arden@creed.tcom.stc.co.uk (Bruce Arden) (03/17/87)
I want to have two macros P_SEG(p) and P_OFF(p) (cf FP_SEG &
FP_OFF) that will let me get the segment and offset of p
independant of the memory model. Here is an example of their
intended use:
char scr[SCRSIZE];
movedata(0xB000, 0x0000, P_SEG(scr), P_OFF(scr), sizeof scr);
I thought I had the answer to this question:
#define P_SEG(p) (unsigned)((long)( (char far *)(p) ) >> 16)
#define P_OFF(p) (unsigned)((long)( (char far *)(p) ))
This appeared to give the correct result (although it did
generate some redundant code) with data pointers. However it
gave register DS as the segment of a near code pointer.
Has anyone solved this problem?
Thanks in advance
--
Bruce Arden
arden@creed.tcom.stc.co.ukmouse@mcgill-vision.UUCP (03/24/87)
In article <247@creed.tcom.stc.co.uk>, arden@creed.tcom.stc.co.uk (Bruce Arden) writes: > #define P_SEG(p) (unsigned)((long)( (char far *)(p) ) >> 16) > #define P_OFF(p) (unsigned)((long)( (char far *)(p) )) > This appeared to give the correct result (although it did generate > some redundant code) with data pointers. However it gave register DS > as the segment of a near code pointer. I don't know the particular compiler, but I'm not surprised. After all, you *are* casting it to a data pointer! Try something like #define CODEP_SEG(p) (unsigned)((long)( (int (far *)())(p) ) >> 16) #define CODEP_OFF(p) (unsigned)((long)( (int (far *)())(p) )) for code pointers. der Mouse Smart mailers: mouse@mcgill-vision.uucp USA: {ihnp4,decvax,akgua,utzoo,etc}!utcsri!musocs!mcgill-vision!mouse think!mosart!mcgill-vision!mouse ARPAnet: think!mosart!mcgill-vision!mouse@harvard.harvard.edu
srp@ethz.UUCP (03/29/87)
In article <703@mcgill-vision.UUCP> mouse@mcgill-vision.UUCP writes: > > der Mouse Shouldn't that be "die Maus" ? -- ----------- Scott Presnell Swiss Federal Institute of Technology (ETH-Zentrum) Department of Organic Chemistry Universitaetsstrasse 16 CH-8092 Zurich Switzerland. uucp: ...seismo!mcvax!cernvax!ethz!srp (srp@ethz.uucp) earn/bitnet: Benner@CZHETH5A