home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.ms-windows.programmer.tools
- Path: sparky!uunet!cis.ohio-state.edu!zaphod.mps.ohio-state.edu!sdd.hp.com!mips!mips!sjsca4!rushmore!poffen
- From: poffen@rushmore (Russ Poffenberger)
- Subject: Re: MFC under BCC?
- Message-ID: <1992Aug14.202143.25307@sj.ate.slb.com>
- Sender: news@sj.ate.slb.com
- Organization: Schlumberger Technologies, ATE division, San Jose, Ca.
- X-Newsreader: Tin 1.1 PL4
- References: <1992Aug13.211344.10040@odi.com>
- Date: Fri, 14 Aug 92 20:21:43 GMT
- Lines: 24
-
- ed@odi.com (Ed Schwalenberg) writes:
- : In article <1992Aug11.020925.18514@microsoft.com> johnkal@microsoft.com (John Kallen) writes:
- :
- : Borland and Microsoft have chosen two different name-mangling schemes for
- : symbols. As you mention, this is unfortunate as it makes linking to .LIB
- : libraries (as opposed to libraries where the source is provided) impossible
- : if the library was built with the "wrong" version.
- :
- : With luck, a standard DOS/Win/OS-2 name-mangling will evolve. Until
- : then, one will have to have sources to libraries, or compiler-vendor-dependent
- : .LIB files (bleargh...)
- :
- : Let's assume that a standard name-mangling exists. You're still in the soup
- : unless object layout (alignment rules, location of vptrs, size of vtbls, etc)
- : is identical and calling conventions are identical.
-
- This is true, even using straight 'C', compiling as only 'C' code, I have never
- been able to mix Borland and MSC object modules. They link OK, but always
- bomb out when executing.
-
- Russ Poffenberger DOMAIN: poffen@sj.ate.slb.com
- Schlumberger Technologies UUCP: {uunet,decwrl,amdahl}!sjsca4!poffen
- 1601 Technology Drive CIS: 72401,276
- San Jose, Ca. 95110 Voice: (408)437-5254 FAX: (408)437-5246
-