Experimental Physics and
Industrial Control System

"Andreas Isenegger" <[email protected]> · Thu, 29 Sep 2005 09:38:17 +0200

It's TSR# 453418.
I don't send copies of the files around. They are property of WRS. It is
[email protected] who sent them to SLS. Unfortunately, he
didn't add a revision number to the files when he changed them recently. So,
the latest revision entries in the files are

dec21x40End.h 22sep03,mni added BOOL restarting member to
DEC21X40_DRV_CTRL structure
dec21x40End.c 22sep03,mni added fix for SPR# 84658
Please check with Mr Niedermair if he did essential changes since.
Kind regards,
Andy
----------------------------------
Andreas Isenegger
SLS Controls
Paul Scherrer Institute, WSLA/207
CH-5232 Villigen PSI, Switzerland
Tel/Fax: ++41 (56) 310 5125/4413
----------------------------------

----- Original Message -----
From: "Ernest L. Williams Jr." <[email protected]>

To: "Andreas Isenegger" <[email protected]>

Cc: "EPICS tech-talk" <[email protected]>

Sent: Thursday, September 29, 2005 12:55 AM

Subject: Re: VxWorks5.5 on MVME230x doesn't reestablish broken N/W link
On Wed, 2005-09-28 at 17:12 +0200, Andreas Isenegger wrote:

Hello All,
Since I posted 2 e-mails in July 2005 seeking for help, I thought it
would

be worth informing you that the problem could be solved.
There are indeed some bugs in Wind River's N/W driver (dec21x40End.h/c)
that

cause the misbehavior of the IOC. I got new revisions of these files from

Martin Niedermair, WRS Austria, which solve the issue. Strangely enough,
a

different support engineer of WRS sent me another revision of the same

files, which doesn't fix the bug. About 2 years back, they were
branched...

. The latest official patch from WRS contains again different revisions
of

the files.
Thus, if you are experiencing troubles with IOC's disconnecting for 'no

reasons', or reset ports of N/W switches to regain links to IOC's, build
a

new VxWorks image with the latest revision of these files.

Can you send us a copy of the files or give us a rev. number to compare
against?
I think we are fine but I want to double-check.
Our MVME2101s also use the "dec21x40End.h/c"

PS: Did WRS give you a TSR or SPR that we could look up on our end?

Thanks,
Ernest L. Williams Jr.
SNS Control Systems Group
ORNL

Kind regards,
Andreas Isenegger
----------------------------------

Andreas Isenegger

SLS Controls

Paul Scherrer Institute, WSLA/207

CH-5232 Villigen PSI, Switzerland

Tel/Fax: ++41 (56) 310 5125/4413

---------------------------------- 

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System