asynDriver: Release Notes

devEpics

• Change logic for output records with asyn:READBACK so they set udf=0 in the outputcallbackcallback() function, and not in the processXX function. This is needed to correctly clear the UDF status and severity.
  • This problem is happening because for EPICS output records the only way to clear a UDF alarm is to put a value into the record, either from an external put from a CA or DB link (in which case code in the dbAccess.c dbPut() routine clears UDF and the record would get processed) or by reading a value through the DOL link during record processing triggered by some other means.
  • In both cases UDF will have been cleared before the call to checkAlarms() in the boRecord::process() routine, which is where the UDF field is checked.
  • Unfortunately for asyn all output records call the checkAlarms() routine before the device support's write_bo() routine, so clearing UDF in processBo() is too late for this process cycle, the UDF_ALARM has already been registered.
  • This also means that if the bo record is configured with non-zero alarms in ZSV or OSV the initial read wouldn't trigger the appropriate STATE_ALARM either, since they get set in that routine as well.

drvAsynUSBTMC

• Add support for output terminator.
• Increase buffer size to 1 MB.
• Add Remote Enable (REN), Local Lockout (LLO), and Go To Local (GTL) support.

asynPortDriver

• Fixed initialization problem for UInt32 parameters. Initially setting a parameter to 0 left the corresponding PV undefined because the callback wasn't called. Thanks to Erico Nogueira for this fix.
• Removed "using std::logic_error" from parameter header files. This is bad practice. This change could potentially break downstream module that would now have to use std::logic_error, rather than just logic_error. This is unlikely since the param classes are essential private to asynPortDriver. Thanks to Luka Krmpotic for this fix.

drvAsynIPServerPort

• Allow specifying an interface to bind the IP server port. The second argument to drvAsynIPServerPortConfigure allows specifying a hostname or IP address for multi-homed machines. If you want to specify any interface, use "localhost:port", an empty host ":port" or the address "0.0.0.0:port". The address has to be an existing interface on the host. If the name cannot be resolved, the call will fail. If you need the loopback interface, use "127.0.0.1:port". Thanks to Lutz Rossa for this.

asynShellCommands

• Provide usage information for asynSetTraceMask and asynSetTraceIOMask. This is helpful for remembering how to specify the bit masks numerically or symbolically. Thanks to Michael Davidsaver for this.

asynManager

• Added asynDisconnected and asynDisabled in strStatus(). Thanks to Kristian Loki for this.

configure/RELEASE

• All external modules are commented out. Users will need to uncomment the ones to be used.

drvVxi11

• Fix potential VXI connection resource leak. Thanks to Freddie Akeroyd for this.

.ci

• Change from Travis to Github Actions. Thanks to Michael Davidsaver for this.
• Add support from sscan, calc, and ipac. Thanks to Freddie Akeroyd for this.

devGPIB and streamSCPI template files

• Changed DB_INSTALLS+= to DB+= to work with recent versions of EPICS.

drvProligixGPIB

• New driver for the Proligix GPIB-Ethernet controller. Thanks to Eric Norum for this. This $200 device makes GPIB devices available over Ethernet. It uses a proprietary protocol, not VXI-11.

asynPortDriver

• Added checks for invalid list (<0 or >=maxAddr).
• Added new error code asynParamInvalidList that is returned if the list passed is not valid.

devCommonGPIB.c

• Added 0 termination after strncpy().

asynInterposeEos

• Improve efficiency when eosOutLen is 0.

asynRecord

• Minor changes to avoid compiler warnings.

Many files

• Changed the code and the Makefile to avoid using shareLib.h from EPICS base. Added a new header file asynAPI.h that is used to control whether functions, classes, and variables are definined internally to the asyn library or externally. This is the mechanism now used in EPICS base 7. It makes it much easier to avoid mistakes in the order of include files that cause external functions to be accidentally exported in the asyn DLL or shareable library. This should work on all versions of base, and have no impact on user code.

devAsynInt32.c, devAsynUInt32Digital.c

• Replaced strncpy()) with memcpy() to avoid compiler warnings on newer versions of gcc (e.g. 8.3.1).

drvAsynIPPort

• Fixed the logic for using SO_REUSEPORT or SO_RESUSEADDR. In R4-40 RTEMS and Windows used SO_REUSEADDR and all others used SO_REUSEPORT. This does not work because Linux kernels before 3.9 also do not support SO_REUSEPORT. Now if SO_REUSEPORT is not defined then SO_REUSEADDR is used instead.

         asynSetTraceMask port,0,ASYN_TRACE_ERROR 
        asynSetTraceIOMask port,0,ascii+escape 
        asynSetTraceInfoMask port,0,1+port+TRACEINFO_SOURCE|ASYN_TRACEINFO_THREAD

devAsynFloat64.c, devAsynInt32.c

• Added support for SCAN=I/O Intr with asynFloat64Average and asynInt32Average device support. Previously only periodic scanning or manual processing was supported. The I/O Intr support computes the average and processes the record once NumAverage callback readings have been received. The SVAL field in the ai record is used to set NumAverage. This is rather a kluge, but there is not another good way to communicate this value to the device support, while allowing it to be changed at run-time.

