home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.sys.intel
- Path: sparky!uunet!hobbes!md
- From: md@sco.COM (Michael Davidson)
- Subject: Re: Parity on iAPX x86 chips
- Organization: The Santa Cruz Operation, Inc.
- Date: Tue, 12 Jan 1993 19:58:05 GMT
- Message-ID: <1993Jan12.195805.19499@sco.com>
- References: <C0r3oq.4Hu@cs.mcgill.ca>
- Sender: news@sco.com (News admin)
- Lines: 17
-
-
- storm@cs.mcgill.ca (Marc WANDSCHNEIDER) writes:
-
- >I've noticed that most of the x86 chips I have programmed have a Parity
- >bit. I was just wondering if they actually compute this every time an
- >operation is done, or if they just get this from memory (ie, is the parity
- >bit valid after EVERY operation, or just loads from memory...?)
-
- The parity flag indicates the parity of the low 8 bits of the result
- of the last arithmetic or logical operation - it has nothing to do with
- memory parity and since move instructions do not affect the contents of the
- flags register a load from or a store to memory will not change the flag.
- However most arithmetic and logical instruction on the X86 will cause it
- to be updated - the only exceptions seem to be AAA, AAS, MUL, IMUL and the
- bit scan and bit test and set/clear/complement instructions after any of
- which the parity flag is undefined.
-
-