delorie.com/archives/browse.cgi | search |
From: | tom burgess <tburgess AT drao DOT nrc DOT ca> |
Newsgroups: | comp.os.msdos.djgpp |
Subject: | Re: asm xor problems |
Date: | Sun, 05 Oct 1997 01:09:42 -0700 |
Organization: | BCTEL Advanced Communications |
Lines: | 15 |
Message-ID: | <34374B46.64B2@drao.nrc.ca> |
References: | <EHI7xn DOT FFs AT nonexistent DOT com> <3436AD7B DOT CA40B171 AT worldonline DOT nl> <EHJt2K DOT 5wv AT nonexistent DOT com> |
NNTP-Posting-Host: | pntn02m01-28.bctel.ca |
Mime-Version: | 1.0 |
To: | djgpp AT delorie DOT com |
DJ-Gateway: | from newsgroup comp.os.msdos.djgpp |
Arthur Bredrick wrote: > > Thanks. Would "and memory, 0" be any faster than using mov? I don't really > know what takes the cpu longer to execute, or even if one would take any > longer than the other at all. But if there is even a tiny difference in > speed, I would like to know. > > Thanks again, > Art mem = mem & 0 requires both a read and write to memory, whereas mem = 0 just requires a write. Logically, the second should be faster, though the only way to know for sure is to benchmark your code. regards, tom
webmaster | delorie software privacy |
Copyright © 2019 by DJ Delorie | Updated Jul 2019 |