drvAsynIPPort.c

• Improved error reporting. Thanks to Dirk Zimoch for this.

asynInterposeEos.c

• Improved asynTrace output to print bytes read and eom, and print warning if low-level driver returns error.

asynInterposeCom.c

• Added support for the asynOption ixon (XON/XOFF) for ports communicating via the RFC 2217 Telnet protocol. In this case, as noted in the standard, ixon implies both outbound and inbound flow control.

asynPortDriver

• Fixed uninitialized value in paramVal.

asynShellCommands.h

• Add asynOctetDisconnect and asynWaitConnect.

drvVxi11.c

• Fixed logic for FLAG_NO_SRQ. Error was introduced in R4-31.

CONFIG_SITE

• Optionally include $(SUPPORT)/configure/CONFIG_SITE.
• Older Linux and Cygwin distributions have packages (e.g. glibc-headers) that put the rpc include files in /usr/include/rpc. Newer Linux and Cygwin distributions use the tirpc package instead. For such systems define TIRPC=YES.

.travis.yml

• Fixed to fetch Debian files from nsls2 with https not http.

devAsynFloat64.c

• Added support for ASLO/AOFF scaling and SMOO value smoothing to the float64 device support for ai and ao records. SMOO only applies to ai records. devAsynFloat64 directly writes to the record's VAL/OVAL fields, so the conversion routines in the record that use RVAL/VAL don't work.

devAsynInt32.c, devAsynUInt32Digital.c, devAsynFloat64.c, devAsynOctet.c

• Fixed a problem with output records that have the asyn:READBACK info tag, i.e. output records that update on callbacks from the driver. Previously it did not correctly distinguish between record processing due to a driver callback (in which case it should not call the driver) and normal record processing (in which case the driver should be called). A new test application, testOutputCallbackApp, was added to test this. It allows testing all combinations of the following 6 settings for all 4 of these device support files:
  1. Synchronous driver, i.e. ASYN_CANBLOCK not set.
  2. Asynchronous driver, i.e. ASYN_CANBLOCK is set.
  3. Driver callback is done in the write() operation.
  4. Driver callback is done in a separate thread. The callbacks are triggered with the TriggerCallbacks record.
  5. Single callback is done in each operation.
  6. Multiple callbacks are done in each operation. The NumCallbacks record selects the number of callbacks.
  The callback values are the following. For the longout records (asynInt32 and asynUInt32Digital) and the ao record (asynFloat64) the callback value is N+1 where N the current record value. Thus if one writes 10 to the longout record and NumCallbacks is 5 it will immediately do callbacks with the values 11, 12, 13, 14, 15. For the bo record the value toggles between 0 and 1 for each callback. For the stringout record the value is a string with the format "Value=N". N increments by 1 on each callback as for the longout records. For each output record there is a corresponding input record with SCAN=I/O Intr. If the test is working correctly the output records should always have the same values as the input records.

devAsynOctet.c

• Fixed bug where the record alarm status was not set correctly if a write or read failed with a synchronous driver.

asynManager.c

• Improved error messages from autoconnect attempts. Messages are now printed only when the autoconnect status changes, rather than each time there is a failure. Thanks to Ben Franksen for this.

drvAsynIPServerPort.c

• Changed the logic so it creates maxClients drvAsynIPPort drivers at initialization rather than on-demand when clients connect. This makes it possible to use these ports before iocInit, which is much more useful. They can be used in input and output links in EPICS records, for example. This change should be backwards compatible, since it was transparent when the ports were created.
• The first drvAsynIPPort created is now named PORT_NAME:0 rather than PORT_NAME:1, where PORT_NAME is the name of the drvAsynIPServer port created by this driver. This might break backwards compatibility, but clients were normally getting the name from a callback, so probably not.
• Set the socket option SO_REUSEADDR when creating the listening socket. Previously if a client was connected and the IOC exited it could not be restarted again immediately. One needed to wait for the operating system to time out the connection. This change allows the IOC to be run again with the same listening port immediately, without waiting for the operating system timeout.
• Added an exit handler to close the listen socket and delete the private structure.
• Improved the report() function so it shows all the drvAsynIPPort drivers that were created.

testIPServerApp

• Added a new datbase testIPServer1.db for testing the enhancements to drvAsynIPServerPort. It contains a stringin record and a stringout record that use the drvAsynIPPort created by drvAsynIPServerPort.
• Added a new startup script, iocBoot/ioctestIPServer/st.cmd.testIPServer1 which creates a drvAsynIPServerPort called P5001 on localhost:5001. It loads the testIPServer1.db database connected to the drvAsynIPPort created by P5001, called P5001:0. One can telnet to port 5001 on the IOC computer. All text typed in the telnet session will go into the stringin record. Text written to the stringout record will appear in the telnet session.

