[Bug tree-optimization/19899] New: ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[Bug tree-optimization/19899] New: ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[commie1 at gmx dot net]

/tmp/gcc4/libexec/gcc/i686-pc-linux-gnu/4.0.0/cc1 -quiet -v render.c -quiet
-dumpbase render.c -msse2 -auxbase render -O -ftree-vectorize -o render.s -version

GNU C version 4.0.0 20050211 (experimental) (i686-pc-linux-gnu)
        compiled by GNU C version 4.0.0 20050211 (experimental).
GGC heuristics: --param ggc-min-expand=30 --param ggc-min-heapsize=4096
render.c: In function 'render':
render.c:2: internal compiler error: tree check: expected real_cst, have
integer_cst in const_binop, at fold-const.c:1490

Configured with: ../gcc/configure --prefix=/tmp/gcc4 --enable-languages=c
--disable-nls


========================================================

void render()
{
  float i;
  for (i=1; i>=0; i--) {}
  i=1;
}

========================================================

-- 
           Summary: ICE: tree check: expected real_cst, have integer_cst in
                    const_binop, at fold-const.c:1490 with -ftree-vectorize
                    -msse2
           Product: gcc
           Version: 4.0.0
            Status: UNCONFIRMED
          Severity: critical
          Priority: P2
         Component: tree-optimization
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: commie1 at gmx dot net
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: i686-pc-linux-gnu
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: i686-pc-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-02-11 13:39 -------
Confirmed.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |dorit at il dot ibm dot com
           Severity|critical                    |normal
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
           Keywords|                            |ice-on-valid-code
   Last reconfirmed|0000-00-00 00:00:00         |2005-02-11 13:39:33
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[dorit at il dot ibm dot com]

------- Additional Comments From dorit at il dot ibm dot com  2005-02-12 11:13 -------
As far as I can see it's related to scalar evolution (nothing gets vectorized, 
the ICE happens during the analysis stage of the vectorizer, when the 
vectorizer calls analyze_scalar_evolution):

