Velocity Reviews - Computer Hardware Reviews

Velocity Reviews > Newsgroups > Programming > VHDL > sane input

Reply
Thread Tools

sane input

 
 
shane
Guest
Posts: n/a
 
      10-03-2008
I'm designing a basic component, and I've realised that the input could be non
sensible. I have 3 input signals (not including a vector that inputs a value)
I want to be sure that the 3 input signals can never be actioned at the same
time (I want to avoid a race condition)
Could someone point me in the right direction?

TIA
--
Hardware n: Parts of the computer you can kick
 
Reply With Quote
 
 
 
 
KJ
Guest
Posts: n/a
 
      10-03-2008
On Oct 3, 1:57*am, shane <(E-Mail Removed)-a-geek.net> wrote:
> I'm designing a basic component, and I've realised that the input could be non
> sensible. *I have 3 input signals (not including a vector that inputs a value)
> I want to be sure that the 3 input signals can never be actioned at the same
> time (I want to avoid a race condition)


1. You don't control inputs, they control your design.
2. Simple logic, Bad_Input <= Inp1 and Inp2 and Inp3.
3. Race condition implies a path somewhere to a storage element but
you haven't given enough details to go much further.
4. Comparing the values of each of the inputs to what they were at
some previous time to see if they are all stable will filter out many
bogus inputs. How long there should be between 'now' and 'previous
time' depends on how long you expect that whatever is creating the
invalid inputs will be doing its thing. Increasing that time
increases your reliability, but at the same time decreases your
responsiveness.

Kevin Jennings
 
Reply With Quote
 
 
 
 
shane
Guest
Posts: n/a
 
      10-03-2008
Jonathan Bromley did scribble:

> On Fri, 03 Oct 2008 18:57:13 +1300, shane <(E-Mail Removed)-a-geek.net>
> wrote:
>
>>I'm designing a basic component, and I've realised that the input could be non
>>sensible. I have 3 input signals (not including a vector that inputs a value)
>>I want to be sure that the 3 input signals can never be actioned at the same
>>time (I want to avoid a race condition)
>>Could someone point me in the right direction?

>
> Only if you can be a bit more specific. What do you mean
> by "actioned"? Are these three signals all clocks, or are they
> enable or trigger signals that can be sampled by a single common
> clock? (I very much hope it's the latter.) More clues needed.


Hi, yes

I don't have a clock in my component ( I am very new to this so feel free to
correct me.. gently )

I have three trigger signals incoming, that my component will react to. My
concern is that I have no control on their timing, so am trying to design a
sane reaction(?)

The trigger signals instruct the component to do the following:
load a value from the vector signal
increment that value
write enable, allow the value stored in the component to be read from the
outside
--
Hardware n: Parts of the computer you can kick
 
Reply With Quote
 
Mike Treseler
Guest
Posts: n/a
 
      10-03-2008
shane wrote:

> I don't have a clock in my component ( I am very new to this so feel free to
> correct me.. gently )


1. Add a clock and use it to synchronize the inputs.
2. Write a synchronous process to do the loads, stores and increments.
Some examples:
http://mysite.verizon.net/miketreseler/

-- Mike Treseler
 
Reply With Quote
 
 
 
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
SANE 2006 early registration deadline is April 7, 2006 Edwin Kremer Computer Security 0 03-30-2006 11:38 AM
SANE 2006 Registration Information and Early Bird Deadline Edwin Kremer Computer Security 0 02-19-2006 09:17 PM
SANE 2006 refereed paper submission deadline is October 24, 2005 Edwin Kremer Computer Security 0 09-22-2005 09:14 PM
SANE 2006: Announcement and Call for Papers Edwin Kremer Computer Security 0 07-08-2005 04:22 PM
Python/PSF at SANE 2004 - Announcement and a request for help Jack Jansen Python 0 09-16-2004 07:48 PM



Advertisments