[Bug tree-optimization/17635] New: [4.0 regression] ICE in verify_ssa: type mismatch

[Bug tree-optimization/17635] New: [4.0 regression] ICE in verify_ssa: type mismatch

[bangerth at dealii dot org]

Here's another verify_ssa ICE: 
 
/home/bangerth/bin/gcc-4.0-pre/bin/c++ -O3 -funroll-loops -funroll-all-loops 
-msse2 -mtune=pentium4 -march=pentium4 -mfpmath=sse -ftree-vectorize -c 
tria.cc 
x.ii: In member function `unsigned int 
Triangulation<dim>::max_adjacent_cells() const [with int dim = 3]': 
x.ii:44743: error: Type mismatch between an SSA_NAME and its symbol. 
x.ii:44743: error: Missing definition 
for SSA_NAME: D.193000_390in statement: 
D.193000_402 = PHI <D.193000_195(62), D.193000_390(26)>; 
x.ii:44743: internal compiler error: verify_ssa failed. 
Please submit a full bug report, 
with preprocessed source if appropriate. 
See <URL:http://gcc.gnu.org/bugs.html> for instructions. 
 
W.

-- 
           Summary: [4.0 regression] ICE in verify_ssa: type mismatch
           Product: gcc
           Version: 4.0.0
            Status: UNCONFIRMED
          Keywords: ice-on-valid-code
          Severity: normal
          Priority: P2
         Component: tree-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: bangerth at dealii dot org
                CC: gcc-bugs at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[bangerth at dealii dot org]

------- Additional Comments From bangerth at dealii dot org  2004-09-23 14:12 -------
Created an attachment (id=7203)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=7203&action=view)
Preprocessed source


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |pinskia at gcc dot gnu dot
                   |                            |org
   Target Milestone|---                         |4.0.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2004-09-23 14:17 -------
I've seen this one, too, but I couldn't get a test case out of it.

Thanks Wolfgang!



-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2004-09-23 14:17:05
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[bangerth at ices dot utexas dot edu]

------- Additional Comments From bangerth at ices dot utexas dot edu  2004-09-23 14:20 -------
Subject: Re:  [4.0 regression] ICE in verify_ssa: type mismatch

> I've seen this one, too, but I couldn't get a test case out
> of it.

Do you want to try to reduce it, or is the testcase in itself enough as is? 
Unfortunately, it takes about 45 seconds before gcc crashes and I don't have 
much time today to track things down...

W.

-------------------------------------------------------------------------
Wolfgang Bangerth              email:            bangerth@ices.utexas.edu
                               www: http://www.ices.utexas.edu/~bangerth/



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[steven at gcc dot gnu dot org]

------- Additional Comments From steven at gcc dot gnu dot org  2004-09-23 14:26 -------
I have tried to reduce instances of this bug and I couldn't do it.

Perhaps Serge Belyshev can do it, he's The Expert on reducing C++
test cases to the bare minimum.  I do hope so, 45s is quite long :-)


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[bangerth at ices dot utexas dot edu]

------- Additional Comments From bangerth at ices dot utexas dot edu  2004-09-23 14:29 -------
Subject: Re:  [4.0 regression] ICE in verify_ssa: type mismatch


> Perhaps Serge Belyshev can do it, he's The Expert on reducing C++
> test cases to the bare minimum.  I do hope so, 45s is quite long :-)

Well, I think I can claim that for myself, too :-)

Anyway, I'm at 5 seconds now, so things go much faster. Hope someone will be 
able to do something with a small testcase then...

W.

-------------------------------------------------------------------------
Wolfgang Bangerth              email:            bangerth@ices.utexas.edu
                               www: http://www.ices.utexas.edu/~bangerth/



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-09-23 14:45 -------
This takes a long time to compile when the compiler is compiled with -O0.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[bangerth at dealii dot org]

------- Additional Comments From bangerth at dealii dot org  2004-09-23 15:36 -------
Here's the smallest I can get it within the constraints of my time: 
---------------------------- 
#include <vector> 
 
template<typename T> 
T max_element(T __first, T __last) { 
  if (__first == __last) 
    return __first; 
  T __result = __first; 
  while (++__first != __last) 
    if (*__result < *__first) 
      __result = __first; 
  return __result; 
} 
 
struct S { 
    std::vector<bool> used; 
}; 
 
struct A { 
    unsigned int foo (); 
    std::vector<S*> s; 
}; 
 
struct Accessor { 
    A *a; 
 
    bool used () { 
      return this->a->s[0]->used[0]; 
    } 
 
    void bar () { 
      this->a->s[0]->used.size(); 
    } 
}; 
 
struct D { 
    bool baz (); 
 
    Accessor accessor; 
 
    void bar () { 
      accessor.bar(); 
      if (this->accessor.used()) 
        return; 
    } 
}; 
 
unsigned int A::foo () { 
  std::vector<int> v; 
  for (D d; d.baz(); d.bar()) +v[0]; 
  return (*max_element (v.begin(), v.end())); 
} 
------------------------------- 
 
g/x> /home/bangerth/bin/gcc-4.0-pre/bin/c++ -O3 -ftree-vectorize -c x.cc 
x.cc: In member function `unsigned int A::foo()': 
x.cc:47: error: Type mismatch between an SSA_NAME and its symbol. 
x.cc:47: error: Missing definition 
for SSA_NAME: D.14109_789in statement: 
D.14109_788 = PHI <D.14109_149(29), D.14109_789(6)>; 
x.cc:47: internal compiler error: verify_ssa failed. 
Please submit a full bug report, 
with preprocessed source if appropriate. 
See <URL:http://gcc.gnu.org/bugs.html> for instructions. 
 
