cell libraries and place and route

elektroda.net NewsGroups Forum Index - Cadence - cell libraries and place and route

Goto page Previous 1, 2, 3 ... , 51, 52, 53 Next

Andrew Beckett
Guest

Wed Mar 04, 2009 2:07 pm

reubenwilcock_at_gmail.com wrote, on 03/03/09 15:44:

Quote:

Unfortunately this problem has returned and now even deleting
the .ws_states file does nothing to help the situation.

Generally if I go to direct plot a result then the waveform window
that opens is completely blank (because the window items are so badly
misplaced that they lie out of the viewable area)....

Any suggestions for debugging?

Reuben

Reuben,

Apart from using a newer subversion, I don't have much in the way of suggestions.

You're using the IC5141 Base release, which came out nearly 5 years ago, and
there have been a _lot_ of hotfixes and update releases since then! I'm amazed
that anyone is still using the Base release...

Is it possible to get a newer subversion? EuroPractice normally updates to the
latest USR for each year's software release.

Regards,

Andrew.

spectrallypure
Guest

Wed Mar 04, 2009 2:26 pm

Thanks so much for your reply, Andrew (and sorry for the delayed
response!). I tried your solutions and obtained these results:

1. [Variables in chart title]: I decided to use Skill for creating/
setting the chart titles, and your function abShowAllVars() solves the
problem nicely; thank you!

2. [Setting the deafult number of strips in chart] I haven't been able
to solve this problem. I added the suggested entry in my .cdsenv, but
nothing seems to happen, and the charts still show 5 strips by
default. I further tried setting/reading the suggested variable from
the CIW, but it doesn't seem to exist. This is what I get:

\i envSetVal("wavescan.rectGraph" "visibleStripChartRows" 'string
"10")
\w *WARNING* envSetVal: Can't set the value of variable
'visibleStripChartRows',
\w in tool[.partition] 'wavescan.rectGraph' - it has not been
registered.
\t nil

\i envGetVal("wavescan.rectGraph" "visibleStripChartRows")
\w *WARNING* envGetVal: Could not find variable
'visibleStripChartRows'
\w in tool[.partition] 'wavescan.rectGraph'.
\t nil

I am using icfb sub-version 5.10.41.500.3.49. What could be going
wrong?

Thanks again for any further ideas!

Regards,

Jorge.

spectrallypure
Guest

Wed Mar 04, 2009 4:41 pm

OK Andrew; thanks anyway for saving me from wasting more time on this.

Quote:

So you're unfortunate -
I know... Smile

Cheers,

Jorge.

I-F AB
Guest

Thu Mar 05, 2009 4:46 am

Quote:

The first ONLY gets the symContactDevice contacts. The second will additionally
return ruleContactDevice and cdsViaDevice contacts.

Regards,

Andrew.

Many thanks.
I've found a techfile which has both cdsViaDevice & symContactDevice
contact definitions.
Now I can try them out.

Best Regards,
I-FAB

S. Badel
Guest

Thu Mar 05, 2009 11:26 am

Quote:

If the cellView is open readonly, geGetEditCellView() still returns
it, because it's the one that the "editor" is working on.

Andrew, I figured what's got me suspicious about geGetEditCellView().

When called interactively (CIW) it often returns unexpected cellview ids. This has probably more to
do with what is assumed to be the current window (I'd assume the last one used, but it seems kind of
random). I wrongly attributed that to my opening some cellviews in read-only mode. Again, thanks for
correcting.

Stéphane

Andrew Beckett
Guest

Thu Mar 05, 2009 7:30 pm

S. Badel wrote, on 03/05/09 10:26:

Quote:

If the cellView is open readonly, geGetEditCellView() still returns
it, because it's the one that the "editor" is working on.

Andrew, I figured what's got me suspicious about geGetEditCellView().

When called interactively (CIW) it often returns unexpected cellview
ids. This has probably more to do with what is assumed to be the current
window (I'd assume the last one used, but it seems kind of random). I
wrongly attributed that to my opening some cellviews in read-only mode.
Again, thanks for correcting.