(gdb) backtrace
#0  tree_check_failed (node=0x414902e0, file=0x49ab2c "../../gcc/gcc/fold-
const.c", line=1490, function=0x49abcc "const_binop") 
at ../../gcc/gcc/tree.c:5450
#1  0x00205504 in const_binop (code=MULT_EXPR, arg1=0x4148de58, 
arg2=0x414902e0, notrunc=0) at ../../gcc/gcc/fold-const.c:1543
#2  0x0021c1ac in fold (expr=0x4140c894) at ../../gcc/gcc/fold-const.c:7222
#3  0x003bbce4 in add_to_evolution (loop_nb=252, chrec_before=0x414902e0, 
code=1095294552, to_add=0x0) at ../../gcc/gcc/tree-scalar-evolution.c:897
#4  0x003bc644 in follow_ssa_edge_in_rhs (loop=0x413044e0, rhs=0x4148de58, 
halting_phi=0x412d9780, evolution_of_loop=0xbffff580) at ../../gcc/gcc/tree-
scalar-evolution.c:1176
#5  0x003bcff0 in follow_ssa_edge (loop=0x413044e0, def=0x4140c3f0, 
halting_phi=0x412d9780, evolution_of_loop=0xbffff580) at ../../gcc/gcc/tree-
scalar-evolution.c:1446
#6  0x003bd1e0 in analyze_evolution_in_loop (loop_phi_node=0x412d9780, 
init_cond=0x414902e0) at ../../gcc/gcc/tree-scalar-evolution.c:1497
#7  0x003bdf24 in analyze_scalar_evolution_1 (loop=0x413044e0, var=0x4140c534, 
res=0x0) at ../../gcc/gcc/tree-scalar-evolution.c:1798
#8  0x003be054 in analyze_scalar_evolution (loop=0x413044e0, var=0x4140c534) 
at ../../gcc/gcc/tree-scalar-evolution.c:1850
#9  0x00115918 in vect_analyze_scalar_cycles (loop_vinfo=0x41304600) 
at ../../gcc/gcc/tree-flow-inline.h:309
#10 0x00119004 in vect_analyze_loop (loop=0x41304600) at ../../gcc/gcc/tree-
vectorizer.c:5697
#11 0x00119244 in vectorize_loops (loops=0x413011a0) at ../../gcc/gcc/tree-
vectorizer.c:5815
#12 0x00093900 in execute_one_pass (pass=0x595cd0) at ../../gcc/gcc/tree-
optimize.c:528
#13 0x000939e0 in execute_pass_list (pass=0x595cd0) at ../../gcc/gcc/tree-
optimize.c:565
#14 0x000939f8 in execute_pass_list (pass=0x595c00) at ../../gcc/gcc/tree-
optimize.c:566
#15 0x000939f8 in execute_pass_list (pass=0x595520) at ../../gcc/gcc/tree-
optimize.c:566
#16 0x00093cec in tree_rest_of_compilation (fndecl=0x41487d24) 
at ../../gcc/gcc/tree-optimize.c:664
#17 0x0001b3e0 in c_expand_body (fndecl=0x4148f0e8) at ../../gcc/gcc/c-
decl.c:6437
#18 0x003b2a6c in cgraph_expand_function (node=0x4148f0e8) 
at ../../gcc/gcc/cgraphunit.c:822
#19 0x003b47d8 in cgraph_expand_all_functions () 
at ../../gcc/gcc/cgraphunit.c:1689
#20 0x003b4b78 in cgraph_optimize () at ../../gcc/gcc/cgraphunit.c:1786
#21 0x0034d234 in compile_file () at ../../gcc/gcc/toplev.c:1009
#22 0x0034f100 in do_compile () at ../../gcc/gcc/toplev.c:2101
#23 0x0034f184 in toplev_main (argc=6042020, argv=0x5c31d0) 
at ../../gcc/gcc/toplev.c:2133
#24 0x000027a8 in _start (argc=20, argv=0xbffffcb0, envp=0xbffffd04) 
at /SourceCache/Csu/Csu-47/crt.c:267
#25 0x8fe1a558 in __dyld__dyld_start ()
(gdb) 
(gdb) p debug_generic_expr(node)
-1
$2 = void
(gdb) up
#1  0x00205504 in const_binop (code=MULT_EXPR, arg1=0x4148de58, 
arg2=0x414902e0, notrunc=0) at ../../gcc/gcc/fold-const.c:1543
1543          tree r2 = TREE_REALPART (arg2);
(gdb) p debug_generic_expr(arg1)
1.0e+0
$3 = void
(gdb) p debug_generic_expr(arg2)
-1
$4 = void
(gdb) up
#2  0x0021c1ac in fold (expr=0x4140c894) at ../../gcc/gcc/fold-const.c:7222
7222            t1 = const_binop (code, arg0, arg1, 0);
(gdb) p debug_generic_expr(expr)
1.0e+0 * -1
$5 = void
(gdb) 

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[stefaan dot deroeck at gmail dot com]

------- Additional Comments From stefaan dot deroeck at gmail dot com  2005-06-13 19:38 -------
still exists on main branch as well

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |stefaan dot deroeck at gmail
                   |                            |dot com


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2005-08-01 11:36 -------
*** Bug 23173 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |micis at gmx dot de


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-01 16:38 -------
*** Bug 23173 has been marked as a duplicate of this bug. ***

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-08-01 16:40 -------
*** Bug 21922 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |p dot van-hoof at qub dot ac
                   |                            |dot uk


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2005-08-01 23:58 -------
The ICE seems to happen when a floating point number is decremented.
The following code snippet in add_to_evolution looks suspicious:

  if (code == MINUS_EXPR)
    to_add = chrec_fold_multiply (type, to_add, 
				  build_int_cst_type (type, -1));

This leads to trouble if type is a REAL_TYPE.