drvVXI11.c

• Improved error messages from connection attempts. Messages are now printed only when the connection status changes, rather than each time there is a connection failure. Thanks to Ben Franksen for this.

testErrorsApp

• Previously the user-defined alarm status and severity were only set on callbacks, i.e. input records with SCAN=I/O Intr and output records with info tag asyn:READBACK=1. Now all calls to the read() and write() methods in each interface also return the user-defined alarm status and severity in pasynUser.

asynPortClient

• Fixed typo in constructor for asynUInt32DigitalClient, it was calling asynInt32SyncIO->connect rather than asynUInt32DigitalClient->connect.

OPI screens

• Added new opi/Makefile. This automatically converts all opi/medm/*.adl files into corresponding files in edm/autoconvert, caqtdm/autoconvert, and boy/autoconvert. It uses new RULES_OPI and CONFIG_SITE.linux-x86_64 files in synApps/support/configure directory. The upper-level edm, caqtdm, and boy directories should be used only for manually tweaked files.
• The medm adl files were improved to make the autoconverted files work better. The size of text graphics and text update widgets were set to the actual text size, and text update widgets were set to the correct datatype, e.g. string type for enum PVs.

Documentation

• asynDriver.html has been improved. All level 3 headings are now links in the Contents at the start of the file. This makes it much easier to find a particular topic and jump to it quickly. Many of the test applications are now documented, including medm screen shots to show how they work.

asynManager.c

• Fixes to queueLockPort() which is used by the asynXXXSyncIO interfaces.
  • Previously queueLockPort() did not specify a timeout value or timeout callback function in the call to pasynManager->queueRequest().
  • This meant that if a port disconnected while a lock request was queued it would hang until the port reconnected. For example, calls to the asynXXXSyncIO functions would not return until the port reconnected.
  • Added a timeout callback for the queueRequest in queueLockPort(). This callback function prints a warning message with ASYN_TRACE_WARNING. It sets the psynUser->auxStatus field to asynTimeout so that queueLockPort() detects an error and returns an error status without locking the port. Callers must be sure to check the return status of queueLockPort and not call any port driver functions if it does. This was already done in all of the asynXXXSyncIO functions.
  • The timeout for the queueLockPort queue request is determined as follows
    • Each port now has a queueLockPortTimeout value. This is set to a default of 2.0 seconds when the port is created.
    • This can be changed with a new iocsh command

      asynSetQueueLockPortTimeout(portName, timeout)

    • If the pasynUser->timeout that is passed for a particular I/O operation is larger than the port timeout value this larger value is used instead.
• In traceVprintIOSource if the traceIOMask is 0 or traceTruncateSize ≤ 0 output a newline, otherwise output is garbled

asynPortDriver

• Replaced "int" with "epicsInt32" in many places for consistency and to avoid compiler warnings on some architectures. Thanks to Scott Baily for this.
• Added new asynPortDriver::getParamType() methods. Thanks to Mark Davis for this.
• Added a new asynPortDriver constructor that removes the paramTableSize argument. Marked the previous constructor as EPICS_DEPRECATED. Added a new initialize() method that both constructors call to do the actual work. Changed all of the asynPortDriver example programs and unit test programs to use the new constructor. Thanks to Martin Konrad for this.

Test application *Main.cpp files

• The main programs in all of the test applications have been updated to the version in the template files in EPICS base 3.15.5. This includes a call to epicsExit() after the iocsh() returns. This is needed for epicsAtExit to work correctly on some platforms, including Windows.

Files in configure/ directories

• The Makefile, CONFIG, CONFIG_SITE, and RULES* files in configure/, makeSupport/top/configure/, and documentation/HowToDoSerial/AB300/configure/ have been updated to the versions in the template files in EPICS base 3.15.5. The new versions are better suited to site-specific customization.

VXI-11

• Added option to omit SRQ channel.

Build system

• Minor change to build with mingw.

Unit tests

• Added unit tests in asynPortDriver/unittest.

asynPortDriver

• Changed the parameter list to be an std::vector. This means that it is no longer necessary to pass the number of parameters in the constructor because the list grows with each call to createParam().
• Changed the internal storage of string parameters from char * to std::string. Added new setStringParam() and getStringParam() methods that take an std::string& argument rather than char *. These do not require specifying a maximum string length, unlike the char * methods.
• Changed the constructor so that it throws an std::runtime_error exception in the case of any error, rather than simply calling "return" and creating a non-functional asynPortDriver object, which it did previously. It still prints an error message as it has done previously.

drvAsynIPPort

• Fixed bugs in Unix sockets.

asynDriver

• Added new fields to the asynUser structure. These are int alarmStatus and int alarmSeverity. They are intended to pass alarm information back to device support, indepenent of the status return from the write() or read() methods or the auxStatus field used for callbacks.

