dpz@dimacs.rutgers.edu (David Paul Zimmerman) (05/14/91)
I had dreamt it would be simple... just get the sources, run them through Turbo C, and voila -- my very own KA9Q net.exe. And so it was... but twice the size of the one Phil distributes!! I've diddled some obvious compilation and linking flags, to no avail. Anyone have any ideas? If this were Unix, I'd think of doing a "strip net.exe", but there doesn't seem to be an analogy. David -- David Paul Zimmerman dpz@dimacs.rutgers.edu Systems Programmer rutgers!dimacs.rutgers.edu!dpz Center for Discrete Mathematics and Theoretical Computer Science (DIMACS)
w8sdz@rigel.acs.oakland.edu (Keith Petersen) (05/14/91)
dpz@dimacs.rutgers.edu (David Paul Zimmerman) writes: >I had dreamt it would be simple... just get the sources, run them through >Turbo C, and voila -- my very own KA9Q net.exe. And so it was... but twice >the size of the one Phil distributes!! I've diddled some obvious compilation >and linking flags, to no avail. Anyone have any ideas? If this were Unix, >I'd think of doing a "strip net.exe", but there doesn't seem to be an analogy. No analogy, just a compression program for MS-DOS executables. It saves disk space but when the program is running it takes just as much space in memory as before. WSMR-SIMTEL20.ARMY.MIL [192.88.110.20] Directory PD1:<MSDOS.FILUTL> Filename Type Length Date Description ============================================== PKLTE105.EXE B 50330 910424 PKLITE v1.05, compress EXE/COM files and run Keith - - - Keith Petersen Maintainer of SIMTEL20's MSDOS, MISC and CP/M archives - [192.88.110.20] Internet: w8sdz@WSMR-SIMTEL20.Army.Mil or w8sdz@vela.acs.oakland.edu Uucp: uunet!umich!vela!w8sdz BITNET: w8sdz@OAKLAND
a0045@rrz.uni-koeln.de (Jochen Roderburg) (05/14/91)
Not a great mystery, the distribution version of NET.EXE is compressed with one of the executable file compressors (PKLITE or such). If you unpack it again, it amounts to the same size than yours. J. Roderburg, U of Cologne, Germany
kurt@photon.tamu.EDU (Kurt Freiberger) (05/15/91)
In article <May.13.22.55.22.1991.7896@dimacs.rutgers.edu>, dpz@dimacs.rutgers.edu (David Paul Zimmerman) writes: |> I had dreamt it would be simple... just get the sources, run them through |> Turbo C, and voila -- my very own KA9Q net.exe. And so it was... but twice |> the size of the one Phil distributes!! I've diddled some obvious compilation |> and linking flags, to no avail. Anyone have any ideas? If this were Unix, |> I'd think of doing a "strip net.exe", but there doesn't seem to be an analogy. |> |> David |> -- |> David Paul Zimmerman dpz@dimacs.rutgers.edu |> Systems Programmer rutgers!dimacs.rutgers.edu!dpz |> Center for Discrete Mathematics and Theoretical Computer Science (DIMACS) Phil uses PKLITE which compresses the program and attaches a header that decompresses it into memory at runtime. kurt -- Kurt Freiberger, wb5bbw kurt@cs.tamu.edu 409/847-8706 Dept. of Computer Science, Texas A&M University DoD #264 *** Not an official document of Texas A&M University ***
acm@RELAY.PROTEON.COM (05/15/91)
David,
I had dreamt it would be simple... just get the sources, run them
through Turbo C, and voila -- my very own KA9Q net.exe. And so
it was... but twice the size of the one Phil distributes!! I've
diddled some obvious compilation and linking flags, to no avail.
Anyone have any ideas? If this were Unix, I'd think of doing a
"strip net.exe", but there doesn't seem to be an analogy.
All you have to do is selcet /dsn for debug, source, none. This will
make a much smaller object file to be linked. If source is on it will
have lots of stuff that is useful for debugging but nothing else.
-Alan Marshall, Proteon
CCMAIL: acm at proteonwebo tel: (508)898-2120
INTERNET: acm@Proteon.com MHS: acm @ ProteonWdpz@dimacs.rutgers.edu (David Paul Zimmerman) (05/15/91)
Well THAT explains it. Much thanks to Scott Laird, Rich Warwick, Jochen Roderburg, Jim Mankin, Keith Petersen, and Kurt Freiberger for setting me straight on the executable compression. My net.exe does indeed take up less than half the disk space now, though memory usage is still the same, as expected. Rich Warwick -- I'll try your #undef ideas... I would indeed like to get some memory back. Alan Marshall -- I appreciate your thoughts, but I'm not using the Turbo menu interface, so your hints don't seem to apply.... David -- David Paul Zimmerman dpz@dimacs.rutgers.edu Systems Programmer rutgers!dimacs.rutgers.edu!dpz Center for Discrete Mathematics and Theoretical Computer Science (DIMACS)