noise integration memory leak?

E

Erik Wanta

Guest
We are running Results->Print->Noise Summary after a spectre noise
simulation in IC5.0. It takes about 400MB of memory and then doesn't
give it back when it is done. Has anyone else seen similar behavior?

Note:
5.0.33.500.4
solaris 8
---
Erik
 
Erik,

There is a known problem with the speed of the noise summary in IC50, and I
expect that some of the reason for this is it using excessive memory. There's a
PCR for this, and I think it's being worked on to fix in a USR (from my memory).

I don't think the memory is leaking, per se, but normally if memory is used in
DFII (or any other executable) it would not be returned to the OS until the
program exits (that's the way many memory managers work), but it would
be available for reuse in DFII. However, it almost certainly shouldn't
be using that amount of memory in the first place.

Cheers,

Andrew.

On 3 Feb 2004 13:21:01 -0800, erikwanta@starband.net (Erik Wanta) wrote:

We are running Results->Print->Noise Summary after a spectre noise
simulation in IC5.0. It takes about 400MB of memory and then doesn't
give it back when it is done. Has anyone else seen similar behavior?

Note:
5.0.33.500.4
solaris 8
---
Erik
Click to expand...
--
Andrew Beckett
Senior Technical Leader
Custom IC Solutions
Cadence Design Systems Ltd
 
Andrew:
Is PCR 610211 the one you are talking about? From what I can gather
that is closed and fixed but we still have the problem.
---
Erik

Andrew Beckett <andrewb@DELETETHISBITcadence.com> wrote in message news:<7af2205bhv7i19077c693upftqtcsavo3g@4ax.com>...
Erik,

There is a known problem with the speed of the noise summary in IC50, and I
expect that some of the reason for this is it using excessive memory. There's a
PCR for this, and I think it's being worked on to fix in a USR (from my memory).

I don't think the memory is leaking, per se, but normally if memory is used in
DFII (or any other executable) it would not be returned to the OS until the
program exits (that's the way many memory managers work), but it would
be available for reuse in DFII. However, it almost certainly shouldn't
be using that amount of memory in the first place.

Cheers,

Andrew.

On 3 Feb 2004 13:21:01 -0800, erikwanta@starband.net (Erik Wanta) wrote:

We are running Results->Print->Noise Summary after a spectre noise
simulation in IC5.0. It takes about 400MB of memory and then doesn't
give it back when it is done. Has anyone else seen similar behavior?

Note:
5.0.33.500.4
solaris 8
---
Erik
Click to expand...
 
Erik,

Ah yes, that was what I was thinking about. So looks like a further problem -
you'll need to contact support via the usual channels.

Andrew.

On 4 Feb 2004 17:39:51 -0800, erikwanta@starband.net (Erik Wanta) wrote:

Andrew:
Is PCR 610211 the one you are talking about? From what I can gather
that is closed and fixed but we still have the problem.
---
Erik
Click to expand...
--
Andrew Beckett
Senior Technical Leader
Custom IC Solutions
Cadence Design Systems Ltd
 
Andrew:
SR32718562 opened on this issue.

If anyone else is seeing that IC50 is taking large amounts of memory
(much more than 446) when running Results->Print->Noise Summary please
open a similar SR so that we can get this issue fixed ASAP.
---
Erik

Andrew Beckett <andrewb@DELETETHISBITcadence.com> wrote in message news:<tlq320tniiau0kuh2in29au37thja54n7r@4ax.com>...
Erik,

Ah yes, that was what I was thinking about. So looks like a further problem -
you'll need to contact support via the usual channels.

Andrew.

On 4 Feb 2004 17:39:51 -0800, erikwanta@starband.net (Erik Wanta) wrote:

Andrew:
Is PCR 610211 the one you are talking about? From what I can gather
that is closed and fixed but we still have the problem.
---
Erik
Click to expand...
 
Andrew:
PCR686584 filed
---
Erik

Andrew Beckett <andrewb@DELETETHISBITcadence.com> wrote in message news:<tlq320tniiau0kuh2in29au37thja54n7r@4ax.com>...
Erik,

Ah yes, that was what I was thinking about. So looks like a further problem -
you'll need to contact support via the usual channels.

Andrew.

On 4 Feb 2004 17:39:51 -0800, erikwanta@starband.net (Erik Wanta) wrote:

Andrew:
Is PCR 610211 the one you are talking about? From what I can gather
that is closed and fixed but we still have the problem.
---
Erik
Click to expand...
 
The memory taken is not available for subsequent runs. That is
running Results->Print->Noise Summary a few times exhausts all my
memory.
---
Erik

erikwanta@starband.net (Erik Wanta) wrote in message news:<84018314.0402050732.335f2922@posting.google.com>...
Andrew:
PCR686584 filed
---
Erik

Andrew Beckett <andrewb@DELETETHISBITcadence.com> wrote in message news:<tlq320tniiau0kuh2in29au37thja54n7r@4ax.com>...
Erik,

Ah yes, that was what I was thinking about. So looks like a further problem -
you'll need to contact support via the usual channels.

Andrew.

On 4 Feb 2004 17:39:51 -0800, erikwanta@starband.net (Erik Wanta) wrote:

Andrew:
Is PCR 610211 the one you are talking about? From what I can gather
that is closed and fixed but we still have the problem.
---
Erik
Click to expand...
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top