>Number:         7050
>Category:       pkg
>Synopsis:       Ghostscript 5.50 failing to start-up
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    gnats-admin (GNATS administrator)
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Feb 25 14:35:02 1999
>Last-Modified:
>Originator:     David Forbes
>Organization:
	
>Release:        1.3I fetched on 17th dec 1998
>Environment:
	
System: NetBSD crowley.our-flat.net 1.3I NetBSD 1.3I (CROWLEY) #0: Mon Jan 18 17:18:36 GMT 1999 david@crowley.our-flat.net:/usr/home/david/netbsd-latest/src/sys/arch/arm32/compile/CROWLEY arm32

NetBSD/arm32 


>Description:

Ghostscript 5.50 built from the pkgsrc tree fails to start-up:

crowley:david$ gs
crowley:david$

version 5.10 runs perfectly.

The only unusual thing I can find is the ktrace output, the relevant section
shown here (in the midst of reading
/usr/pkg/share/ghostscript/5.50/gs_init.ps)

          /userdict .systemvar
        } .bind odef
        /.uservar {             % <name> .uservar <value>
          .userdict exch get
        } .bind odef

        % If we're delaying binding, remember everything that needs to be
boun\
        d later.
        DELAYBIND NOBIND not and
         { .currentglobal false .setglobal
           systemdict /.delaybind 1500 array .forceput
           .setglobal
           userdict /.delaycount 0 put
                % When we've done the delayed bind, we "
  5934 gs       RET   read 8192/0x2000
  5934 gs       CALL  break(0x23bffc)
  5934 gs       RET   break 0
  5934 gs       CALL  lseek(0,0,0,0,0x1)
  5934 gs       RET   lseek 157774/0x2684e
  5934 gs       CALL  __fstat13(0,0xefbfcd94)
  5934 gs       RET   __fstat13 0
  5934 gs       CALL  break(0x24cffc)
  5934 gs       RET   break 0
  5934 gs       CALL  ioctl(0,TIOCGETA,0xefbfcdd8)
  5934 gs       RET   ioctl 0
  5934 gs       CALL  lseek(0,0,0x2684e,0,0)
  5934 gs       RET   lseek 157774/0x2684e
  5934 gs       CALL  close(0x3)
  5934 gs       RET   close 0
  5934 gs       CALL  exit(0xff)


As you can see the file appear to be truncated when reading although it
exists in it's entirety on my system (and has the appropriate permissions,
etc).  I can send the full kdump output is necessary.

David Forbes.
>How-To-Repeat:
>Fix:
>Audit-Trail:
>Unformatted: