Mailing-List: contact cygwin-help AT sourceware DOT cygnus DOT com; run by ezmlm
List-Unsubscribe: <mailto:cygwin-unsubscribe-archive-cygwin=delorie DOT com AT sourceware DOT cygnus DOT com>
List-Subscribe: <mailto:cygwin-subscribe AT sourceware DOT cygnus DOT com>
List-Archive: <http://sourceware.cygnus.com/ml/cygwin/>
List-Post: <mailto:cygwin AT sourceware DOT cygnus DOT com>
List-Help: <mailto:cygwin-help AT sourceware DOT cygnus DOT com>, <http://sourceware.cygnus.com/ml/#faqs>
Sender: cygwin-owner AT sourceware DOT cygnus DOT com
Delivered-To: mailing list cygwin AT sourceware DOT cygnus DOT com
Reply-To: <ssiddiqi AT nspm DOT ipass DOT net>
From: "Suhaib Siddiqi" <sssiddiqi AT ipass DOT net>
To: <cygwin AT sourceware DOT cygnus DOT com>, <ssiddiqi AT nspm DOT ipass DOT net>
Subject: RE: cc1.exe in Cygwin v 1.0 doing coredump
Date: Wed, 3 Nov 1999 12:17:41 -0500
Message-ID: <NDBBLHPPKKBKPBLLMFJBGEFICEAA.sssiddiqi@ipass.net>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0)
In-Reply-To: <19991103121953.C22192@cygnus.com>
Importance: Normal
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2314.1300

It is PGCC 2.95.2 loop.c file, which has a lot of intel optimization
stuff.  Compiling it with -O2 flag is causing stack overflow and
cc1.exe to coredump, without optimization flags, it compiles, though
returns
100s of warnings. I am not sure if it is PGCC code in loop.c which
needs to be fixed?  I hae sent a message to PGCC 2.95.2 folks too.

Strange thing the same code with gcc-2.95 compiles with -O2 flags.

Suhaib

****************************
Suhaib M. Siddiqi
Senior Research Scientist
Inspire Pharmaceuticals, Inc.
4222 Emperor Blvd., Suite 470
Durham, NC 27703

