NU113738@NDSUVM1.BITNET (08/05/89)
Ok, maybe I'm doing something stupid that I don't recognize, maybe someone
can tell me what.
I've been messing around with some code to work with the EGA/VGA
multiple pages. I wrote 2 routines that look simple but aren't working
in the least.
Routine 1:
void set_display_page(int page){
union REGS regs;
regs.h.ah=0x05;
regs.h.al=page;
int86(0x10,®s,®s);
}
Simple enough, just int 10H to call Set Active Display Page.
Routine 2:
void set_write_ofs(int page){
#define PEEK_WORD(seg,ofs) (*(int far *) ((long)(seg) << 16 | (ofs)))
gr_page_ofs=(char far *)(0xA00000000L);
/* gr_page_ofs pointer to tell my line routine where to go. */
while (page){
gr_page_ofs+=PEEK_WORD(0x40,0x4C);
page--;
}
}
The PEEK_WORD define is straight out of MS DOS BIBLE by the Waite Group
and 0x40,0x4C is supposed to contain the page length of the current mode.
Problem: In 640x200 16 color (0x0E), The write function correctly
defines where to write to which page. But, the set_display_page function,
correctly shows page 0, but if called with set_dp(1) (page 1), it
displays the memory somewhat a page 1/2 would be like. Half of both
pages show up. If you call Page 2 with the display function, then
page 1 correctly shows up.
Now, in 640x350 mode, Page 0 is again correct, but trying to show
page 1 again gives a half page type output. But, if you try to call
Page 2, (which isn't supposed to exist), it still stays in the half
page output.
Question: Am I doing something extremely stupid that only I can't see..
or is this p.o.s. clone got a weirdo VGA board? Anyhelp would
be appreciated. If you've got a better method to direct the write and
display of pages I'd love to see it.
Also, if anyone has a good, fast, efficient algorithm or Assembly implementatio
n of a Page to Page copy function, I'd really like to get my hands on one.
The One I implemented is dogged...
Jeff Bakke
NU113738@NDSUVM1.BITNET
I appreciate any help whatsoever....
Thanx.