[12015] in Athena Bugs
sun4 7.6r suncc bug
daemon@ATHENA.MIT.EDU (alexp@MIT.EDU)
Fri May 6 14:16:56 1994
From: alexp@MIT.EDU
Date: Fri, 6 May 94 14:16:37 EDT
To: phils@MIT.EDU, bugs@MIT.EDU
Cc: phils@MIT.EDU, alexp@MIT.EDU
>From phils@MIT.EDU Fri Apr 22 19:35:03 1994
>Received: by gregor.MIT.EDU (5.57/4.7) id AA29834; Fri, 22 Apr 94 19:35:02 -04>00
>Received: from OLMSTED.MIT.EDU by MIT.EDU with SMTP
> id AA25367; Fri, 22 Apr 94 19:34:50 EDT
>From: phils@MIT.EDU
>Received: by olmsted.MIT.EDU (AIX 3.2/UCB 5.64/4.7) id AA13008; Fri, 22 Apr 19>94 19:34:48 -0400
>Date: Fri, 22 Apr 1994 19:34:48 -0400
>Message-Id: <9404222334.AA13008@olmsted.MIT.EDU>
>To: bugs@MIT.EDU
>Subject: sun4 7.6r suncc bug
>Cc: phils@MIT.EDU
>
>
>
>System name: sunra
>Type and version: SPARC/Classic 7.6R (1 update(s) to same version)
>Display type: cgthree
>
>
>I can't compile the simplest programs (see below) with the -fast option
>with suncc. I get:
>
>% suncc -fast test.c -lm
>Undefined first referenced
> symbol in file
>__cg92_libc (command line)
>ld: fatal: Symbol referencing errors. No output written to a.out
>
>
>/* test.c */
>#include <math.h>
>
>main()
>{
> int i;
>
> for (i=0; i < 100000; i++)
> sqrt((double)i);
>}
>
Hi, I have been working on your bug report and have reproduced the problem.
There have been a number of problems with the floating point libraries of
the Sun C compiler and we may be able to apply a Sun patch to fix it
(I will have to check with someone else on this); for what it's worth, the
problem does not show up on an IPX in case you have the option of working
on that model. I will get back to you if we can fix this.
Alex Prengel, Jr.
MIT Academic Computing Services
Software Acquisition Coordinator
