RE: [Q or Help!] change of scanning period causes IOC crash.

[Noboru Yamamoto <yamamoto@acsad1.kek.jp>]

Hi,
 
 Thank you Jeff. After I sent the previous e-mail, my colleague found that
reduced size of buffer area removed the problem. In the original program,
he used rather big array (20Kbytes) in the subroutine for reading data from VXI.
It was necessary at somepoint, but the current version can work with 
smaller buffer size.  And this version works very well.

Thanks again to all of you.

 Noboru YAMAMOTO
 KEKB control group
 KEK, JAPAN

=============================================================================
Jeff wrote:
>
>Noburo,
>
>As I recall, APS experienced some problems with the National Instruments
>MXI hardware timing out during VXI bus transfers when they first set up their 
>MXI systems. Ned Arnold (nda@aps.anl.gov) might be able to provide some hints on
>setting up MXI bus systems.
>
>Jeff

>On Tuesday, June 02, 1998 2:08 AM, Noboru Yamamoto [SMTP:yamamoto@acsad1.kek.jp] wrote:
>> Hi,
>>  
>>  We are writing a device support routines for a VXI oscillo scope. We use
>> MXI-bus connection to cnnect VME bus, where IOC is, and VXI bus , where
>> an Osciloscope. When you satrted databse at one scanning period, the device
>> support works OK. Some records send commands to setup measurement and a
>> waveform record reads trace data from the osciloscope. 
>> However, we changes scanning period of the processing, IOC crashes.
>> Actually it freeze.
>>  
>>  Does anybody has a similar problem in developing device support? Which part
>> of device support we should look at?
>> 
>>  Your comments will be grately appricated.
>> 
>