MASM Optimizations Not Recognized by Some Processors

Last reviewed: January 23, 1995
Article ID: Q69987
The information in this article applies to:
  • Microsoft Macro Assembler for MS-DOS, versions 5.1, 5.1a, 6.0, 6.0a, 6.0b, 6.10, 6.10a, and 6.11
  • Microsoft Macro Assembler for OS/2, versions 5.1, 5.1a, 6.0, 6.0a, and 6.0b

SUMMARY

For optimization reasons, MASM may generate the opcode 83 for logical AND, OR, and XOR instructions in some cases, rather than opcode 81. Unfortunately, opcode 83 was not documented by Intel for 80x86/8088 processors prior to the 80386. Therefore, some processors (such as the NEC V25 and V35 controllers) and some in-circuit emulators for the 80x86 family do not support this opcode.

In the sample code below, "Listing 1" shows code that will cause the 83 opcode to be generated. "Listing 2" illustrates a way to work around this situation for those who are using processors that do not support this opcode and who are using versions of the assembler prior to 6.0. Note that opcode 83 does work properly on all Intel 80x86/8088 processors, and its generation is by design.

The sign-extension optimization may be disabled in MASM 6.0 and later with the OPTION directive. To override the default sign-extended opcodes for AND, OR, and XOR, place the following line at the beginning of the source file:

   OPTION NOSIGNEXTEND

MORE INFORMATION

The opcode 83 takes advantage of the sign-extension feature available on the Intel 80x86 processors. If the high byte of a word has all bits set or all bits cleared, and the most significant bit of the low byte is in the same state as the bits in the high byte, then the word may be represented by one byte. The opcode 83 instructs the machine to build the word by taking the byte and extending the sign bit (most significant bit) of the byte into the rest of the word.

For example, if the word is FFF0h, the word may be represented by F0h. By the same method, if the word is 000Fh, it may be stored as 0F.

The work-around for MASM versions prior to 6.0 shown below works because pass one of the assembler does not know if the second operand will meet the above criteria, and will generate the opcode 81. Zero is defined as the number 0, so the operand value does not change.

Sample Code

;Listing 1:

.MODEL small .DATA val DW 0

.CODE mine PROC

    and val, 0FFF0h
mine ENDP END

;Listing 2:

.MODEL small .DATA val DW 0

.CODE mine PROC

    and val, ZERO + 0FFF0h
mine ENDP

ZERO EQU 0 END


Additional reference words: kbinf 5.10 5.10a 6.00 6.00a 6.00b 6.10 6.10a
KBCategory: kbprg
KBSubCategory: MASMLngIss


THE INFORMATION PROVIDED IN THE MICROSOFT KNOWLEDGE BASE IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND. MICROSOFT DISCLAIMS ALL WARRANTIES, EITHER EXPRESS OR IMPLIED, INCLUDING THE WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL MICROSOFT CORPORATION OR ITS SUPPLIERS BE LIABLE FOR ANY DAMAGES WHATSOEVER INCLUDING DIRECT, INDIRECT, INCIDENTAL, CONSEQUENTIAL, LOSS OF BUSINESS PROFITS OR SPECIAL DAMAGES, EVEN IF MICROSOFT CORPORATION OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. SOME STATES DO NOT ALLOW THE EXCLUSION OR LIMITATION OF LIABILITY FOR CONSEQUENTIAL OR INCIDENTAL DAMAGES SO THE FOREGOING LIMITATION MAY NOT APPLY.

Last reviewed: January 23, 1995
© 1998 Microsoft Corporation. All rights reserved. Terms of Use.