Here's a algorithm for a short locator:

A first crude approximation to this function might go along with the code that 
highlights nets. 
 It updates as soon as you add a physical object and rerun it -- it does not depend on 
the 
netlist for that, so maybe it has possibility for adding to.   Net highlighting must 
find connectivity by overlap
just like VLS tools do.  They compare 2D position of areas for deciding overlap.  We 
already have that 
happening, so finding that in the code is first step to adding a short locator that 
highlights 
only the "in between" nets.    

If the highlighting code can be isolated from netlist highlighting, we can reuse it to 
identify
overlap with selected objects, just like LVS by Cadence et al.   To simply show a 
"movie" of 
next overlapping and therefore connected objects from the two selected ones would be 
helpful.
The "movie" could be played a step at a time or int sets of 4 or 7 steps at a 
time...until shorted.
This would not always show the minimum path.   Human processor power can be used next 
to choose 
a better starting pair of objects and play or fast forward another "until shorte" 
movie...stopping
and NOT displaying any more objects than necessary to connect the selected two for the 
first time.

John G

paraphrased from Dan M:
[A high dollar LVS][jg] tool offered by cadence, has
> a really nifty short locator.

>  you select 2 objects, typically 2 metal segments,
> and run the short locator.  What it does is to highlight the path by which those two 
> metal segments are connected.[leaving out anything not "in between" -- a tough 
> definition 
for a cad system...depends a lot on the data format how easy or hard it
is.][jg]

>   It even has a feature where you keep hitting a next button
> and it moves the zoom and additional highlight to the next segment in the connecting
> path.  I've been able to find shorts with this tool in minutes that could have 
> potentially
> taken days of manually looking over the layout. <snip>
> -Dan
-- 
John Griessen    Cibolo Design       Austin Texas, linux counter #249315

Reply via email to