[Bug c/22576] New: ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[Bug c/22576] New: ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[black dot hole dot sun16 at gmail dot com]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain, Size: 1422 bytes --]

Hello. Sorry if the summary is a little vague, I dont really know a whole bunch
about gcc terminology ;) So anyway, I wrote a simple program in C that computes
the factorials of numbers 1 through 100 (yes, after 7 they get super big). When
this program is compiled as 

gcc -ffast-math factorial.c

I get an ICE;

paul@nephilim:~/c$ gcc -Wall -ffast-math factorials.c
factorials.c:8: warning: return type defaults to ‘int’
factorials.c: In function ‘main’:
factorials.c:24: internal compiler error: in instantiate_virtual_regs_lossage,
at function.c:1442
paul@nephilim:~/c$ gcc -dumpversion
4.0.2

This gcc is from Ubuntu-breezy, gcc-4.0 branch, CVS 20050718. Running on a
Pentium 4.

It told me to report this bug, so here I am :) Regardless of whether this can be
done in a simpler manner, and I'm /sure/ it can, I dont think gcc should die
because of it. The offending code is attached. -save-temps output also attached.

-- 
           Summary: ICE with simple factorial program compiled with -ffast-
                    math on gcc 4.0.2
           Product: gcc
           Version: 4.0.2
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: c
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: black dot hole dot sun16 at gmail dot com
                CC: gcc-bugs at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug c/22576] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[black dot hole dot sun16 at gmail dot com]

------- Additional Comments From black dot hole dot sun16 at gmail dot com  2005-07-20 19:29 -------
Created an attachment (id=9309)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=9309&action=view)
bug causing program...


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug c/22576] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[black dot hole dot sun16 at gmail dot com]

------- Additional Comments From black dot hole dot sun16 at gmail dot com  2005-07-20 19:29 -------
Created an attachment (id=9310)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=9310&action=view)
-save-temps output


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug c/22576] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[black dot hole dot sun16 at gmail dot com]

------- Additional Comments From black dot hole dot sun16 at gmail dot com  2005-07-20 19:30 -------
Created an attachment (id=9311)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=9311&action=view)
-save-temps output


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug c/22576] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[black dot hole dot sun16 at gmail dot com]

------- Additional Comments From black dot hole dot sun16 at gmail dot com  2005-07-20 19:32 -------
If it is at all relevant, I'm using the glibc that breezy provides (2.3.5).

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-07-20 19:33 -------
This is a target bug.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|c                           |target
 GCC target triplet|                            |i686-pc-linux-gnu
           Keywords|                            |ice-on-valid-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[black dot hole dot sun16 at gmail dot com]

------- Additional Comments From black dot hole dot sun16 at gmail dot com  2005-07-20 19:33 -------
Created an attachment (id=9312)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=9312&action=view)
#included file


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] [4.0/4.1 regression] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2005-07-20 21:26 -------
Reduced testcase:

========================
int foo(long double d)
{
    return d==0;
}
========================

Comparison of long doubles with zero seems to be a problem.
This is a regression from gcc 3.4.x.


-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |reichelt at gcc dot gnu dot
                   |                            |org
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2005-07-20 21:26:27
               date|                            |
            Summary|ICE with simple factorial   |[4.0/4.1 regression] ICE
                   |program compiled with -     |with simple factorial
                   |ffast-math on gcc 4.0.2     |program compiled with -
                   |                            |ffast-math on gcc 4.0.2
   Target Milestone|---                         |4.0.2


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] [4.0/4.1 regression] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[black dot hole dot sun16 at gmail dot com]

------- Additional Comments From black dot hole dot sun16 at gmail dot com  2005-07-20 23:13 -------
(From update of attachment 9309)
>#include <stdio.h>
>
>/* Program prints the first 10 postitive integers and their factorials */
>
>main()
>{
>long double i, j, k = 1;
>
>for(i = 1; i <= 100; i++) 
>	{
>	for(j = i; j > 0; j = j - 1)
>		{		
>		k = k * j; 			
>		if(j == 1) 	
>			{
>			printf("factorial of %Lf is %Lf\n", i, k);
>			k = 1; 
>			}
>		}
>	}
>	return 0;
>}


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] [4.0/4.1 regression] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[uros at kss-loka dot si]

------- Additional Comments From uros at kss-loka dot si  2005-07-21 08:01 -------
Patch that fixes attached testcase and reduced testcase from comment #7.

BTW: In the attached testcase, there is a comment that program falls apart if 
long double isn't used. With (patched) gcc-4.1 CVS, it works for me with "only" 
doubles, with SSE or x87 math, -ffast-math or without it.

...
factorial of 99 is 9.33262e+155
factorial of 100 is 9.33262e+157



-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                URL|                            |http://gcc.gnu.org/ml/gcc-
                   |                            |patches/2005-
                   |                            |07/msg01391.html
           Keywords|                            |patch


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] [4.0/4.1 regression] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[reichelt at gcc dot gnu dot org]

------- Additional Comments From reichelt at gcc dot gnu dot org  2005-07-21 17:07 -------
The problem is related to PR22585 where another ICE with long doubles occurs.
Unfortunately Uros' patch doesn't fix the problem there.


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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] [4.0/4.1 regression] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-07-21 19:55 -------
Subject: Bug 22576

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	rth@gcc.gnu.org	2005-07-21 19:55:03

Modified files:
	gcc            : ChangeLog 
	gcc/config/i386: i386.md 

Log message:
	PR target/22576
	* config/i386/i386.md (cmpxf): Change operand constraints
	to "nonmemory_operand".

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=2.9504&r2=2.9505
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/config/i386/i386.md.diff?cvsroot=gcc&r1=1.649&r2=1.650



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] [4.0/4.1 regression] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-07-21 19:56 -------
Subject: Bug 22576

CVSROOT:	/cvs/gcc
Module name:	gcc
Branch: 	gcc-4_0-branch
Changes by:	rth@gcc.gnu.org	2005-07-21 19:56:30

Modified files:
	gcc            : ChangeLog 
	gcc/config/i386: i386.md 

Log message:
	PR target/22576
	* config/i386/i386.md (cmpxf): Change operand constraints
	to "nonmemory_operand".

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=2.7592.2.325&r2=2.7592.2.326
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/config/i386/i386.md.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=1.618.4.4&r2=1.618.4.5



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] [4.0/4.1 regression] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2005-07-21 20:33 -------
Fixed.

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576

[Bug target/22576] [4.0/4.1 regression] ICE with simple factorial program compiled with -ffast-math on gcc 4.0.2

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-09-26 19:20 -------
Subject: Bug 22576

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	janis@gcc.gnu.org	2005-09-26 19:19:43

Modified files:
	gcc/testsuite  : ChangeLog 
Added files:
	gcc/testsuite/gcc.target/i386: pr22576.c pr22585.c pr23268.c 
	                               builtin-apply-mmx.c 

Log message:
	2005-08-26  Uros Bizjak  <uros@kss-loka.si>
	
	* gcc.target/i386/builtin-apply-mmx.c: New test.
	
	PR target/22576
	* gcc.target/i386/pr22576.c: New test.
	
	PR target/22585
	* gcc.target/i386/pr22585.c: New test.
	
	PR target/23268
	* gcc.target/i386/pr23268.c: New test.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.6104&r2=1.6105
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.target/i386/pr22576.c.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.target/i386/pr22585.c.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.target/i386/pr23268.c.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.target/i386/builtin-apply-mmx.c.diff?cvsroot=gcc&r1=NONE&r2=1.1



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=22576