home *** CD-ROM | disk | FTP | other *** search
- Path: li.net!usenet
- From: bsilvern@li.net (Bob Silvern)
- Newsgroups: comp.lang.c++
- Subject: Re: memory model question
- Date: Mon, 15 Jan 1996 06:47:44 GMT
- Organization: Harmony Graphics
- Message-ID: <4dctau$rdj@linet02.li.net>
- References: <4d94go$sp0@cantaloupe.srv.cs.cmu.edu>
- NNTP-Posting-Host: lisuser53.li.net
- X-Newsreader: Forte Free Agent 1.0.82
-
- bmm+@BMM.PC.CS.CMU.EDU (Bruce Maggs) wrote:
-
-
- >Hi,
-
- >I'm compiling a C program using Borland's C++ for DOS. It's a large
- >program, but doesn't use much static memory. I've selected the
- >"small" memory model. The program compiles and runs without any
- >problems, but the length of the .exe file that the compiler produces
- >is 148K. I thought that in the small memory model the length of the
- >code would be limited to one segment, i.e., 64K. I'm curious about
- >what's going on here. Does the compiler automatically split my code
- >into two or more segments?
-
- >Thanks,
-
- >Bruce Maggs
-
-
- One factor that can cause your exe to be large is the inclusion of debug
- information (variable names, addresses, etc.). Check your linker options and
- disable linking debug info. You won't be able to debug the program though
- unless you link with debug info included.
-
-
-