home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!iadpsa!iowegia!james
- From: james@iowegia.uucp (james shoemaker)
- Newsgroups: comp.os.os2.advocacy
- Subject: Re: NTs thunking overhead more than OS/2s?
- Message-ID: <uXeqwB2w165w@iowegia.uucp>
- Date: Fri, 01 Jan 93 22:13:53 CST
- References: <C057uA.5wA@news.iastate.edu>
- Distribution: world
- Organization: Iowegia Public Access Usenet/UUCP, Clive IA USA.
- Lines: 21
-
- W.FY4@isumvs.iastate.edu (Timothy I Miller) writes:
-
- > I was reading a little bit of Dr. Dobbs Programming Journal (or
- > something like that) today, and it had an article about programming
- > in 32 bits. Although OS/2 is only mentioned a couple of times (it
- > was mostly about writing Win32 programs from 16-bit programs), what
- > was said about it wasn't bad. One very interesting thing was that
- > it said that there was more overhead in the Windows NT thunking than
- > there is in OS/2's thunking. Can this be verified? And if this is
- > true, why?
- >
- > Timothy Miller
- >
-
- Sounds reasonable. remember OS/2 has the 512 meg limit to facilitate
- easy thunking. NT has no such limit so a thunk probably involves creating a
- new selector that contains the address you want a 16-bit pointer to. going
- the other way is also more complicated. for similar reasons.
-
- JWS
-
-