gcc2.95-pre bug report

gcc2.95-pre bug report

[Vadim Gouterman]

Short Description:
	egcs prints a bogus error message when bitfield instance 
	variable is used by the fully qualified name

Software version:
        gcc version gcc-2.95pre (latest CVS snapshot)

OS Version:
        SunOS 5.5.1 Generic sun4u sparc SUNW,Ultra-1

Command line used to invoke compiler:
        gcc test.cc -c -o test.o

Full description:
        egcs prints a bogus error message when bitfield instance 
	variable is used by the fully qualified name: 
	Classname::instanceName. It works fine if unqualified name 
	is used. 
	
	The following error message is printed:
	/home/vadimg/test.cc: In method `void P::inspectFlag()':
	/home/vadimg/test.cc:16: illegal pointer to bit field `G::iFlags'

Testcase:
        The testcase file "test.cc" is attached to this mail.
        Please contact me if you need any additional information.



-- 
Vadim Gouterman
Sr. Member of Scientific Staff
mailto:vadimg@cadabradesign.com
Cadabra Design Automation Inc.
Phone (613) 226-7046
http://www.cadabradesign.com

class G {
  public:
    unsigned char iFlags : 8;
};

class P : public G {
  public:
    void inspectFlag();
};

void
P::inspectFlag()
{
    iFlags;
    G::iFlags;
}
>From martin.kahlert@provi.de Wed Jul 07 10:38:00 1999
From: Martin Kahlert <martin.kahlert@provi.de>
To: craig@jcb-sc.com
Cc: egcs-bugs@egcs.cygnus.com
Subject: Re: Bug with g77 and -mieee on Alpha Linux
Date: Wed, 07 Jul 1999 10:38:00 -0000
Message-id: <19990707194012.A291@keksy.linux.provi.de>
References: <199907062042.WAA00509@keksy.linux.provi.de> <19990707140435.1429.qmail@deer>
X-SW-Source: 1999-07/msg00262.html
Content-length: 2443

Quoting craig@jcb-sc.com (craig@jcb-sc.com):
> >Perhaps i have done something wrong, but i thought, -mieee should 
> >prevent these FPE's from occuring?
> 
> I don't know, it might/should, though I don't know offhand what the
> FP value you're trying to print actually is.  (It seems unlikely to
> be a signaling NaN, but, if that *was* what you're trying to print,
> I could understand why you got an FPE.)
The value is 1.448625e-313.

> There are two things that might be working, together or separately, to
> give you trouble:
> 
>   1.  You're supplying your own main(), but not running the libg2c
>       initialization code libg2c's main() normally runs.  See the
>       g77 docs for more info.  (Though in your example you wouldn't
>       need to call the routine to set up command-line info, you might
>       in the product code from which you derived the example.  And,
>       in both examples, setting up the signal stuff might be necessary.)
That's not the case (i changed that with no success).

> 
>   2.  You're using libg2c to print the FP value, but libg2c *itself*
>       has (presumably) not been compiled with -mieee.
This must be the reason. It has problems with 
'if(n<0)' in lwrite.c: l_g. Here it generates an FPE.

So i ask you to incorporate -mieee into the flags for libg2c.

>       I had assumed (erroneously, as I recently discovered) that gcc
>       automatically used multilibs to produce -mieee-compiled libraries
>       for Alpha targets, because, without that, libraries can crash (and
>       libg2c apparently *will* crash doing, e.g., output of floating-point
>       values that require -mieee).
> 
>   3.  You're not supplying `-mieee' when linking the object files, so
>       even if there *was* a libg2c.a built with -mieee available (and
>       properly configured, whatever that means), you'd pick up the default
>       one and, if #2 is correct, crash anyway.
This didn't make any difference, either.

> My own opinion is that this sort of thing is insanely painful to deal with,
> and that the default should have been -mieee all along (both for gcc and
> for Digital's compilers, but I assume the latter have all-along offered
> easier ways to get the -mieee behavior), with an option (-mno-ieee) being
> needed to get -mno-ieee performance.
I have to agree heavily on that.

Thanks,
Martin.

