Experimental Physics and
Industrial Control System

"Pearson, Matthew R." <[email protected]> · Thu, 18 Jul 2013 15:03:59 -0400

Hi,

While we're talking about the PC6K driver… has anyone started work to port it to one based on asynPortDriver? (ie. a motor model 3 driver)? We're heavily invested in the Parker controller here at the SNS and I was planning to update the driver to an asynPortDriver version.

Cheers,
Matt

On Jul 18, 2013, at 2:41 PM, Mark Rivers <[email protected]> wrote:

> Hi Martin,
> 
> I don't know about the intended behavior, since I did not write the driver.  I am just saying what the actual behavior is. (Note that there is a comment in the driver code that says that I am the original author, but that is not true.  It is a remnant of someone's cut and paste from my Newport MM4000 driver on which this driver was based).
> 
> I tend to agree with you that this seems like something that should be done before iocInit.
> 
> PC6K_num_cards is set correctly after the PC6KConfig commands, and the required motor_state structures have been allocated.
> 
> However, the motor_init() code is where the connection to the asyn driver occurs, and that does not happen until iocInit.
> 
> If the code that initializes the connection to the hardware, like:
> 
> 	/* Initialize communications channel */
> 	success_rtn = pasynOctetSyncIO->connect(cntrl->asyn_port, 
>                                            cntrl->asyn_address, &cntrl->pasynUser, NULL);
> 
> were moved from motor_init() into PC6KConfig() then you would be able to call PC6kUpLoad before iocInit().
> 
> Mark
> 
> -----Original Message-----
> From: Konrad, Martin [mailto:[email protected]] 
> Sent: Thursday, July 18, 2013 1:13 PM
> To: Mark Rivers
> Cc: [email protected]
> Subject: Re: motor record & PC6K: program upload fails
> 
> Hi Mark,
>> So your PC6KUpLoad command must not be given before iocInit, it must be after iocInit.
> But would mean we reprogram the controller while the IOC is running and
> record processing is in full progress, doesn't it? That sounds quite
> dangerous to me. I was expecting it to program the controller before
> record processing starts. Can you please confirm that running it after
> iocInit really is the intended behavior?
> 
> Martin
> 

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System