[Bug middle-end/30164] New: Gimplifier does not produce valid gimple for global_vectora = CONSTRUCTOR

[Bug middle-end/30164] New: Gimplifier does not produce valid gimple for global_vectora = CONSTRUCTOR

[pinskia at gcc dot gnu dot org]

Testcase (only with the C++ front-end because there is no other way to produce
this kind of internal IR, reduced from g++.dg/init/vector1.C):
typedef signed char v8qi __attribute__ ((vector_size (8)));
int num (void);
v8qi b = (v8qi){ num (), num (), num (), num (), 6, 6, 6, 6 };


-- 
           Summary: Gimplifier does not produce valid gimple for
                    global_vectora = CONSTRUCTOR
           Product: gcc
           Version: 4.3.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: middle-end
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: pinskia at gcc dot gnu dot org
OtherBugsDependingO 30142
             nThis:


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30164

[Bug middle-end/30164] Gimplifier does not produce valid gimple for global_vectora = CONSTRUCTOR

[pinskia at gcc dot gnu dot org]

------- Comment #1 from pinskia at gcc dot gnu dot org  2006-12-13 15:54 -------
I just exposed this on the mainline with my GIMPLE_REG vector patch for a lot
more cases so I am going to take it.


-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |pinskia at gcc dot gnu dot
                   |dot org                     |org
             Status|UNCONFIRMED                 |ASSIGNED
     Ever Confirmed|0                           |1
   Last reconfirmed|0000-00-00 00:00:00         |2006-12-13 15:54:45
               date|                            |
            Summary|Gimplifier does not produce |Gimplifier does not produce
                   |valid gimple for            |valid gimple for
                   |global_vectora = CONSTRUCTOR|global_vectora = CONSTRUCTOR


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30164

[Bug middle-end/30164] Gimplifier does not produce valid gimple for global_vectora = CONSTRUCTOR

[pinskia at gcc dot gnu dot org]

------- Comment #2 from pinskia at gcc dot gnu dot org  2006-12-14 05:19 -------
Actually there are three seperate issues:
1) for my recent vector patch, it was a c-gimplifier issue with respect of not
setting GIMPLE_REG_P on the decls for the compount literals, this helps with
complex typed variables also.
2) we don't record constant vector constructors as min invarient
3) we still don't gimplifier correctly for the below testcase with the
constructor.

I have a patch for each of those three issues too.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30164

[Bug middle-end/30164] Gimplifier does not produce valid gimple for global_vectora = CONSTRUCTOR

[pinskia at gcc dot gnu dot org]

------- Comment #3 from pinskia at gcc dot gnu dot org  2006-12-22 07:39 -------
(In reply to comment #2)
> Actually there are three seperate issues:
> 1) for my recent vector patch, it was a c-gimplifier issue with respect of not
> setting GIMPLE_REG_P on the decls for the compount literals, this helps with
> complex typed variables also.
I posted a semi correct patch for this but it needs to be improved so I
withdrew it but I am about to post a new patch.

> 2) we don't record constant vector constructors as min invarient
This was fixed via:
http://gcc.gnu.org/ml/gcc-patches/2006-12/msg01366.html

> 3) we still don't gimplifier correctly for the below testcase with the
> constructor.

I am about to post a patch for this.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30164

[Bug middle-end/30164] Gimplifier does not produce valid gimple for global_vectora = CONSTRUCTOR

[rguenth at gcc dot gnu dot org]

------- Comment #4 from rguenth at gcc dot gnu dot org  2008-09-13 16:32 -------
Any news here?  What is the testcase that produces invalid gimple?


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30164

[Bug middle-end/30164] Gimplifier does not produce valid gimple for global_vectora = CONSTRUCTOR

[pinskia at gcc dot gnu dot org]

------- Comment #5 from pinskia at gcc dot gnu dot org  2008-09-14 05:16 -------
(In reply to comment #4)
> Any news here?  What is the testcase that produces invalid gimple?

typedef signed char v8qi __attribute__ ((vector_size (8)));
int num (void);
v8qi b = (v8qi){ num (), num (), num (), num (), 6, 6, 6, 6 };

And as mentioned this is reduced from g++.dg/init/vector1.C.

Also when I say below, I had meant above (where the comment field is got me
confused :)).


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30164

[Bug middle-end/30164] Gimplifier does not produce valid gimple for global_vectora = CONSTRUCTOR

[pinskia at gcc dot gnu dot org]

------- Comment #6 from pinskia at gcc dot gnu dot org  2009-01-01 21:57 -------
This has been fixed with tuples merge.


-- 

pinskia at gcc dot gnu dot org changed:

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=30164