Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!ames!sgi!calvin@dinkum.SGI.COM
From: calvin@dinkum.SGI.COM (Calvin H. Vu)
Newsgroups: comp.sys.sgi
Subject: Re: fortran (once again)
Summary: Bug not reproducible, version required
Message-ID: <29669@sgi.SGI.COM>
Date: 30 Mar 89 01:44:32 GMT
References: <8903291158.aa16902@SMOKE.BRL.MIL>
Sender: daemon@sgi.SGI.COM
Organization: Silicon Graphics, Inc., Mountain View, CA
Lines: 58

In article <8903291158.aa16902@SMOKE.BRL.MIL>, LAAKSONE@FINFUN.BITNET writes:
> 
> Please tell me what I'm doing wrong. I have a short file with some
> fortran subroutines but I'm totally unable to compile the file.
> The fortran compiler grabs more and more of memory and after a while
> the system tells: DANGER: Out of swap space and I lose the control
> of the system totally. The system I'm using is a 4D/70GTB and the
> fortran compiler is version 1.31.
> 
> I can get the same effect by just compiling the following subroutine:
> 
> I have tried with and without optimization but always the same problem.
> ----------------------cut here--------------------------------------
> C
> C
> C PLOTTER OFF CALL    ESC.)
>         SUBROUTINE HPLOF
> C BEFORE OFF SEND A TERMINATING CHARACTER
>         CALL HPOUT(char(125)//char(27)//'.)')
>         RETURN
>         END
> 
> -----------------------cut here------------------------------------
> I have tried to compile the code on a VAX and a SUN and there was no
> problem.
> 
> Thanks in advance,
> 
> Leif Laaksonen
> CSC ESPOO FINLAND

	I tried your test case with the 3.1 Rev C release which is the oldest
	1.31 compiler I can lay my hands on but could not reproduce the
	problem.   It compiles successfully with all different versions
	of the fortran compiler I currently have so I'm not sure what the
	problem is/was.

	I suggest the followings:

	- check the version of the fortran compiler you have by typing
	'versions ftn'.
	[Rev C release should have 41 as the last two digits in the Version
	column.]

	- if your alpha number is less than 41 you can get Rev C from your
	local SGI office.   You might be able to get Rev D too.

	- if your alpha number is 41 or greater you might have problems with
	the distribution tape.   It sounds far-fetched but it's the only 
	thing I can think of.   You can try to get a replacement tape from
	the SGI office if that's the case.   I tend to believe this is what
	happened though since this is the first time we have ever heard of
	this problem.   It would have caused an uproar if this actually
	happened to everybody :-).


Calvin Vu
Silicon Graphics Computer Systems