devEpics

• Previously when the status return or auxStatus were not asynSuccess then device support did not update the value, and set the record STAT and SEVR based upon rules in the generic device support. If the new alarmStatus and alarmSeverity fields are 0 (default) then the current behavior is retained. If these fields are non-zero, then the record STAT and SEVR will be set to these values, independent of the status return or the auxStatus values. This allows a driver to set the record value, while still setting the STAT and SEVR to values other than NO_ALARM. The behavior of the function asynStatusToEpicsAlarm() was changed, and this function is now always called at the end of the process() function, even if the status or auxStatus is asynSuccess.

asynPortDriver

• Added new methods setParamAlarmStatus(), setParamAlarmSeverity, getParamAlarmStatus(), getParamAlarmSeverity to write and read the new pasynUser->alarmStatus and pasynUser->alarmSeverity fields.
• Changed the constructor so that it sets the ASYN_MULTIDEVICE flag if maxAddr>1 even if the caller neglected to set it.
• Fixed destructor eliminate memory leaks. Thanks to Henrique Almeida.

devAsynOctet

• Added support for initial readback from driver for stringout and waveform output records to support bumpless reboots. This support is only enabled if the info tag asyn:INITIAL_READBACK is present for these records in the database. The testErrorsApp application was updated to test this feature.
• Check that FTVL is CHAR or UCHAR and return an error in init_record if it is not.
• Changed the waveform record device support asynOctetWrite to only support string output. The numchars argument in pasynOctet->write() is obtained by calling strnlen(). This makes it consistent with the behavior of the stringout record.
• Added a new waveform record device support asynOctetWriteBinary. This passes numchars=NORD and does not call strnlen(), so it should be used for binary data.

drvAsynSerialPort

• Added support for RS-485 on Linux. This is only supported on ports with hardware RS-485 capability, and only on more recent Linux kernels. Thanks to Florian Feldbauer. The following new options were added for the asynOption interface in this driver.
  • rs485_enable Y/N
  • rs485_rts_on_send Y/N
  • rs485_rts_after_send Y/N
  • rs485_delay_rts_before_send msec_delay
  • rs485_delay_rts_after_send msec_delay

testErrorsApp

• Added new records AlarmStatus and AlarmSeverity that allow this driver to test the new alarm handling behavior.

drvAsynIPPort

• Revert the change made in R4-27 to drvAsynIPPortConfigure that changed the noProcessEos flag to userFlags. The final argument to drvAsynIPPortConfigure is now noProcessEos again.
• Implemented the asynOption interface which allows configuring options using key/value pairs. This interface is used in the drvAsynSerialPort driver for configuring baud rate, stop bits, etc. It is also used on drvAsynIPPort for configuring these same items if the port was configured with the COM qualifier, used for Ethernet/Serial terminal servers that use the TELNET RFC 2217 protocol. For that drvAsynIPPort uses the asynInterposeCOM to interpose the asynOctet and asynOption interfaces.
• drvAsynIPPort now also directly supports the asynOption interface for 2 key/value pairs.
  • key="disconnectOnReadTimeout", value="Y" or "N". This option replaces the USERFLAG_CLOSE_ON_READ_TIMEOUT that was introduced in R4-27. The advantage of using the asynOption interface is that this behavior can now be changed at run-time, rather than being set once when the driver is created.
  • key="hostInfo", value="host:port[:localport] [protocol]", i.e. the Internet host name, port number, optional local port number and optional protocol. This uses the same syntax as the drvAsynIPPortConfigure command. This allows changing at run time the Internet host and port to which this asyn port is connected. The only restriction is that the setting of the COM (TELNET RFC 2217) protocol cannot be changed from that specified with drvAsynIPPortConfigure. This is because if COM is specified in the drvAsynIPPortConfigure command then asynOctet and asynOption interpose interfaces are used, and asynManager does not support removing interpose interfaces.
• Improved debugging output with ASYN_TRACEIO_DRIVER. Previously there was no asynTrace output if recv() or recvfrom() returned ≤0. It is often useful to know if these functions returned 0, so the asynTrace output is now only suppressed if these functions return <0.
  There was previously no ASYN_TRACEIO_DRIVER output from the flushIt() function, which repeatedly calls recv() until the return value is ≤0. ASYN_TRACEIO_DRIVER will now print a message with the total number of bytes flushed in this function, if the number is greater than 0.

asynRecord

• Added 2 new record fields to read and control the new drvAsynIPPort options:
  • DRTO (Disconnect on Read Timeout)
  • HOSTINFO (hostInfo string)
• Added new OPI screen (asynIPPortSetup.[adl,.opi,.edl,.ui]). This screen allows display and control of the new DRTO and HOSTINFO fields. It also allows display and control of the serial port parameters for drvAsynIPPorts created with the COM (RFC 2217) protocol. There was previously no GUI to control these options for IP ports. This screen is accessed via the More menu in asynRecord.adl.

