Experimental Physics and
Industrial Control System

[email protected] (Rozelle Wright) · Mon, 28 Jul 97 13:45:43 MDT

> From [email protected] Mon Jul 28 09:18:51 1997
> Return-Path: <[email protected]>
> Subject: Sequencer bug
> To: [email protected] (EPICS tech-talk)
> Date: Mon, 28 Jul 1997 11:17:29 -0400 (EDT)
> X-Mailer: ELM [version 2.4 PL25]
> Mime-Version: 1.0
> Content-Type> : > text/plain> ; > charset=US-ASCII> 
> Content-Transfer-Encoding: 7bit
> Content-Length: 717
> X-Lines: 14
> 
> 
> Sequencer version 1.9 always returns FALSE from the function pvGetComplete.
> 
> Changing line 133 in seq_if.c from pDB=pSP->pChan to pDB=pSP->pChan+pvId
> makes pvGetComplete operate correctly.
> 
> -- 
>  ----------------------------------------------------------------------    
>   William E. Spengler                  [email protected]   
>   COMSO Inc.                           Mail Stop 931                       
>   NASA Langley Research Center         Phone:   (757) 766-8255           
>   Hampton, Virginia 23681-0001         Fax:     (757) 766-2571           
>   Subcontracted by CSC; Assigned to Information Sys and Serv Division      
>  ----------------------------------------------------------------------     
> 
> In synchronous mode the pvGet function always times out (error 10).
> This was traced back to seq_main.c not calling semBCreate.
>    
> Changing line 317 in seq_main.c to  if(!(pSP->options & OPT_ASYNC))
> corrects the problem.
> 

Hi William,
I think you may have an old version of the sequencer,
your line numbers don't match the ones that are
put out in the sequencer version 1.9.2 which is
then one distributed with 3.13 and 3.12.2 patches.

As far as I can tell the two bugs you mentioned are fixed in
these version.
seq_if.c has the following code somewhat after line 133:
/*
 * seq_pvGetComplete() - returns TRUE if the last get completed.
 */
long seq_pvGetComplete(SS_ID ssId, long pvId)
{
	SPROG		*pSP;	/* ptr to state program */
	CHAN		*pDB;	/* ptr to channel struct */

	pSP = ((SSCB *)ssId)->sprog;
	pDB = pSP->pChan + pvId;
	return pDB->getComplete;
}

and seq_main.c has the following code some place after line 317 :
		/* Create a binary semaphore for synchronous pvGet() (-a) */
		if ((pSP->options & OPT_ASYNC) == 0)
		{
			pSS->getSemId =
			 semBCreate(SEM_Q_FIFO, SEM_FULL);
			if (pSS->getSemId == NULL)
			{
				logMsg("can't create getSemId\n", 0,0,0,0,0,0);
				return;
			}

I think William Lupton is taking over maintenance on the sequencer,
but the message I sent him indicates he is on vacation, so I glanced
at the problem with the above results.

Good Luck,
Rozelle

---------------------------------------------------------
|                                                       |
|Rozelle Wright              Phone (505) 667-4804       |
|Los Alamos Natl Labs LANSCE-8  FAX (505) 665-5107      |
|PO Box 1663 MS-H820         Group Office (505) 667-6087|
|Los Alamos, NM 87545        email : [email protected]   |
|                                                       |
---------------------------------------------------------

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System