Stéphane

Stéphane,

You can see which window is current. In IC5141 it has the banner (just above the
pulldown menus) shown in dark grey rather than light grey (it shows a line "Cmd:
.... Sel: 0 ..." in the schematic window).

A window is made current typically by hitting a bindkey in it (or even just the
shift or control key) or using a pulldown menu.

hiGetCurrentWindow() will return the windowId of the current window. This will
return something like:

window:5

where 5 is the number shown in the top right of the window.

In IC613, there's a little square in the bottom left corner of each window which
becomes pink when the window is current. The square has both the session window
and window numbers in (because you have multiple tabs in windows).

So it's not (quite) random ;->

It might seem that way though...

Andrew.

Linus Torwalds
Guest

Mon Apr 06, 2009 3:34 am

New Book Service wrote:

Quote:

Just thought people might be intrested in this new free book service. Here is a link to a signup page http://www.eyemagic.com or it is offered free at about 4,000 libraries in the US. I really enjoy it!

We have released 2009 version of our

universal CAD and drafting solution for Linux.

www.LinuxCAD.com

check it out, it's been 15 years since it was

out initially.

Is a match for AutoCAD 2009 now.

Andrew Beckett
Guest

Mon Apr 20, 2009 10:35 pm

hamidrezah_at_yahoo.com wrote, on 04/19/09 16:38:

Quote:

I do not get pin names and symbol name when I generate a symbol view
from a schematic view. I only get a box with red squares (supposedly
pins) around it (but no name associated with box or squares).

Any idea what is wrong?

Hrh

Have you got some strange template loaded? When you do Create CellView from
CellView, make sure that the "Edit Options" is checked, and then on the Symbol
Generation Options, check the "Load/Save" checkbox. See what Template path it
has, and try loading "analog" or "digital" or some other template from the
cyclic choice next to the "Load" button.

In the CIW, you can type

tsgTemplateMasters

to see what choices are available, and

envGetVal("schematic" "tsgTemplateType")

to show the default.

Perhaps you can report back here what your settings are, and what happens when
you do the steps I mention above on the create cellView from cellView forms?

Regards,

Andrew.

Andrew Beckett
Guest

Wed May 27, 2009 8:57 am

Riad KACED wrote, on 05/13/09 00:39:

Quote:

Hi,

Yes, there are loads of docs out there. You don't need them all
though. I have working with Cadence for 8 years and I'm still using
only a dozen of docs out of the ~100 available. The following is very
good for beginners:
http://cadence.wikispaces.com/Documentation+index

I would advice to use the cdnshelp/cdsdoc.

Regards,
Riad.

I just updated that wiki page to mention cdnshelp. I always use cdnshelp with
IC5141 (cdnshelp is in all current streams - it's not in IC5141 because it's too
old - however, you can use cdnshelp from one stream to open the help from another).

In order to prepare IC5141 for use with cdnshelp, you need to do:

cdnshelp -refresh -hierarchy $ICHOME/doc

(you'll need write access to the installation) and then follow the prompts.

Afterwards, users can then use cdnshelp and do File->Library->Open Library and
point to the $CICHOME/doc/xmlreg/library.lbr file that got created by the above
command. Having added it once, it will appear in the list of libraries.

cdnshelp doesn't involve Java, has a built-in search engine, and doesn't need a
separate web browser. Much easier to use and more reliable.

Regards,

Andrew.

terminator
Guest

Sun Jun 21, 2009 9:53 am

On Jun 15, 9:28 pm, terminator <raghavanamrith...@gmail.com> wrote:

Quote:

when i open my library manager i want the cell names to be sorted with
the latest modified cell view first and that modified long back at the
end. How to write a code in skill to accomplish this?

thank you,

Is the above operation not possible?

