[Bug c/10735] ICE on incorrect type conversion with vector

[Bug c/10735] ICE on incorrect type conversion with vector

[shurick@pisem.net]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735



------- Additional Comments From shurick@pisem.net  2003-06-02 14:30 -------
Subject: Re:  ICE on incorrect type conversion with vector

pinskia@physics.uc.edu wrote:

>PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.
>
>http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735
>
>
>pinskia@physics.uc.edu changed:
>
>           What    |Removed                     |Added
>----------------------------------------------------------------------------
>            Summary|Crash on incorrect type     |ICE on incorrect type
>                   |conversion                  |conversion with vector
>
>  
>
Really this type conversion is right. The second example (with warning 
instead of ICE) does not contain conversion by a mistake (my fault).

Correct example is:
typedef int __v4hi __attribute__ ((__mode__ (__V4HI__)));

__v4hi test3;

void test()
{
        long long test4;
        test4 = (long long)test3; // <== "(long long)" was skipped
        if (test4 != 0)
                return;
}

So, it should be called like "ICE on correct type conversion with vector".





------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

[Bug c/10735] ICE on type conversion with vector

[dhazeghi@yahoo.com]

PLEASE REPLY TO gcc-bugzilla@gcc.gnu.org ONLY, *NOT* gcc-bugs@gcc.gnu.org.

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735


dhazeghi@yahoo.com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|dhazeghi@yahoo.com          |
  GCC build triplet|                            |i686-pc-linux-gnu
   GCC host triplet|                            |i686-pc-linux-gnu
 GCC target triplet|                            |i686-linux-pc-gnu
   Last reconfirmed|0000-00-00 00:00:00         |2003-06-05 03:49:50
               date|                            |


------- Additional Comments From dhazeghi@yahoo.com  2003-06-05 03:49 -------
Reconfirmed with gcc 3.3 branch and mainline (20030604).



------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.

[Bug c/10735] ICE on type conversion with vector

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2003-12-28 06:20 -------
The problem is that vector types do not have a min or a max value.
I think I have a patch for this.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735

[Bug c/10735] ICE on type conversion with vector

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2003-12-28 14:29 -------
Even with the fix I had in mind, it still ICEs.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735

[Bug c/10735] ICE on type conversion with vector

[bonzini at gcc dot gnu dot org]

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735

[Bug c/10735] ICE on correct type conversion with vector

[bonzini at gcc dot gnu dot org]

------- Additional Comments From bonzini at gcc dot gnu dot org  2004-10-14 11:51 -------
http://gcc.gnu.org/ml/gcc-patches/2004-09/msg02540.html

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735

[Bug c/10735] ICE on correct type conversion with vector

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-07-21 00:46 -------
I cannot reproduce this at all anymore, maybe I am doing something wrong.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735

[Bug c/10735] ICE on correct type conversion with vector

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2005-07-21 09:02 -------
The testcase from comment #2 still crashes on mainline:

bug.c:1: warning: specifying vector types with __attribute__ ((mode)) is
deprecated
bug.c:1: warning: use __attribute__ ((vector_size)) instead
bug.c: In function 'test':
bug.c:7: internal compiler error: tree check: expected integer_type or
enumeral_type or boolean_type or char_type or real_type, have vector_type in
shorten_compare, at c-common.c:2002
Please submit a full bug report, [etc.]

Well, the testcase could be improved a little:
__v4hi should actually look like this:
  typedef short __v4hi __attribute__ ((__vector_size__ (8)));
and the if statement has no real effect.

So here's a new proposed testcase:

=====================================================
typedef short v4hi __attribute__((vector_size(8)));

int foo()
{
    v4hi v;
    return (long long)v != 0;
}
=====================================================


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735

[Bug c/10735] ICE on correct type conversion with vector

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2005-08-25 06:02 -------
Fixed on mainline and the 4.0 branch by Paolo Bonzini's
patch for PR23517.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|ASSIGNED                    |RESOLVED
         Resolution|                            |FIXED
   Target Milestone|---                         |4.0.2


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=10735