcc dot gnu dot org]

------- Comment #4 from rguenth at gcc dot gnu dot org  2008-11-30 12:01 -------
PRE now does the correct thing:

  pretmp.13_12 = *entry_ptr_1(D);

<bb 5>:
  # prephitmp.14_14 = PHI <current_automata_list_3(D)(3), pretmp.13_12(4)>
  D.1235_4 = prephitmp.14_14;
  D.1239_5 = (int *) D.1235_4;

with pretmp and the prephitmp being (void *) pointers.


-- 

rguenth at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |4.4.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26307