[comp.protocols.nfs] "Packed file is corrupt" error and MS-DOS 5.0

geoff@bodleian.East.Sun.COM (Geoff Arnold @ Sun BOS - R.H. coast near the top) (06/25/91)

Quoth victor@ccwf.cc.utexas.edu (Victor Menayang) (in <51169@ut-emx.uucp>):
#In article <1991Jun24.195425.19543@jarvis.csri.toronto.edu> west@turing.toronto.edu (Tom West) writes:
#>
#>  I have been getting "Packed file is corrupt" errors when trying to
#>run certain applications after installing MS-DOS 5.0.  Any idea what 
#>might cause this?  Specifically qbasic and PC-NFS telnet give this
#>error when I try to run them.
#>
#
#Perhaps the fix is simple: loadfix program.
#It was mentioned in one online doc in the Beta. You're right though, it has
#something to do with you loading DOS high (loadfix puts the program
#abovethe first 64k of conventional memory).

It seems that the bug is in the EXEPACK distributed with Microsoft C versions
before 6.0. The unpacking function assumes that the load address will
always be >64K. This was a reasonable assumption prior to DR-DOS and MS-DOS 5.0,
but no more. The immediate answer, as Victor pointed out, is LOADFIX. Long-term,
the solution is to build with MSC 6.x tools.

#I thought loadfix and wina20.386 would only be needed in the Beta
#versions :-)

But how....? After all, there's nothing in DOS 5.0 (beta or FCS) that can
change the way the EXEPACK-generated stuff works. (Or are you suggesting
that DOS 5.0 should run a signature check on an EXE, recognize that it's
EXEPACKed, and automatically LOADFIX it?)

Or did I misinterpret the smiley?

Geoff

--Geoff Arnold, PC-NFS architect(geoff@East.Sun.COM or geoff.arnold@Sun.COM)--
------------------------------------------------------------------------------
--       Sun Technology Enterprises : PC Networking group                   --
--   (officially from July 1, but effectively in place right now)           --