Tel: 919-941-9777 Ext 238
FAX: 919-941-9797
http://www.inspirepharm.com


> -----Original Message-----
> From: cygwin-owner AT sourceware DOT cygnus DOT com
> [mailto:cygwin-owner AT sourceware DOT cygnus DOT com]On Behalf Of
> Chris Faylor
> Sent: Wednesday, November 03, 1999 12:20 PM
> To: ssiddiqi AT nspm DOT ipass DOT net
> Cc: Suhaib Siddiqi; Cygwin AT Sourceware DOT Cygnus DOT Com
> Subject: Re: cc1.exe in Cygwin v 1.0 doing coredump
>
>
> On Wed, Nov 03, 1999 at 11:01:52AM -0500, Suhaib Siddiqi wrote:
> >Never mind.  It is something in my code.  With optimization (-O2)
> >turned on cc1.exe does core dump, without optimization flags, it
> >work.
>
> Hmm.  I hate to say this but there is nothing that you should be
> able to do to your code which should cause the compiler
> to core dump.
> The back-end of gcc on the CD has been completely rewritten
> to produce better quality x86 code.  It seems that you have hit
> something which we hadn't tested for.
>
> cgf
>
> >> -----Original Message-----
> >> From: cygwin-owner AT sourceware DOT cygnus DOT com
> >> [mailto:cygwin-owner AT sourceware DOT cygnus DOT com]On Behalf Of
> >> Suhaib Siddiqi
> >> Sent: Wednesday, November 03, 1999 10:38 AM
> >> To: Cygwin AT Sourceware DOT Cygnus DOT Com
> >> Subject: cc1.exe in Cygwin v 1.0 doing coredump
> >>
> >>
> >>
> >> I am trying to compile something, and cc1.exe from gcc in
> >> Cygwin 1.0
> >> (Cd Distribution) is doing stack dump.
> >>
> >> loop.c:3907: internal error--unrecognizable insn:
> >> (insn 394 225 232 (set (reg/v:SI 72)
> >>         (minus:SI (reg:SI 70)
> >>             (const_int 1))) -1 (nil)
> >>     (expr_list:REG_DEAD (reg:SI 70)
> >>         (nil)))
> >>     0       0 [main]
> >> d:\cygwin\usr\lib\gcc-lib\i686-cygwin\2.9-cygwin-990830\cc1
> >> .exe 8455 handle_exceptions: Exception: STATUS_ACCESS_VIOLATION
> >>     0   22742 [main] cc1 8455 stackdump: Dumping stack trace to
> >> cc1.exe.stackdum
> >> p
> >> make[1]: *** [loop.o] Error 1
> >>
> >>
> >> 21601   44343 [main] cc1 8455 exception: trapped!
> >> 44174   88517 [main] cc1 8455 exception: code 0xC0000005 at
> >> 0x6105E871
> >> 41132  129649 [main] cc1 8455 exception: ax 0x0 bx 0x1 cx
> >> 0xFFFFFFFF
> >> dx 0x5CD
> >> 41987  171636 [main] cc1 8455 exception: si 0x0 di 0x5CD bp
> >> 0x25BE4E4 sp 0x25BE4E0
> >> 58070  229706 [main] cc1 8455 exception: exception is:
> >> STATUS_ACCESS_VIOLATION
> >> 42009  271715 [main] cc1 8455 stack: Stack trace:
> >> 49620  321335 [main] cc1 8455 stack: frame 0: sp =
> 0x25BE300, pc =
> >> 0x6100AE52
> >> 41965  363300 [main] cc1 8455 stack: frame 1: sp =
> 0x25BE314, pc =
> >> 0x6100B230
> >> 49691  412991 [main] cc1 8455 stack: frame 2: sp =
> 0x25BE330, pc =
> >> 0x77F96666
> >> 50425  463416 [main] cc1 8455 stack: frame 3: sp =
> 0x25BE354, pc =
> >> 0x77F8912B
> >> 49589  513005 [main] cc1 8455 stack: frame 4: sp =
> 0x25BE3E0, pc =
> >> 0x77F763BA
> >> 42164  555169 [main] cc1 8455 stack: frame 5: sp =
> 0x25BE4E4, pc =
> >> 0x61061D1E
> >> 41158  596327 [main] cc1 8455 stack: frame 6: sp =
> 0x25BE71C, pc =
> >> 0x61061328
> >> 58662  654989 [main] cc1 8455 stack: frame 7: sp =
> 0x25BE734, pc =
> >> 0x610612D7
> >> 41417  696406 [main] cc1 8455 stack: frame 8: sp =
> 0x25BEBB4, pc =
> >> 0x61061443
> >> 50287  746693 [main] cc1 8455 stack: frame 9: sp =
> 0x25BEDF4, pc =
> >> 0x61061328
> >> 49619  796312 [main] cc1 8455 stack: frame 10: sp =
> >> 0x25BEE0C, pc =
> >> 0x6105C4F0
> >> 42253  838565 [main] cc1 8455 stack: frame 11: sp =
> >> 0x25BEE20, pc =
> >> 0x405578
> >> 66152  904717 [main] cc1 8455 stack: frame 12: sp =
> >> 0x25BEE44, pc =
> >> 0x405618
> >> 42012  946729 [main] cc1 8455 stack: frame 13: sp =
> >> 0x25BEE54, pc =
> >> 0x4FE0E1
> >> 49617  996346 [main] cc1 8455 stack: frame 14: sp =
> >> 0x25BEE7C, pc =
> >> 0x46AFBD
> >> 50392 1046738 [main] cc1 8455 stack: frame 15: sp =
> >> 0x25BEEE4, pc =
> >> 0x46A5AC
> >> 41244 1087982 [main] cc1 8455 stack: End of stack trace
> >> (more stack
> >> frames may be present)
> >>
> >>
> >> ****************************
> >> Suhaib M. Siddiqi
> >> Senior Research Scientist
> >> Inspire Pharmaceuticals, Inc.
> >> 4222 Emperor Blvd., Suite 470
> >> Durham, NC 27703
> >>
> >> Tel: 919-941-9777 Ext 238
> >> FAX: 919-941-9797
> >> http://www.inspirepharm.com
> >>
> >>
> >>
> >> --
> >> Want to unsubscribe from this list?
> >> Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com
> >>
> >
> >
> >--
> >Want to unsubscribe from this list?
> >Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com
> >
>
> --
> cgf AT cygnus DOT com
> http://www.cygnus.com/
>
> --
> Want to unsubscribe from this list?
> Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com
>


--
Want to unsubscribe from this list?
Send a message to cygwin-unsubscribe AT sourceware DOT cygnus DOT com