Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Can interconnection tracks (wires) between CLBs be controlled by more than 1 blocks?

Status
Not open for further replies.

Zhong Guan

Newbie level 2
Newbie level 2
Joined
Jun 26, 2013
Messages
2
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Visit site
Activity points
14
Can interconnection track (wire) between CLBs be controlled by more than 1 blocks?

I'm currently doing a research which requires me to find a unidirectional current path in FPGA.

My question is

Can interconnection track (wire) between CLBs be controlled by more than 1 buffer (CLB output)? If it is controlled by a single buffer, then definitely its current is bidirectional, which is sad for me...:cry:
 

Do you mean to ask if there can be more than 1 driver on an internal wire, as inside the fpga fabric? AFAIK all modern fpga's have only 1 active driver per interconnect. Not entirely sure if this is what you are asking though...

Can you describe in a bit more detail what you need that "unidirectional current path" for?
 
Do you mean to ask if there can be more than 1 driver on an internal wire, as inside the fpga fabric? AFAIK all modern fpga's have only 1 active driver per interconnect. Not entirely sure if this is what you are asking though...

Can you describe in a bit more detail what you need that "unidirectional current path" for?

That's exactly what I mean here.

I'm going to check if there is electromigration reliability issue (suppose to happen in P/G lines) in FPGA signal lines. :razz:

Thx a lot.
 

So something like take the longest routes from a routing box (because they have to drive harder due to higher capacitance compared to short routes). Then put them all at a static state, all high, all low, or probably high/low/high/low alternating for wires next to each other for best migration effect. ;) Make sure it doesn't get optimized away by the synthesizer. Then burn this sucker in at elevated temperature / voltage because we are all impatient. And then after a looong time plonk in a new bit stream with switching patterns and then see if you can get some errors? Sounds like fun! XD

The amusing part is going to be to figure out which logic lines are physically closest to each other on the die to optimize your migration fun.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top