Andrew Beckett
Guest

Tue Jun 23, 2009 4:32 pm

terminator wrote, on 06/21/09 08:53:

Quote:

On Jun 15, 9:28 pm, terminator <raghavanamrith...@gmail.com> wrote:
when i open my library manager i want the cell names to be sorted with
the latest modified cell view first and that modified long back at the
end. How to write a code in skill to accomplish this?

thank you,

Is the above operation not possible?

No, it's not possible, at least not without writing your own library manager.

Regards,

Andrew.

Charley
Guest

Tue Jul 21, 2009 9:22 am

On 7æœˆ20æ—¥, ä¸‹åˆ8æ™‚40åˆ†, Andrew Beckett <andr...@DcEaLdEeTnEcTe.HcIoSm>
wrote:

Quote:

Charley wrote, on 07/20/09 08:41:

Hi All,

I have a Â cell with hierarchy e.g. A (cellview) -- B(cellview) -- C
(cellview )
There are 3 paths inside C (cellview) and Â I want do re-draw (By
dbCopyFig or dbCreatePath) Â the same path on the top level.

which Â rotation Â ï¼ŒA is Â "R0" ã€B is "R0 " & C is "R90"

How do write the code ?

Thank youï¼Œ
Charley

Charley,

You'd need to build the overall transform by using dbConcatTransform() with each
of the ~>transform attributes of the instances listed in geGetInstHierPath().

In the example below I kept it simple by not worrying about mosaics (arrays).
That makes it a bit more complicated, because you need to factor the position
within the array into the transformation. But I'll leave that as an exercise for
you if you're even using mosaics...

; this code makes absolutely no attempt to deal with mosaics...
xform=list(0:0 "R0" 1)
; reverse the list to get the transform the right way around
foreach(instPath reverse(geGetInstHierPath(hiGetCurrentWindow()))
Â Â xform=dbConcatTransform(xform car(instPath)~>transform)
)

; now copy selected shapes to the top cellView, with the right transformation
topCv=geGetTopLevelCellView()
newShapes=foreach(mapcar shape geGetSelSet()
Â Â dbCopyFig(topCv shape xform)
)

Regards,

Andrew.

Hi Andrew,

It works . That is what I need .
PS : dbCopyFig(topCv shape xform) ==> dbCopyFig(shape topCv xform)

Thank you,
Charley

Andrew Beckett
Guest

Tue Jul 21, 2009 10:30 am

Charley wrote, on 07/21/09 08:22:

Quote:

; now copy selected shapes to the top cellView, with the right transformation
topCv=geGetTopLevelCellView()
newShapes=foreach(mapcar shape geGetSelSet()
dbCopyFig(topCv shape xform)
)

Regards,

Andrew.

Hi Andrew,

It works . That is what I need .
PS : dbCopyFig(topCv shape xform) ==> dbCopyFig(shape topCv xform)

Thank you,
Charley

Charley,

Apologies for that last bit. That bit of the example I'd not actually tested,
and so made a typo when putting it together, and got the args back to front.

Regards,

Andrew.

Tarek
Guest

Tue Oct 20, 2009 2:34 pm

I just looked a bit closer at your save statement. Try "save
M*:oppoint", or simply "save *:oppoint". refer to "spectre -h save"
for options.

Guest

Tue Oct 20, 2009 7:16 pm

Thank you for your replies. Unfortunately I don't think it is really a
problem of how one defines the save oppoint. Because with the way I do
it I have no problems seeing the gds curve in wavescan - so, it
extracts the data from the DC sweep. It is just that when I try to put
it in the calculator and then plot back in wavescan, that it kind of
"loses" it. But I'll try what you recommended.

On Oct 20, 5:34 am, Tarek <tarek.k...@gmail.com> wrote:

Quote:

I just looked a bit closer at your save statement. Try "save
M*:oppoint", or simply "save *:oppoint". refer to "spectre -h save"
for options.

Goto page Previous 1, 2, 3 ... , 51, 52, 53 Next

elektroda.net NewsGroups Forum Index - Cadence - cell libraries and place and route