asynPortDriver

• Add paramName to error messages for easier debugging.
• Throw exception if NULL pointer is passed to setStringParam().

asynDriver

• asynTrace now does flush after each write on files, previously it only did this on stdin and stderr. This prevents trace information in files being lost if the application crashes.
• Fixed syntax of call to cantProceed().
• Added definition of ASYN_EXCEPTION_STRINGS which provides strings corresponding to each asynException enum value.

testBroadcastApp

• Added new test application testBroadcastBurst. This application periodically sends a burst of broadcast packets. The number of packets per burst and the interval between bursts can be specified on the command line. It is useful for testing devices to see how tolerant they are of broadcast bursts.

testConnectApp

• Added missing extern "C" required to build dynamically on Windows.

Many source files

• Replaced tab characters with spaces. Tabs in source files are bad practice.

drvAsynIPPort

Fixed problem with UDP broadcast sockets. Previously sending broadcast messages worked correctly, but any responses from clients to those messages were rejected by the IOC system. This is because connect() and send() were being called. This is not correct, in this case connect() should not be called and sendto() should be called rather than send(). Also changed the code to call recvfrom() rather than recv()for UDP sockets. This returns the source address information, which is now printed with the source message length and contents if ASYN_TRACEIO_DRIVER is set.

Test applications

New test application, testOutputReadbackApp. Tests that the initial values of output records are set correctly.

New test application directory, testBroadcastApp. It has 2 applications for sending broadcasts and reading responses. One uses asyn, and the other native OS calls. The native OS call version only builds on Linux. This is used for testing the UDP broadcast problem fixed in this release.

Repository location

The source code repository was moved from https://svn.aps.anl.gov/epics/asyn to https://github.com/epics-modules/asyn. This will make it much easier for others to collaborate in the development of asyn.

drvAsynSerialPort

• Fix to save the previous values of the baud rate and termios structures before attempting to change them. If the change fails then the previous values are restored. Previously if a call to setOption() failed, then subsequent calls to getOption() would return the value that failed. Now subsequent calls to getOption() will return the previous valid option, which is presumably the one that the underlying OS driver is still using. Thanks to Ron Sluiter for the initial fix to this problem.

drvAsynSerialPortWin32

• Fix to automatically prefix COM port names with "\\.\" unless the port name already begins with "\\.\". This is valid syntax for all COM ports and is required for all ports except COM1-COM9. Note that when passing the port string from the iocsh if the port name is in quotes then backslash characters should not be escaped, e.g. "\\.\COM10". If the port name is not in quotes then the backslash characters must be escaped, e.g. \\\\.\\COM10. Thanks to Freddie Akeroyd for the initial version of this fix.

drvAsynIPPort

• Added the option to automatically close the port when there is a read timeout. This was done by changing the syntax of the drvAsynIPConfigure command from
  drvAsynIPPortConfigure("portName","hostInfo",priority,noAutoConnect,noProcessEos) to
  drvAsynIPPortConfigure("portName","hostInfo",priority,noAutoConnect,userFlags)
  userFlags: bit-wise ORed:
  • Bit 0 is USERFLAG_NO_PROCESS_EOS. If 0 then asynInterposeEosConfig is called specifying both processEosIn and processEosOut.
  • Bit 1 is USERFLAG_CLOSE_ON_READ_TIMEOUT. If 1 then the (TCP) socket will be closed when read() returns a timeout.
  • Bit 2--31 are reserved, set to zero.
  This change is backwards compatible since bit 0 is the same as the previous noProcessEos. Thanks to Torsten Bogershausen for this change.

devEpics

• Changed error reporting if pasynManager->queueRequest returned an error. Previously error messages were printed on every call, which would cause a large number of messages if the port was disconnected and records were periodically processing. Now an error message is only printed when the status return from pasynManager->queueRequest changes, so there will be single message for each record when a port disconnects and a single message when it reconnects.
• Changed error handling when any of the following errors are encountered:
  • pasynManager->queueRequest returns an error
  • The I/O call to the driver (e.g. pasynInt32->read()) returns an error.
  • The interrupt callback function is passed an error code in pasynUser->auxStatus
  Previously such errors did correctly set the record alarm status and severity. However, they did not return an error code to the function that called the device support write() or read() function. They also updated the record VAL or RVAL field when there was an error, which they should not. Now these functions return an error code and do not update VAL or RVAL when there is an error.

asynPortDriver

• Changed the status return from pasynManager->queueRequest from asynError to asynDisconnected if a port with ASYN_CANBLOCK=0 is disconnected.

asynShellCommands

• Added new command asynSetMinTimerPeriod, and corresponding C function. This command is currently only implemented on the WIN32 architectures, and simply prints an error message on all other architectures. It allows reducing the minimum time for epicsThreadSleep() from the default of 10 ms to the minimum supported Windows system value. On most systems this is 1 ms. Note that epicsThreadSleepQuantum() does not seem to report the correct value whether or not this function is called. On my system it always reports 0.0156 seconds. Note also that this function really has nothing to do with asyn, and the equivalent functionality should probably be moved into EPICS base at some time.