-- 
Your mouse has moved, Windows must be restarted for changes
to take affect - restart now?
>From ronis@ronispc.chem.mcgill.ca Wed Jul 07 11:02:00 1999
From: David Ronis <ronis@ronispc.chem.mcgill.ca>
To: egcs-bugs@egcs.cygnus.com
Subject: Strange Bootstrap Failure
Date: Wed, 07 Jul 1999 11:02:00 -0000
Message-id: <199907071802.OAA14133@ronispc.chem.mcgill.ca>
X-SW-Source: 1999-07/msg00263.html
Content-length: 1663

Today's CVS gave me the following build failure (make boostrap):

stage1/xgcc -Bstage1/ -B/usr/i586-pc-linux-gnulibc1/bin/ -c  -DIN_GCC    -DUSE_GNULIBC_1 -W -Wall -O2 -O3 -march=pentium -funroll-loops -fomit-frame-pointer -fstrict-aliasing  -DHAVE_CONFIG_H    -I. -I../../egcs/gcc -I../../egcs/gcc/config -I../../egcs/gcc/../include ../../egcs/gcc/sched.c
../../egcs/gcc/sched.c: In function `insn_cost':
../../egcs/gcc/sched.c:878: warning: implicit declaration of function `x86_adjust_cost'
../../egcs/gcc/sched.c: In function `schedule_block':
../../egcs/gcc/sched.c:2631: warning: `regs_sometimes_live' might be used uninitialized in this function
../../egcs/gcc/sched.c:2632: warning: `sometimes_max' might be used uninitialized in this function
../../egcs/gcc/sched.c:3536: Internal compiler error in `compute_use_by_pseudos'
, at reload1.c:561
Please submit a full bug report to `egcs-bugs@egcs.cygnus.com'.
See <URL: http://egcs.cygnus.com/faq.html#bugreport > for details.
make[2]: *** [sched.o] Error 1
make[2]: Leaving directory `/home/ronis/project/objdir/gcc'
make[1]: *** [bootstrap] Error 2
make[1]: Leaving directory `/home/ronis/project/objdir/gcc'
make: *** [bootstrap] Error 2

What's strange is that when I ran the command from the command line
(with -save-temps to get intermediate files for a bug report) it
worked; I got the same warnings but no ICE.  I'm running linux-2.2.10
on an i586.  I reran, checking memory usage, and never exceeded 20% of
RAM.  In addition, the machine was fairly inactive during the bootstrap.

1.  Any idea what's going on? 

2.  Is there some way to continue with the bootstrap from this point onwards?

David
>From wilson@cygnus.com Wed Jul 07 11:17:00 1999
From: Jim Wilson <wilson@cygnus.com>
To: law@cygnus.com
Cc: "Kaveh R. Ghazi" <ghazi@caip.rutgers.edu>, egcs-bugs@egcs.cygnus.com
Subject: Re: gcc-2.95 19990627 irix6 bootstrap fails with "-O3 -funroll-all-loops" 
Date: Wed, 07 Jul 1999 11:17:00 -0000
Message-id: <199907071817.LAA23502@rtl.cygnus.com>
References: <14848.931333084@upchuck.cygnus.com>
X-SW-Source: 1999-07/msg00264.html
Content-length: 487

	  > 	* unroll.c (unroll_loop): Don't delete NOTE_INSN_DELETED_LABEL notes.
	Thanks.  I went ahead and installed this patch.

It doesn't work.  My bootstrap still fails.  Kaveh posted a message about this.

I wrote a second different patch on Friday, started another bootstrap, and saw
that it had failed again when I came in Tuesday morning.  I haven't gotten back
to it yet.

I think I can get it working if I combine my two patches, but the patch is
starting to get pretty ugly.

Jim
>From maeder@glue.ch Wed Jul 07 11:18:00 1999
From: Thomas Maeder <maeder@glue.ch>
To: Alexandre Oliva <oliva@dcc.unicamp.br>
Cc: egcs-bugs@egcs.cygnus.com
Subject: Re: Internal compiler error.
Date: Wed, 07 Jul 1999 11:18:00 -0000
Message-id: <378387C0.BE45EC15@glue.ch>
References: <3780F0B8.57E64911@glue.ch> <orso71v9vr.fsf@cupuacu.lsd.dcc.unicamp.br>
X-SW-Source: 1999-07/msg00265.html
Content-length: 385

Alexandre Oliva schrieb:
> 
> > Here's another one, this time without date.
> 
> > //h/Probase/Source/Geometry/Square/Occupy.cpp:37: Internal compiler > > error.
> 
> Thanks, fixed in the upcoming gcc 2.95.

Again thanks for the quick answer. Is there an further information
available:

What's the problem?
Is there a known simple workaround?
When will gcc 2.95 be released?

Thomas