Chinaunix首页 | 论坛 | 博客
  • 博客访问: 633517
  • 博文数量: 75
  • 博客积分: 7001
  • 博客等级: 少将
  • 技术积分: 1465
  • 用 户 组: 普通用户
  • 注册时间: 2007-07-11 17:39
文章分类

全部博文(75)

文章存档

2010年(1)

2009年(25)

2008年(49)

我的朋友

分类: LINUX

2008-07-29 20:00:34

How to comprehend the verb Assert and de-assert?
============================================================================================================================================
The engineers here use asserted and de-asserted to describe what happens to certain signals.
"De-asserted" is not defined in my Webster's 9th New Collegiate. Nor is
"asserted" defined in an electronics context. How is a signal "asserted"?
Most digital logic signals have two states. They also have names that
indicate the condition that the states represent. For example, a signal
called "Data Ready" might maintain one of two voltages: -3.3 volts to
indicate that signals on other lines contain valid data, 0 volts to indicate
that those other lines do not contain valid data.
In this example, you might write
The bus controller asserts Data Ready when the data on lines D0-D7 is stable.
Engineers sometimes use the term de-assert to indicate a return of the
signal to its other state. I prefer saying something like
The bus controller asserts Data Ready for as long as the data on D0-D7 is
stable.
The bus controller asserts Data Ready for 10ns when the data on D0-D7
becomes stable.
The bus controller asserts Data Ready when the data becomes stable, then
removes it when the data lines no longer contain valid data. ...RM
----------------------------------------------------------------------------------------------------------------------------------------------------------------
If you do not use the word deasserted, and the engineer catches it, he
will simply mark it to be changed back to deasserted. I belive that there
is a technical reason for the use of teh word deasserted.
The "technical reason" has to do with clarity. Time was when you simply wrote
such-and-such a line (or signal) was "raised" or "held high." This referred to
the voltage on the line ("high" voltage was a "1" or "true" in digital terms).
But design constraints caused some signal lines to assert the condition with a
low (0 or "false") signal. Traditionally this was signified by drawing a
horizontal line over the signal name (the traditional logical symbol for "not").
"Read" with a line above it was pronounced "not Read."
This is hard to get across in ASCII documentation (used frequently for working
notes) and it's quite possible to misunderstand it even in typeset documents.
(Imagine the condition above: When the microprocessor raised the "not Read" line
it signalled a write operation. Now think of reading a statement to that effect
in an engineer's ASCII design notes file, without the benefit of the horizontal
line -- "When the microprocessor raises the Read line it signals a write
operation.")

Have fun,
Arlen
----------------------------------------------------------------------------------------------------------------------------------------------------------------
The verb "assert" in this context is a hardware reference, meaning to
drive the voltage high. "De-assert" is acceptable, but if you dislike
it, see if you can get the engineers to use "drop" instead.
--
--- Aahz (@netcom.com)
 
 
 
阅读(2042) | 评论(0) | 转发(0) |
给主人留下些什么吧!~~