Deadlock issue

• There is a potential deadlock issue when using input records with SCAN=I/O Intr. This is discussed in the KnownProblems document. Thanks to Ambroz Bizjak for finding this problem.

devEpics

• Added capability for output records to update their values with a callback from the driver. This is enabled by adding the following info tag to the output record:
  info(asyn:READBACK, "1")
  The output record value will update any time the driver does a callback with a new value. The output record alarm status and severity will also update to reflect the value passed by the driver in pasynUser->auxStatus. The driver callbacks use the same ring buffer mechanism that input records do. The default ring buffer size is 10 for all records except stringin, stringout, and waveform records, for which the default is 0.
• Updated the test application testErrorsApp to demonstrate and test output record updates.
• Added ring buffer support to devAsynOctet. The default ring buffer size is 0 for all records in devAsynOctet, i.e. stringin, stringout, waveform in and out. Non-zero values can be specified with the asyn:FIFO info tag.
• Changed the info tag that is used to control the size of the ring buffer for driver callbacks. The tag has been changed from FIFO to asyn:FIFO, i.e. adding the asyn: namespace to prevent future conflicts with any other facility. NOTE:This change is not backwards compatible, any databases using the FIFO info tag will need to be changed to asyn:FIFO. It is not expected that this will affect many users, since the default ring buffer size is used in nearly all cases.

asynPortDriver

• Bug fix: If setParamStatus() was called for an array or generic pointer parameter then callParamCallbacks() could return prematurely and not do the callbacks for all modified parameters.

devAsynOctet

Fixed 2 bugs:

1. The interrupt callback function for stringin records (e.g. with SCAN=I/O Intr) did not null-terminate the string if the driver returned MAX_STRING_SIZE or more characters. Thanks to Freddie Akeroyd for fixing this.
2. The interrupt callback function for stringin and waveform records were not properly locking the record when modifying the record fields.

devAsynXXXArray

Added optional support for ring buffers with waveform records. Ring buffers were added to asyn device support for scalar (non-array) records in R4-10. To enable ring buffer support on a waveform record the record info tag FIFO can be set to a value greater than 0. For example this line in a db file for a waveform record sets a ring buffer size of 20 elements.

       info("FIFO", "20") 
    

Ring buffers are only used when records have SCAN=I/O Intr. They allow the record to process all of the arrays from a rapid burst of callbacks from the driver. However, because Channel Access does not provide any buffering for arrays, even if the record processes for each new array callback, Channel Access will not necessarily send events for each value, because it just sends the current record value when the CA callbacks are done.

A new test application, testArrayRingBufferApp was added to test this array ring buffer support. A new iocBoot/ioctestRingBuffer directory was also added.

Interfaces

Added new asynOptionSyncIO interface. This interface is needed so that the asynOption interface can be called synchronously when it is OK to block.

Building asyn with only using libCom from EPICS base

It has always been asserted that except for devEpics asyn only really depends on libCom from EPICS base. People who are interested in using asyn drivers from other control systems, want to minimize the dependencies of libraries from EPICS base. The following lines have been added to asyn/configure/CONFIG_SITE:

-------------------------------------------------------------------------------
# If you want to build asyn so the only dependency on EPICS base is libCom then set the following flag
#EPICS_LIBCOM_ONLY=YES
-------------------------------------------------------------------------------

If EPICS_LIBCOM_ONLY is YES then the build is done so that only libCom is needed. This does the following:

• Omits building all of the device support for EPICS records
• Omits building the test applications that create IOC applications
• Sets asyn_LIBS to Com rather than $(EPICS_BASE_IOCS_LIBS) when creating the library
• Changes the logic in asynPortDriver which uses interruptAccept. If EPICS_LIBCOM_ONLY is set then dbDefs.h is not included and interruptAccept is defined as a static variable set to 1.

A new C++ class was added, asyn/asynPortClient/asynPortClient.cpp. This class makes it easy to write a C++ application that starts existing asyn port drivers and communicates with them over standard asyn interfaces without running an IOC.

A new test application directory was added testAsynPortClient. This tests running C++ applications that communicate with asyn port drivers without running an IOC. This currently contains a single test application, testAsynIPPortClient.cpp. This program creates an asynIPPort driver, and uses the command line arguments to set the hostInfo string, a single command string to send to the server, and optionally the input and output EOS. It then prints out the response from the server. There are 3 example shell scipts that show how to use testAsynIPPortClient to communicate with a Web server, XPS motor controller, and a telnet host respectively.

drvAsynUSBTMC

Bruno Martins found and fixed a problem with data transfers that spanned multiple USB packets.

drvAsynIPPort.c

Added capability to specify the local port that the server should use for the connection. Normally the local host choses a random local port that it binds to and passes to the server. There are a few servers that only accept a specific local port or range of local ports, for which this capability is required. The new syntax is:

<host>:<port>[:localPort] [protocol]

For example

164.54.160.100:5000:10101 UDP

where 10101 is the optional local port number.

devEpics

Fixed all initialization routines so that if there is an error they do the following:

• Call recGblSetSevr(precord,LINK_ALARM,INVALID_ALARM)
• Set precord->pact=1
• return(INIT_ERROR), where INIT_ERROR=-1

Thanks to Nick Rees for these fixes.

