---

GLAST/LAT > DAQ and FSW > FSW > Doxygen Index > LSEP / V1-2-0

Constituent: lsepw     Tag: mv2304

---

         

---

LSEPW.c File Reference

Manages information and adds the application specific root contribution to the output datagram. More...

#include "LSEPW_pvtdef.h"
#include "LSEP/LSEPW.h"
#include "LSEP/LSEP_ids.h"
#include "LSEP/LSEP_etxCtb.h"
#include "LSEP/LSEPW_evt0.h"
#include "LSEP/LSEP_evtBdy0.h"
#include "LSE/LSEW_dgmFw.h"
#include "LDT/BFP.h"
#include "EDS/EDS_fw.h"
#include "EDS/EDS_rsd.h"
#include "EDS/EBF_gemLocate.h"
#include "EDS/FFS.h"
#include <string.h>

Include dependency graph for LSEPW.c:

Defines
#define	ADVANCE(_ptr, _nbytes)   (void *)(((unsigned char *)(_ptr) + (_nbytes)))
	Advance a generic pointer by the specified number of bytes.
Functions
static __inline unsigned int	pack_rsdDscs (unsigned int *wrds, unsigned int position, unsigned int active, const EDS_rsdDsc dscs[32])
	Packs the byte length of the result summary data specified by active into the output bit array.
static __inline unsigned int	pack_rsds (unsigned int *wrds, unsigned int position, unsigned int active, unsigned int rsd, const EDS_rsdDsc dscs[32])
	Packs the result summary information for those handlers that are both active and have non-zero result summary data.
static __inline unsigned int	pack_cfgs (unsigned int *wrds, unsigned int position, unsigned int active, const unsigned char cfgs[32])
	Packs the cfgs ids specified by active into the output bit array.
int	LSEPW_sizeof (int level)
	Return the size of the LSEP data structure.
const EDS_fwPostServices *	LSEPW_constructC (LSEPW *lsepw, int level, const void *cfg, unsigned short int apid, unsigned int dgm_id, LSF_CPU_K cpu, LSF_ORIGIN_K origin, unsigned int sw_key, LSEW_dgmFwPoster postRtn, void *postPrm, unsigned int *buf, int buf_size)
	Constructs/initializes the physics application specific context.
const EDS_fwPostServices *	LSEPW_construct (LSEPW *lsepw, int level, unsigned short int apid, unsigned int dgm_id, LSF_CPU_K cpu, LSF_ORIGIN_K origin, unsigned int sw_key, LSEW_dgmFwPoster postRtn, void *postPrm, unsigned int *buf, int buf_size)
	Constructs/initializes the physics application specific context.
int	LSEPW_start (LSEPW *lsepw, LSF_REASON_OPEN_K reason, unsigned int run_id, unsigned int startTime, LSF_MODE_K mode, unsigned int hw_key, unsigned int lpa_key)
	Start or more accurately, prepare a posting stream to receive events.
int	LSEPW_flush (LSEPW *lsepw, int reason)
	Flushes the current datagram.
int	LSEPW_levelSet (LSEPW *lsepw, int level)
	Modifies to the current working level to the minimum of level and the maximum allowable compression level that LSEPW is configured to.
int	LSEPW_notify (LSEPW *lsepw, unsigned int mode, unsigned int active, const unsigned char cfgs[32], const EDS_rsdTbl *rsdTbl)
	Called when the configuration has changed, for example by a mode change.
unsigned int *	LSEPW_ctbAdd (LSEPW *lsepw, unsigned int *cur, int len32)
	Adds the Physics specific context contribution to the datagram.
int	LSEPW_evtAdd (LSEPW *lsepw, EDS_fwIxb *ixb)
	Formats the event specified in by the EDS_fw Information Exchange block, flushing the current record to the output stream if the datagram is full.

---

Detailed Description

Manages information and adds the application specific root contribution to the output datagram.

