[Bug regression/20009] New: [4.0 regression] cris-elf testsuite failures: gcc.c-torture/unsorted/USIcmp.c, -O1 and above

[Bug regression/20009] New: [4.0 regression] cris-elf testsuite failures: gcc.c-torture/unsorted/USIcmp.c, -O1 and above

[hp at gcc dot gnu dot org]

With LAST_UPDATED: "Wed Feb 16 04:21:14 UTC 2005", these tests succeeded.
With LAST_UPDATED: "Wed Feb 16 14:53:35 UTC 2005" they fail:
FAIL: gcc.c-torture/unsorted/USIcmp.c,  -O1
FAIL: gcc.c-torture/unsorted/USIcmp.c,  -O2
FAIL: gcc.c-torture/unsorted/USIcmp.c,  -O3 -fomit-frame-pointer
FAIL: gcc.c-torture/unsorted/USIcmp.c,  -O3 -g
FAIL: gcc.c-torture/unsorted/USIcmp.c,  -Os

The message in the log is:
Executing on host: /home/hp/cvs_areas/combined/cris-regobj/gcc/xgcc
-B/home/hp/cvs_areas/combined/cris-regobj/gcc/  -w  -O1   -c   -isystem
/home/hp/cvs_areas/combined/cris-regobj/cris-unknown-elf/.\
/newlib/targ-include -isystem
/home/hp/cvs_areas/combined/combined/newlib/libc/include  -o /home/hp\
/cvs_areas/combined/cris-regobj/gcc/testsuite/USIcmp.o
/home/hp/cvs_areas/combined/combined/gcc/testsuite/gcc.c-torture/unsorted/USIcmp.c   
(timeout = 300)
/home/hp/cvs_areas/combined/combined/gcc/testsuite/gcc.c-torture/unsorted/USIcmp.c:
In function 'imm0limm1':^M
/home/hp/cvs_areas/combined/combined/gcc/testsuite/gcc.c-torture/unsorted/USIcmp.c:103:
internal compiler error: tree check: expected ssa_name, have addr_expr in
vrp_hash, at tree-ssa-dom.c:3310^M

Jakub's change in this time-frame is the only one I see that should matter, so
CC:ed.
(Jeff's being RTL shouldn't cause a tree-check problem.)

-- 
           Summary: [4.0 regression] cris-elf testsuite failures: gcc.c-
                    torture/unsorted/USIcmp.c, -O1 and above
           Product: gcc
           Version: 4.0.0
            Status: UNCONFIRMED
          Keywords: ice-checking
          Severity: normal
          Priority: P2
         Component: regression
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: hp at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org,jakub at redhat dot com
  GCC host triplet: i686-pc-linux-gnu
GCC target triplet: cris-axis-elf


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20009

[Bug tree-optimization/20009] [4.0 regression] cris-elf testsuite failures: gcc.c-torture/unsorted/USIcmp.c, -O1 and above

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-02-16 19:04 -------
I see this every once in a while on a different testcase.

CCing Jeff as he is the person who knows DOM a lot.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |law at gcc dot gnu dot org
          Component|regression                  |tree-optimization
           Keywords|                            |ice-on-valid-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20009

[Bug tree-optimization/20009] [4.0 regression] cris-elf testsuite failures: gcc.c-torture/unsorted/USIcmp.c, -O1 and above

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-02-16 19:32 -------
This also can be seen on x86:
http://gcc.gnu.org/ml/gcc-regression/2005-02/msg00052.html

Confirmed.  Most likely DOM is doing some thing wrong.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2005-02-16 19:32:35
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20009

[Bug tree-optimization/20009] [4.0 regression] cris-elf testsuite failures: gcc.c-torture/unsorted/USIcmp.c, -O1 and above

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-02-17 13:45 -------
Actually this is not a problem in DOM at all but in fold, say we have (which is the reduced testcase):
unsigned int glob1;
int imm0limm1 (int r0)
{
  if (22 <= ((unsigned int)&glob1))
    return 1;
  else
    return 0;
}

fold looks like it is removing the cast to "unsigned int" and does not convert 22 to pointer type.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.0.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20009

[Bug tree-optimization/20009] [4.0 regression] cris-elf testsuite failures: gcc.c-torture/unsorted/USIcmp.c, -O1 and above

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-02-17 13:49 -------
This is caused obviously by:
        * convert.c (convert_to_integer) <case POINTER_TYPE>: Pass
        TYPE_UNSIGNED (type) as type_for_size's UNSIGNEDP argument.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20009

[Bug tree-optimization/20009] [4.0 regression] cris-elf testsuite failures: gcc.c-torture/unsorted/USIcmp.c, -O1 and above

[jsm28 at gcc dot gnu dot org]

------- Additional Comments From jsm28 at gcc dot gnu dot org  2005-02-17 20:00 -------
Also fails on ia64-hpux.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |jsm28 at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20009

[Bug tree-optimization/20009] [4.0 regression] cris-elf testsuite failures: gcc.c-torture/unsorted/USIcmp.c, -O1 and above

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-02-17 23:57 -------
Subject: Bug 20009

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	jakub@gcc.gnu.org	2005-02-17 23:56:55

Modified files:
	gcc            : ChangeLog convert.c 
	gcc/testsuite  : ChangeLog 
Added files:
	gcc/testsuite/gcc.c-torture/compile: 20050217-1.c 
	gcc/testsuite/g++.dg/opt: switch3.C 

Log message:
	PR c++/20023
	PR tree-optimization/20009
	* convert.c (convert_to_integer): Revert 2005-02-16 change.
	
	* gcc.c-torture/compile/20050217-1.c: New test.
	* g++.dg/opt/switch3.C: New test.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=2.7518&r2=2.7519
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/convert.c.diff?cvsroot=gcc&r1=1.56&r2=1.57
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.5045&r2=1.5046
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.c-torture/compile/20050217-1.c.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/g++.dg/opt/switch3.C.diff?cvsroot=gcc&r1=NONE&r2=1.1



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20009

[Bug tree-optimization/20009] [4.0 regression] cris-elf testsuite failures: gcc.c-torture/unsorted/USIcmp.c, -O1 and above

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-02-18 00:22 -------
Fixed.

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=20009