Many source files

Fixed problem with location of #define epicsExportSharedSymbols and/or #include <epicsExport.h>. In previous versions these were placed immediately before the #include statements defining symbols for that source file . However, this was incorrect, they must be placed before all of the #include statements defining symbols for that DLL . This mistake causes the same symbol being defined with both dllExport and dllImport when building the DLL. The Visual Studio compiler does not even warn about this error, but it produces a fatal error with the GCC compiler under Cygwin.

asynManager.c

Fixed a bug in pasynManager->memMalloc. It could return a pointer that was not a multiple of 8 bytes. This led to subtle problems on some architectures (e.g. ARM) if a double was stored in the memory returned by memMalloc. Fixed the code so the pointer is always a multiple of 16 bytes (for future safety).

drvAsynUSBTMC

• Added driver for USB TMC (Test and Measurement Class) devices.
• Works on any system that provides libusb-1.0.
• Requires no kernel modifications.
• Tested with StreamDevice on Darwin, Linux and Windows.
• Enabled by setting DRV_USBTMC=YES in configure/CONFIG_SITE.

drvAsynSerialPort

Fixed bug: the port was not always being disconnected when the OS returned serious errors. This was preventing USB serial devices from disconnecting and reconnecting properly.

drvAsynSerialPortWin32 (Windows serial port driver)

Fixed 2 bugs in pasynOctet->read() when pasynUserTimeout=0:

• It was not returning immediately, it was waiting 16 ms.
• If there were no characters to be read it was not returning asynTimeout, it was returning asynSuccess. This prevented StreamDevice from working correctly.

drvVxi11

Fixed several 32/64 bit issues.

drvAsynIPServerPort

Added support for UDP servers, in addition to TCP servers. Thanks to Helge Brands and Dirk Zimoch from PSI for this.

drvLinuxGpib

Fix to Makefile to link with system libgpib library.

devEpics (asynInt32)

Sign-extend positive or unsigned values as well as negative values when DTYP=asynInt32 and asynMask value is non-zero.

devEpics (asynUInt32Digital)

Apply mask to longin and longout records.

asynPortDriver

Undid the change in R4-22 that changed the asynStdInterfaces from protected to private. That change was not compatible with areaDetector R1-9-1.

Removed paramList.h and paramList.cpp, put the code into asynPortDriver.cpp. These two classes had many interdependencies, so needed to include each other's header files. This was difficult when building dynamically on Windows, where imported and exported symbols need to be distinguished.

Added links to asynPortDriver.html to detailed tutorial on how to write a driver using asynPortDriver.

asynOctetSyncIO

Fixed a bug that could cause the values of *nbytesOut, *nbytesIn, and *eomReason to be garbage if an I/O error occurred.

Many files

Handled exporting symbols consistently, which is important when building dynamically for Windows with Visual Studio. Eliminated all references to epicsSharedSymbols, now just #include <epicsexport.h> just before the #include for the header file that defines symbols for this file, and after all other #include statements. Thanks to Peter Heesterman for the initial version of this fix.

configure directory

Updated the files in the configure directory to the versions from EPICS base 3.14.12.3. Paths to EPICS base and other modules are still defined in configure/RELEASE. However, other configuration values, such as LINUX_GPIB and DRV_USBTMC are now defined in configure/CONFIG_SITE, not in RELEASE.

asynDriver

Added support functions for setting timestamps in asyn port drivers. These can be used to set the timestamp when the port driver received data. The driver can then set the asynUser->timeStamp field to this value for all input records on read and callback operations. Records that have TSE=-2 will have this timestamp. There is support for registering a user-supplied function to provide the timestamp, which will override the default source that just calls epicsTimeGetCurrent().

Added the following new functions to pasynManager for timestamp support.

asynStatus (*registerTimeStampSource)(asynUser *pasynUser, void *userPvt, timeStampCallback callback);
asynStatus (*unregisterTimeStampSource)(asynUser *pasynUser);
asynStatus (*updateTimeStamp)(asynUser *pasynUser);
asynStatus (*getTimeStamp)(asynUser *pasynUser, epicsTimeStamp *pTimeStamp);
asynStatus (*setTimeStamp)(asynUser *pasynUser, const epicsTimeStamp *pTimeStamp);

Added the following shell commands for timestamp support.

asynRegisterTimeStampSource(const char *portName, const char *functionName);
asynUnregisterTimeStampSource(const char *portName);
   

Added a new bit to asynTraceMask, ASYN_TRACE_WARNING. This is intended to be used for messages that are less serious than ASYN_TRACE_ERROR, but more serious than ASYN_TRACE_FLOW.