Author:

JJRussell - russell@slac.stanford.edu

   CVS $Id: LSEPW.c,v 1.10 2007/02/14 06:19:23 russell Exp $

Overview

This gathers all the stuff need to add both the public and private application specific contributions to the datagram. This structure should be allocated and pasted to LSEPW_construct routine.

Private

The private aspects are driven by the needs of event formatting and deformatting. These needs break into two pieces,

1. Those that are static
2. Those that dependent of the event data

The private static needs are things such as the CAL pedestals used in the encoding. The dynamic needs are the compression tables that are built on the fly from the data.

Public

The public aspects also fall into two pieces

1. Those that are static
2. Those that change in response to changes in the operating conditions

The public static needs are things such as the file keys for the relevant configuration files, LATc, LPA_DB, etc. The dynamics needs are the list of active handlers and their active configuration.

The challenge is how to gather this information and, particularly, for those values that change when the conditions change, to decide how to update the information. Here one has to choices

1. Query EDS_fw everytime the information is needed
2. Have EDS_fw notify one of such at change.

---

Define Documentation

#define ADVANCE (  _ptr, _nbytes   )     (void *)(((unsigned char *)(_ptr) + (_nbytes)))

Advance a generic pointer by the specified number of bytes. Returns: The advanced pointer, recast to a void Parameters: _ptr  The pointer to advance _nbytes  The nubmer of bytes to advance

---

Function Documentation

const EDS_fwPostServices* LSEPW_construct (  LSEPW *  lsepw, int  level, unsigned short int  apid, unsigned int  dgm_id, LSF_CPU_K  cpu, LSF_ORIGIN_K  origin, unsigned int  sw_key, LSEW_dgmFwPoster  postRtn, void *  postPrm, unsigned int *  buf, int  buf_size )

Constructs/initializes the physics application specific context. Parameters: lsepw  The physics application specific context to construct level  The maximum compression level that will be used apid  The application id to use in the CCSDS packets dgm_id  The LAT Science Format Data Id (LSF) to use cpu  The originating CPU origin  The data source type sw_key  The master software configuration key; this is the key to the Secondary Boot File. postRtn  Routine used to post a formatted event postPrm  Parameter passed to the post routine buf  The first, preallocated buffer to fill buf_size  The size, in bytes, of the ouptut buffer. Events will be packed into this buffer until full

const EDS_fwPostServices* LSEPW_constructC (  LSEPW *  lsepw, int  level, const void *  cfg, unsigned short int  apid, unsigned int  dgm_id, LSF_CPU_K  cpu, LSF_ORIGIN_K  origin, unsigned int  sw_key, LSEW_dgmFwPoster  postRtn, void *  postPrm, unsigned int *  buf, int  buf_size )

Constructs/initializes the physics application specific context. Parameters: lsepw  The physics application specific context to construct level  The maximum compression level that will be used cfg  Level level specific configuration parameters apid  The application id to use in the CCSDS packets dgm_id  The LAT Science Format Data Id (LSF) to use cpu  The originating CPU origin  The data source type sw_key  The master software configuration key; this is the key to the Secondary Boot File. postRtn  Routine used to post a formatted event postPrm  Parameter passed to the post routine buf  The first, preallocated buffer to fill buf_size  The size, in bytes, of the ouptut buffer. Events will be packed into this buffer until full This routine is something of a royal kludge. The original call for LSEPW_construct did not include a configuration parameter. To keep things backwardly compatiable, this routine was invented. It should be become LSEPW_construct if a reorganization permits it.

unsigned int* LSEPW_ctbAdd (  LSEPW *  lsepw, unsigned int *  cur, int  len32 )

Adds the Physics specific context contribution to the datagram. Parameters: lsepw  The physics application specific context cur  Pointer to the where to add the informaton len32  The length, in 32-bit words, of the available space in the datagram

int LSEPW_evtAdd (  LSEPW *  lsepw, EDS_fwIxb *  ixb )

