GCC Bugzilla Bug 10083

ICE while building event.c from swi-prolog-packages on Linux/alpha.  Happens
with 3.3, 3.2, and 2.95:

gcc -c -Wall -O2 -fno-strict-aliasing -I. -I../src -I/usr/include  -fPIC
-funsigned-char -DHAVE_CONFIG_H evt/event.c -o evt/event.o
evt/event.c: In function `getTimeEvent':
evt/event.c:235: Internal compiler error in simplify_unary_operation, at
simplify-rtx.c:636

gcc-2.95 gives a slightly different error message:
evt/event.c:235: internal error--unrecognizable insn:
(insn 35 34 36 (set (reg:DI 81)
        (truncate:DI (lshiftrt:TI (mult:TI (zero_extend:TI (reg:DI 80))
                    (zero_extend:TI (const_int 5 [0x5])))
                (const_int 64 [0x40])))) -1 (insn_list 34 (nil))
    (nil))

Release:
3.3 20030301 (prerelease)

Environment:
Reading specs from /usr/lib/gcc-snapshot/lib/gcc-lib/alpha-linux/3.3/specs
Configured with: /build/buildd/gcc-snapshot-20030301/src/configure -v
--enable-languages=c,c++,java,f77,objc,ada --prefix=/usr/lib/gcc-snapshot
--infodir=/share/info --mandir=/share/man --enable-shared --with-system-zlib
--enable-nls --without-included-gettext --enable-__cxa_atexit
--enable-clocale=gnu --enable-java-gc=boehm --enable-java-awt=xlib
--with-cpp-install-dir=bin --enable-objc-gc alpha-linux
Thread model: posix

From: Richard Henderson <rth@redhat.com>
To: Falk Hueffner <falk.hueffner@student.uni-tuebingen.de>
Cc: gcc-gnats@gcc.gnu.org, 184749@bugs.debian.org, gcc-bugs@gcc.gnu.org,
   rmurray@debian.org
Subject: Re: c/10083: alpha: ICE while building swi-prolog-packages
Date: Sun, 16 Mar 2003 11:59:39 -0800

 On Sun, Mar 16, 2003 at 03:56:47PM +0100, Falk Hueffner wrote:
 > 		      (zero_extend:TI (const_int 65 [0x41])))
 
 This is indeed illegal.  Should be just the const_int here.
 
 
 r~

From: Falk Hueffner <falk.hueffner@student.uni-tuebingen.de>
To: gcc-gnats@gcc.gnu.org, 184749@bugs.debian.org, gcc-bugs@gcc.gnu.org,
   rmurray@debian.org
Cc:  
Subject: Re: c/10083: alpha: ICE while building swi-prolog-packages
Date: 16 Mar 2003 14:36:14 +0100

 Hi,
 
 a slightly smaller test case is:
 
 static unsigned long getTimeEvent(unsigned long time)
 {
     return time % ((1UL << 62) - 1);
 }
 
 % gcc -v
 gcc version 3.4 20030310 (experimental)
 % gcc -c -O2   event.i
 event.i: In function `getTimeEvent':
 event.i:4: internal compiler error: in simplify_unary_operation, at simplify-rtx.c:707
 Please submit a full bug report,
 with preprocessed source if appropriate.
 See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
 
 -- 
 	Falk

From: Falk Hueffner <falk.hueffner@student.uni-tuebingen.de>
To: gcc-gnats@gcc.gnu.org
Cc: 184749@bugs.debian.org, gcc-bugs@gcc.gnu.org, rmurray@debian.org
Subject: Re: c/10083: alpha: ICE while building swi-prolog-packages
Date: 16 Mar 2003 15:56:47 +0100

 Falk Hueffner <falk.hueffner@student.uni-tuebingen.de> writes:
 
 > a slightly smaller test case is:
 > 
 > static unsigned long getTimeEvent(unsigned long time)
 > {
 >     return time % ((1UL << 62) - 1);
 > }
 
 The problem is that expand_divmod generates a zero extension of a
 constant, which is VOIDmode:
 
 (insn 11 10 12
       (nil)
       (set (reg:DI 72)
 	   (truncate:DI (lshiftrt:TI (mult:TI (zero_extend:TI (reg/v:DI 70 [time]))
 					      (zero_extend:TI (const_int 65 [0x41])))
 				     (const_int 64 [0x40])))) -1 (nil)
 				     (nil))
 
 Should expand_divmod not generate this, or should
 simplify_unary_operation be able to handle it?
 
 -- 
 	Falk

State-Changed-From-To: open->analyzed
State-Changed-Why: Has already been analyzed by others

From: Falk Hueffner <falk.hueffner@student.uni-tuebingen.de>
To: gcc-gnats@gcc.gnu.org, 184749@bugs.debian.org, rmurray@debian.org
Cc:  
Subject: Re: c/10083: alpha: ICE while building swi-prolog-packages
Date: 25 Mar 2003 07:40:48 +0100

 Hi,
 
 here's a patch:
 
 http://gcc.gnu.org/ml/gcc-patches/2003-03/msg02027.html
 
         Falk

From: rth@gcc.gnu.org
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: c/10083
Date: 30 Mar 2003 19:00:12 -0000

 CVSROOT:	/cvs/gcc
 Module name:	gcc
 Changes by:	rth@gcc.gnu.org	2003-03-30 19:00:12
 
 Modified files:
 	gcc            : ChangeLog 
 	gcc/config/alpha: alpha.md 
 Added files:
 	gcc/testsuite/gcc.c-torture/compile: 20030330-1.c 
 
 Log message:
 	PR c/10083
 	* config/alpha/alpha.md (umuldi3_highpart): Change to expander;
 	don't zero_extend const inputs.
 	* gcc.c-torture/compile/20030330-1.c: New.
 
 Patches:
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&r1=1.17252&r2=1.17253
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/config/alpha/alpha.md.diff?cvsroot=gcc&r1=1.206&r2=1.207
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.c-torture/compile/20030330-1.c.diff?cvsroot=gcc&r1=NONE&r2=1.1
 

From: rth@gcc.gnu.org
To: gcc-gnats@gcc.gnu.org
Cc:  
Subject: c/10083
Date: 30 Mar 2003 19:02:56 -0000

 CVSROOT:	/cvs/gcc
 Module name:	gcc
 Branch: 	gcc-3_3-branch
 Changes by:	rth@gcc.gnu.org	2003-03-30 19:02:56
 
 Modified files:
 	gcc            : ChangeLog 
 	gcc/config/alpha: alpha.md 
 Added files:
 	gcc/testsuite/gcc.c-torture/compile: 20030330-1.c 
 
 Log message:
 	PR c/10083
 	* config/alpha/alpha.md (umuldi3_highpart): Change to expander;
 	don't zero_extend const inputs.
 	* gcc.c-torture/compile/20030330-1.c: New.
 
 Patches:
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.16114.2.382&r2=1.16114.2.383
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/config/alpha/alpha.md.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=1.199.4.2&r2=1.199.4.3
 http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gcc.c-torture/compile/20030330-1.c.diff?cvsroot=gcc&only_with_tag=gcc-3_3-branch&r1=NONE&r2=1.1.2.1
 

State-Changed-From-To: analyzed->closed
State-Changed-Why: Fixed by the commit in the audit trail.