[Bug c/26490] New: tree check fail on valid C code

[Bug c/26490] New: tree check fail on valid C code

[dcb314 at hotmail dot com]

I just tried to compile Suse Linux package graphviz-2.6-9 with a recent
GNU C++ compiler version 4.2 snapshot 20060225. 

The compiler snapshot said

~dcb/gnu/42-20060225/results/bin/gcc -DHAVE_CONFIG_H -I. -I. -I../.. -I../..
-I../../lib/common -I../../lib/gvc -I../../lib/pack -I../../lib/pathplan
-I../../lib/graph -I../../lib/cdt -O2 -g -fmessage-length=0 -D_FORTIFY_SOURCE=2
-W -Wall -Wno-unused-parameter -Wno-unknown-pragmas -Wstrict-prototypes
-Wpointer-arith -ffast-math -Wno-unused-parameter -Wno-unknown-pragmas
-Wstrict-prototypes   -c adjust.c  -fPIC -DPIC
adjust.c: In function "countOverlap":
adjust.c:317: internal compiler error: tree check: expected ssa_name, have
type_memory_tag in verify_ssa, at tree-ssa.c:735
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.

Preprocessed source code attached.  -O2 Flags required.


-- 
           Summary: tree check fail on valid C code
           Product: gcc
           Version: 4.2.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: dcb314 at hotmail dot com
  GCC host triplet: x86_64-suse-linux


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26490

[Bug c/26490] tree check fail on valid C code

[dcb314 at hotmail dot com]

------- Comment #1 from dcb314 at hotmail dot com  2006-02-27 23:10 -------
Created an attachment (id=10928)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=10928&action=view)
C source code


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26490

[Bug tree-optimization/26490] tree check fail on valid C code

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2006-02-27 23:18 -------
Reducing.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26490

[Bug tree-optimization/26490] [4.2 Regression] tree check fail on valid C code

[pinskia at gcc dot gnu dot org]

------- Comment #3 from pinskia at gcc dot gnu dot org  2006-02-28 00:08 -------
Reduced testcase:
extern int nsites;
typedef struct Site {
} Site;
typedef struct ptitem {
  Site site;
  int overlaps;
} Info_t;
extern Info_t *nodeInfo;
int countOverlap(int iter)
{
  int i, j;
  Info_t *ip = nodeInfo;
  Info_t *jp;
  for (i = 0; i < nsites - 1; i++)
    {
      jp = ip + 1;
      if (polyOverlap (ip->site))
        jp->overlaps = 1;
      ip++;
    }
}


-- 

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         |2006-02-28 00:08:28
               date|                            |
            Summary|tree check fail on valid C  |[4.2 Regression] tree check
                   |code                        |fail on valid C code
   Target Milestone|---                         |4.2.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26490

[Bug tree-optimization/26490] [4.2 Regression] tree check fail on valid C code

[rguenth at gcc dot gnu dot org]

------- Comment #4 from rguenth at gcc dot gnu dot org  2006-02-28 10:10 -------
Happens in forwprop, after it we have

;; basic block 3, loop depth 1, count 0
;; prev block 2, next block 4
;; pred:       2 [90.0%]  (true,exec) 5 [90.0%]  (dfs_back,true,exec)
;; succ:       4 [50.0%]  (true,exec) 5 [50.0%]  (false,exec)
# ivtmp.31_1 = PHI <ip_4(2), ivtmp.31_2(5)>;
# TMT.5_34 = PHI <TMT.5_17(2), TMT.5_20(5)>;
# nodeInfo_33 = PHI <nodeInfo_3(2), nodeInfo_19(5)>;
# nsites_32 = PHI <nsites_6(2), nsites_14(5)>;
# i_30 = PHI <0(2), i_12(5)>;
<L13>:;
#   nsites_18 = V_MAY_DEF <nsites_32>;
#   nodeInfo_19 = V_MAY_DEF <nodeInfo_33>;
#   VUSE <TMT.5>;
D.1539_10 = polyOverlap (ivtmp.31_1->site);
if (D.1539_10 != 0) goto <L1>; else goto <L2>;

where the VUSE missed renaming.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26490

[Bug tree-optimization/26490] [4.2 Regression] tree check fail on valid C code

[pinskia at gcc dot gnu dot org]

------- Comment #5 from pinskia at gcc dot gnu dot org  2006-02-28 18:11 -------
(In reply to comment #4)
> where the VUSE missed renaming.
Except this is changing from a V_MAY_DEF but is forwardprop even touching this
statement, it does nothing to it.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26490

[Bug tree-optimization/26490] [4.2 Regression] ICE in verify_ssa, with structs pointers and globals

[reichelt at gcc dot gnu dot org]

------- Comment #6 from reichelt at gcc dot gnu dot org  2006-03-26 12:27 -------
Even shorter testcase:

===============================================
struct A
{
  int i;
  struct {} s;
};

int N;

void foo(struct A *p)
{
  while (N && bar((++p)->s))
    p->i = 0;
}
===============================================


-- 

reichelt at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |reichelt at gcc dot gnu dot
                   |                            |org
           Keywords|                            |monitored


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26490

[Bug tree-optimization/26490] [4.2 Regression] ICE in verify_ssa, with structs pointers and globals

[pinskia at gcc dot gnu dot org]

------- Comment #7 from pinskia at gcc dot gnu dot org  2006-04-13 16:35 -------
*** Bug 27103 has been marked as a duplicate of this bug. ***


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |debian-gcc at lists dot
                   |                            |debian dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26490

[Bug tree-optimization/26490] [4.2 Regression] ICE in verify_ssa, with structs pointers and globals

[reichelt at gcc dot gnu dot org]

------- Comment #8 from reichelt at gcc dot gnu dot org  2006-05-09 08:47 -------
Fixed by Dan's patch for PR27093.


-- 

reichelt at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=26490