home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.dsp
- Path: sparky!uunet!psinntp!isc-newsserver!ritvax.isc.rit.edu!JXR1092
- From: jxr1092@ritvax.isc.rit.edu (Jens Rodenberg)
- Subject: Re: Adress of memory beyond 1M on PC
- Message-ID: <1992Jul27.011202.5006@ultb.isc.rit.edu>
- Sender: news@ultb.isc.rit.edu (USENET News System)
- Nntp-Posting-Host: vaxa.isc.rit.edu
- Reply-To: jxr1092@ritvax.isc.rit.edu
- Organization: Rochester Institute of Technology
- References: <1992Jul22.064616.3401@swanee.ee.uwa.oz.au>
- Date: Mon, 27 Jul 1992 01:12:02 GMT
- Lines: 27
-
-
- Zhang Ya Xin writes:
-
- > We have a TANG IBM/PC 386 which has 4M of memory. Normally the adress
- >of memory in the PC consists of a segment plus an offset, say, ****:****.
- >1M is the limitation. My question is how can I access the memory beyond 1M?
- >In other words, is there an adress of the memory beyond 1M in the PC?
-
- For my thesis (a prototype for a real-time image processing board), I
- needed to access memory above 1M on a 386, for storing images above 1M and
- also accessing the board's memory which is at 8M, just above the PC's
- memory. An MS-DOS function call (int 15H, AH = 87H) is all that's
- needed for the memory transfer, which can transfer a buffer to extended
- memory and vice-versa. The source and destination addresses are specified
- before the fuction call. More details and example assembly code can be
- found in an MS-DOS book, which I don't recall the name of (maybe it's just
- "MS-DOS"). A memory transfer of up to 64K can be done, but Turbo C wouldn't
- allow me to declare such a large array, so I had to go with 32K.
-
- Jens
-
- P.S. To any potential employers: I would like to get a DSP job, as my thesis
- has given me experience with the design, simulation, and construction of
- an image processing board. E-mail for more details if you're looking
- for a design engineer (BS in Electrical and Computer Engineering, almost
- MS in Computer Engineering). Thanks.
-
-