W. 

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-09-23 17:08 -------
The type of the SSA_NAME is NULL_TREE.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dpatel at gcc dot gnu dot
                   |                            |org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[dpatel at apple dot com]

------- Additional Comments From dpatel at apple dot com  2004-09-23 22:06 -------
Subject: Re:  [4.0 regression] ICE in verify_ssa: type mismatch

I am not able to reproduce this on powerpc-darwin using one week old 
FSF mainline sources.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-09-23 22:29 -------
Note I could reproduce this with a cross compiler to x86-linux (from powerpc-darwin).

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
 GCC target triplet|                            |i?86-pc-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2004-09-24 00:08 -------
Here's a reduced testcase:

========================
void foo(int i)
{
  while (1)
    if (i) ++i;
}
========================


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |monitored


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2004-09-24 00:11 -------
Just compile with "-O -ftree-vectorize" to see the failure with
the code snippet of the previous comment.
The failure happens with the C and C++ frontend.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-09-24 00:18 -------
And I can reproduce it with todays compiler on powerpc-apple-darwin with -O1 -ftree-vectorize.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[dpatel at apple dot com]

------- Additional Comments From dpatel at apple dot com  2004-09-24 00:51 -------
Subject: Re:  [4.0 regression] ICE in verify_ssa: type mismatch


On Sep 23, 2004, at 5:18 PM, pinskia at gcc dot gnu dot org wrote:

>
> ------- Additional Comments From pinskia at gcc dot gnu dot org  
> 2004-09-24 00:18 -------
> And I can reproduce it with todays compiler on powerpc-apple-darwin 
> with -O1 -ftree-vectorize.

I am able to reproduce it after I updated my sources. I have fix that I 
will test overnight.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-09-24 17:30 -------
Patch here: <http://gcc.gnu.org/ml/gcc-patches/2004-09/msg02562.html>.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |patch


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-01 18:08 -------
Okay I think I know what is going on with this one and PR 17757, they are both related to the same 
bug.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |pinskia at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-01 18:09 -------
I was wrong (in sense it is related to but not the same) but I get a different error wth the fix for PR 
17757:
pr17635.c:2: error: Found an SSA_NAME that had been released into the free pool
pr17635.c:2: internal compiler error: Bus error
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|pinskia at gcc dot gnu dot  |unassigned at gcc dot gnu
                   |org                         |dot org
             Status|ASSIGNED                    |NEW


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-01 18:10 -------
So we are using a SSA_NAME which has been released.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[dpatel at apple dot com]

------- Additional Comments From dpatel at apple dot com  2004-10-06 01:14 -------
Patch:
  http://gcc.gnu.org/ml/gcc-patches/2004-10/msg00390.html

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-14 18:40 -------
Fixed.

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2004-11-13 13:23 -------
The testcase from comment #12 ICEs again:

PR17635.c: In function 'foo':
PR17635.c:2: internal compiler error: in clean_predicate_lists, at
tree-if-conv.c:661

Devang, your patch
http://gcc.gnu.org/ml/gcc-cvs/2004-10/msg00927.html
contains a ChangeLog entry for the testsuite, but the test never got committed.
Could you please commit the testcase?


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|RESOLVED                    |REOPENED
           Keywords|patch                       |
         Resolution|FIXED                       |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2004-11-13 13:54 -------
Devang, your patch
http://gcc.gnu.org/ml/gcc-cvs/2004-11/msg00591.html
is responsible for the new ICE.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[dpatel at apple dot com]

------- Additional Comments From dpatel at apple dot com  2004-11-17 22:13 -------
Subject: Re:  [4.0 regression] ICE in verify_ssa: type mismatch


On Nov 13, 2004, at 5:55 AM, reichelt at gcc dot gnu dot org wrote:

>
> ------- Additional Comments From reichelt at gcc dot gnu dot org  
> 2004-11-13 13:54 -------
> Devang, your patch
> http://gcc.gnu.org/ml/gcc-cvs/2004-11/msg00591.html
> is responsible for the new ICE.

I've fixed this one.
But it seems, I am not out of wood yet.

-
Devang



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635

[Bug tree-optimization/17635] [4.0 regression] ICE in verify_ssa: type mismatch

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2004-11-18 12:46 -------
Fixed.
The testcase was committed, too.


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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17635