hw/i386.reference/s16_02.htm

   1 <!DOCTYPE HTML PUBLIC "-//IETF//DTD HTML 2.0//EN">
   2 <HTML>
   3 <HEAD>
   4 <TITLE>80386 Programmer's Reference Manual -- Section 16.2</TITLE>
   5 </HEAD>
   6 <BODY>
   7 <B>up:</B> <A HREF="c16.htm">
   8 Chapter 16 -- Mixing 16-Bit and 32 Bit Code</A><BR>
   9 <B>prev:</B> <A HREF="s16_01.htm">16.1  How the 80386 Implements 16-Bit and 32-Bit Features</A><BR>
  10 <B>next:</B> <A HREF="s16_03.htm">16.3  Sharing Data Segments Among Mixed Code Segments</A>
  11 <P>
  12 <HR>
  13 <P>
  14 <H1>16.2  Mixing 32-Bit and 16-Bit Operations</H1>
  15 The 80386 has two instruction prefixes that allow mixing of 32-bit and
  16 16-bit operations within one segment:
  17 <UL>
  18 <LI> The operand-size prefix (66H)
  19 <LI> The address-size prefix (67H)
  20 </UL>
  21 <P>
  22 These prefixes reverse the default size selected by the D-bit. For example,
  23 the processor can interpret the word-move instruction <A HREF="MOV.htm">MOV</A> mem, reg in any of
  24 four ways:
  25 <UL>
  26 <LI> In a USE32 segment:
  27 <OL>
  28 <LI> Normally moves 32 bits from a 32-bit register to a 32-bit
  29 effective address in memory.
  30 <LI> If preceded by an operand-size prefix, moves 16 bits from a 16-bit
  31 register to 32-bit effective address in memory.
  32 <LI> If preceded by an address-size prefix, moves 32 bits from a 32-bit
  33 register to a16-bit effective address in memory.
  34 <LI> If preceded by both an address-size prefix and an operand-size
  35 prefix, moves 16 bits from a 16-bit register to a 16-bit effective
  36 address in memory.
  37 </OL>
  38 <LI> In a USE16 segment:
  39 <OL>
  40 <LI> Normally moves 16 bits from a 16-bit register to a 16-bit
  41 effective address in memory.
  42 <LI> If preceded by an operand-size prefix, moves 32 bits from a 32-bit
  43 register to 16-bit effective address in memory.
  44 <LI> If preceded by an address-size prefix, moves 16 bits from a 16-bit
  45 register to a32-bit effective address in memory.
  46 <LI> If preceded by both an address-size prefix and an operand-size
  47 prefix, moves 32 bits from a 32-bit register to a 32-bit effective
  48 address in memory.
  49 </OL>
  50 </UL>
  51 These examples illustrate that any instruction can generate any combination
  52 of operand size and address size regardless of whether the instruction is in
  53 a USE16 or USE32 segment. The choice of the USE16 or USE32 attribute for a
  54 code segment is based upon these criteria:
  55 <OL>
  56 <LI> The need to address instructions or data in segments that are larger
  57 than 64 Kilobytes.
  58 <LI> The predominant size of operands.
  59 <LI> The addressing modes desired. (Refer to
  60 <A HREF="c17.htm">Chapter 17</A> for an explanation
  61 of the additional addressing modes that are available when 32-bit
  62 addressing is used.)
  63 </OL>
  64 Choosing a setting of the D-bit that is contrary to the predominant size of
  65 operands requires the generation of an excessive number of operand-size
  66 prefixes.
  67 <P>
  68 <HR>
  69 <P>
  70 <B>up:</B> <A HREF="c16.htm">
  71 Chapter 16 -- Mixing 16-Bit and 32 Bit Code</A><BR>
  72 <B>prev:</B> <A HREF="s16_01.htm">16.1  How the 80386 Implements 16-Bit and 32-Bit Features</A><BR>
  73 <B>next:</B> <A HREF="s16_03.htm">16.3  Sharing Data Segments Among Mixed Code Segments</A>
  74 </BODY>