The following patch (neither bootstrapped nor regtested) seems to fix
the problem (also in PR23173). Is this a proper fix?

===================================================================
--- tree-scalar-evolution.c	27 Jul 2005 13:26:54 -0000	2.32
+++ tree-scalar-evolution.c	1 Aug 2005 23:43:59 -0000
@@ -237,6 +237,7 @@ Software Foundation, 51 Franklin Street,
 #include "tm.h"
 #include "ggc.h"
 #include "tree.h"
+#include "real.h"
 
 /* These RTL headers are needed for basic-block.h.  */
 #include "rtl.h"
@@ -866,8 +867,9 @@ add_to_evolution (unsigned loop_nb, 
     }
 
   if (code == MINUS_EXPR)
-    to_add = chrec_fold_multiply (type, to_add, 
-				  build_int_cst_type (type, -1));
+    to_add = chrec_fold_multiply (type, to_add, TREE_CODE (type) == REAL_TYPE
+				  ? build_real (type, dconstm1)
+				  : build_int_cst_type (type, -1));
 
   res = add_to_evolution_1 (loop_nb, chrec_before, to_add);
 
===================================================================


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2005-08-02 16:05 -------
Testing a patch.


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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-08-03 14:19 -------
Subject: Bug 19899

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	reichelt@gcc.gnu.org	2005-08-03 14:18:56

Modified files:
	gcc            : ChangeLog Makefile.in tree-scalar-evolution.c 
	gcc/testsuite  : ChangeLog 
Added files:
	gcc/testsuite/gcc.dg/tree-ssa: scev-1.c scev-2.c 

Log message:
	PR tree-optimization/19899
	* Makefile.in (tree-scalar-evolution.o): Add real.h.
	* tree-scalar-evolution.c: Include real.h.
	(add_to_evolution): Build constant -1 of correct type.
	
	* gcc.dg/tree-ssa/scev-1.c: New test.
	* gcc.dg/tree-ssa/scev-2.c: New test.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=2.9646&r2=2.9647
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/Makefile.in.diff?cvsroot=gcc&r1=1.1533&r2=1.1534
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/tree-scalar-evolution.c.diff?cvsroot=gcc&r1=2.32&r2=2.33
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.5872&r2=1.5873
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/tree-ssa/scev-1.c.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/tree-ssa/scev-2.c.diff?cvsroot=gcc&r1=NONE&r2=1.1



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-08-03 14:41 -------
Subject: Bug 19899

CVSROOT:	/cvs/gcc
Module name:	gcc
Branch: 	gcc-4_0-branch
Changes by:	reichelt@gcc.gnu.org	2005-08-03 14:40:53

Modified files:
	gcc            : ChangeLog Makefile.in tree-scalar-evolution.c 
	gcc/testsuite  : ChangeLog 
Added files:
	gcc/testsuite/gcc.dg/tree-ssa: scev-1.c scev-2.c 

Log message:
	PR tree-optimization/19899
	* Makefile.in (tree-scalar-evolution.o): Add real.h.
	* tree-scalar-evolution.c: Include real.h.
	(add_to_evolution): Build constant -1 of correct type.
	
	* gcc.dg/tree-ssa/scev-1.c: New test.
	* gcc.dg/tree-ssa/scev-2.c: New test.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=2.7592.2.349&r2=2.7592.2.350
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/Makefile.in.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=1.1447.2.2&r2=1.1447.2.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/tree-scalar-evolution.c.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=2.17.4.2&r2=2.17.4.3
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=1.5084.2.316&r2=1.5084.2.317
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/tree-ssa/scev-1.c.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=NONE&r2=1.1.2.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.dg/tree-ssa/scev-2.c.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=NONE&r2=1.1.2.1



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899

[Bug tree-optimization/19899] ICE: tree check: expected real_cst, have integer_cst in const_binop, at fold-const.c:1490 with -ftree-vectorize -msse2

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2005-08-03 14:42 -------
Fixed on mainline and for GCC 4.0.2.


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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=19899