Xref: news2.mv.net comp.graphics.algorithms:20780 comp.os.msdos.djgpp:4184
Newsgroups: comp.os.msdos.djgpp,comp.graphics.algorithms
From: penquerc AT enssat DOT fr (Vincent Penquerc'h)
Subject: Re: Inline ASM, Mode 13h problem
Message-ID: <1996May23.061756.10087@cnet.fr>
Sender: usenet AT cnet DOT fr
Nntp-Posting-Host: lat1072.lannion.cnet.fr
Organization: CNET
References: <31A26377 DOT 29C6 AT postoffice DOT ptd DOT net> <4nva9j$5l5 AT wapping DOT ecs DOT soton DOT ac DOT uk>
Date: Thu, 23 May 1996 06:17:56 GMT
Lines: 33
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

In article <4nva9j$5l5 AT wapping DOT ecs DOT soton DOT ac DOT uk>, rjc94 AT ecs DOT soton DOT ac DOT uk (Robert Cooper) says:
>
>
>>void putpixel(short x,short y,short c) {
>>  __asm__ __volatile__("
>>    movw %0, %%es;
>>    movl $0xA0000, %%edi;
>>    movw %2, %%ax;
>>    imulw $320, %%ax;
>>    addw %1, %%ax;
>>    addw %%ax, %%di;
>>    movb %3, %%es:(%%edi);"
>>    :
>>    : "g" (glib_selector), "g" (x), "g" (y), "g" (c)
>>    );
>>I'm developing under RHIDE beta4.  I get an assembler error in the 
>>putpixel function, and I believe it is with the following line:  
>>       movb %3, %%es:(%%edi);"  
>
>>ERROR:  OPERANDS GIVEN DON'T MATCH ANY KNOWN 386 INSTRUCTION.
>
>Hmm.  Shouldn't those commands be something like movsb/movsw.  Rather 
>than movb/movw.  I think that may be right, but then again, I don't know 
>a huge amount about ASM either.
>
>Ica

Hmmm ... Are these 80x86 opcodes ?? Uses es, edi, but seem so weird ...
I think operands are in reversed order. Where does this assembler
come from ???

Vincent 'Indy' Penquerc'h
penquerc AT enssat DOT fr