Formats the event specified in by the EDS_fw Information Exchange block, flushing the current record to the output stream if the datagram is full. Returns: Status Parameters: lsepw  The physics application specific context ixb  The Information Exchange Block

int LSEPW_flush (  LSEPW *  lsepw, int  reason )

Flushes the current datagram. Parameters: lsepw  The physics application specific context reason  The reason for the flush

int LSEPW_levelSet (  LSEPW *  lsepw, int  level )

Modifies to the current working level to the minimum of level and the maximum allowable compression level that LSEPW is configured to. Returns: The actual level is was set to, i.e. the minimum of level and the maximum allowable compression level. On failure this value is -1. Parameters: lsepw  The physics application specific context level  The new compression level

int LSEPW_notify (  LSEPW *  lsepw, unsigned int  mode, unsigned int  active, const unsigned char  cfgs[32], const EDS_rsdTbl *  rsdTbl )

Called when the configuration has changed, for example by a mode change. Returns: 0 if successful, non-zero if not Parameters: lsepw  The physics application specific context mode  The new mode active  The list of active handlers cfgs  The current configuration for each handler. rsdTbl  The result summary data table

int LSEPW_sizeof (  int  level  )

Return the size of the LSEP data structure. Parameters: level  The maximum compression level that will be used. Resources will be allocated consistent with supporting events written in this and all lesser compression levels.

int LSEPW_start (  LSEPW *  lsepw, LSF_REASON_OPEN_K  reason, unsigned int  run_id, unsigned int  startTime, LSF_MODE_K  mode, unsigned int  hw_key, unsigned int  lpa_key )

Start or more accurately, prepare a posting stream to receive events. Returns: Status, currectly what LSEW_dgmFwRunStart returns. Parameters: lsepw  The physics application specific context reason  The reason why this run is being started run_id  The run identifier. Only the first 24-bits are used. startTime  The run start time rounded to the nearest second mode  The operational mode hw_key  The master hardware configuraion key; this is the key to the LATc file lpa_key  The LPA_DB file key

static __inline unsigned int pack_cfgs (  unsigned int *  wrds, unsigned int  position, unsigned int  active, const unsigned char  cfgs[32] )  [static]

Packs the cfgs ids specified by active into the output bit array. Returns: The new bit position in the output buffer Parameters: wrds  The beginning of 32-bit output buffer position  The current bit position in the output buffer active  Bit mask indicating which of the elements of cfgs should be packed cfgs  The array of up-to 32 configuration ids of the handlers Warning: This piece of code assumes that active is non-zero

static __inline unsigned int pack_rsdDscs (  unsigned int *  wrds, unsigned int  position, unsigned int  active, const EDS_rsdDsc  dscs[32] )  [static]

Packs the byte length of the result summary data specified by active into the output bit array. Returns: The new bit position in the output buffer Parameters: wrds  The beginning of 32-bit output buffer position  The current bit position in the output buffer active  Bit mask indicating which of the elements of rsdDscs should be packed dscs  The result summary data descriptor array

static __inline unsigned int pack_rsds (  unsigned int *  wrds, unsigned int  position, unsigned int  active, unsigned int  rsd, const EDS_rsdDsc  dscs[32] )  [static]

Packs the result summary information for those handlers that are both active and have non-zero result summary data. Returns: The new bit position in the output buffer Parameters: wrds  The beginning of 32-bit output buffer position  The current bit position in the output buffer active  The defining set of bits, i,e. set of all handlers rsd  The set of handlers with non-zero result summary data. This bit list is collapsed by removing all the corresponding bits in active that are zero. Since only active handlers can contribute result summary data, the set is reduced to only the active handlers. Thus for each set bit in active a result bit is emitted that is 0/1 if that handler does/does not produce any summary data. dscs  The result summary data descriptor array This routine composes a bit field from fld2 considering only the set bits in fld1. For every bit set in fld1, a bit is added to the output which is set to 0/1 if the corresponding bit in fld2 is not/is set. The resulting bits are packed into the output buffer.

---