Added new asynTraceInfoMask. This mask controls the information printed at the beginning of each message by asynPrint and asynPrintIO. Thanks to Ulrik Pedersen for help with this. The mask is defined with the following bits:

#define ASYN_TRACEINFO_TIME 0x0001
#define ASYN_TRACEINFO_PORT 0x0002
#define ASYN_TRACEINFO_SOURCE 0x0004
#define ASYN_TRACEINFO_THREAD 0x0008
    

• ASYN_TRACEINFO_TIME prints what has been printed in previous versions of asyn, the date and time of the message.
• ASYN_TRACEINFO_PORT prints [port,addr,reason], where port is the port name, addr is the asyn address, and reason is pasynUser->reason. These are the 3 pieces of "addressing" information in asyn.
• ASYN_TRACEINFO_SOURCE prints the file name and line number, i.e. [__FILE__,__LINE__] where the asynPrint statement occurs.
• ASYN_TRACEINFO_THREAD prints the thread name, thread ID and thread priority, i.e. [epicsThreadGetNameSelf(), epicsThreadGetIdSelf(), epicsThreadGetPrioritySelf()].

Added a new shell command to control this mask.

    asynSetTraceInfoMask port,addr,mask
    

Added asynTrace information to the output of asynReport if details >=1.

asynOctetSyncIO

Use simple lock/unlock operations rather than queueLockPort/queueUnlockPort for end-of-string manipulations (setInputEos, getInputEos, setOutputEos, getOutputEos). This ensures that these operations can take place even with the device disconnected.

devEpics

Finished the support for setting the record TIME field for input records, both for read operations and for callback operations (i.e. records with SCAN=I/O Intr). This work was begun in R4-20, but it was not complete. Fixes were made in devAsynOctet, devAsynInt32, devAsynFloat64, and devAsynUInt32Digital.

Changed the ring buffer overflow messages that are printed in devAsynInt32, devAsynFloat64, and devAsynUInt32Digital. These now use the new ASYN_TRACE_WARNING mask. Prior to asyn R4-20 these used ASYN_TRACE_ERROR, and in asyn R4-20 and R4-21 they used ASYN_TRACEIO_DEVICE.

drvAsynIPPort

Fixed a bug in calling poll(). Previously the status return from poll() was not being checked; it was assumed that when poll() returned the port either had new data or had timed out. This is not correct, because poll() can return prematurely with errno=EINTR if a Posix signal occurs before data is received or the timeout expires. This can happen, for example, when the Posix high-resolution timer routines (timer_create, etc.) are used in the IOC. The Prosilica vendor library uses the Posix timer routines, and there were problems using asyn IP ports in IOCs that were also running Prosilica cameras. The problem was fixed by calling poll() again if it returns EINTR and the desired timeout time has not expired since poll() was first called.

Added support for AF_UNIX sockets on systems that provide them.

drvAsynSerialPort

Support for all baud rates supported by the operating system.

Fix for XON/XOFF support from Dirk Zimoch.

Changes to work with EPICS base 3.15

Removed include of asyn.dbd, from drvAsynSerialPort.dbd, drvAsynIPPort.dbd, drvVxi11.dbd, and drvGsIP488.dbd.

Fixed Makefile rule for vxi11intr.h for parallel builds.

drvLinuxGpib

Fixes for eomReason and EOS handling from Dirk Zimoch.

asynOctetSyncIO interface

Removed the openSocket() function from this interface. This method really did not belong in this interface, since it just wrapped the call to drvAsynIPPortConfigure().

asynRecord

Added 4 additional baud rates to the BAUD field (460800, 576000, 921600, 1152000). However, the BAUD field can still only support 16 fixed baud rates because it is type DBF_ENUM.

Added a new LBAUD (long baud rate) field of type DBF_LONG. This allows selection of any baud rate, not just those in the BAUD menu. Changing the BAUD field changes the LBAUD field. Changing the LBAUD field changes the BAUD field to the appropriate selection if it is a supported value, or to "Unknown" if not.

Added support for ASYN_TRACE_WARNING and asynTraceInfoMask, including the opi screens.

asynPortDriver

Added new virtual methods to support Eos operations on the asynOctet interface. These are setInputEosOctet(), getInputEosOctet(), setOutputEosOctet(), getOutputEosOctet(). Changed the report() method to print the current values of the inputEos and outputEos.

Added new virtual methods for timestamp support. These are updateTimeStamp(), setTimeStamp(), getTimeStamp().

Added new function getAsynStdInterfaces() to access the asynStdInterfaces structure, which is now private.

Changed the functions that do callbacks when callParamCallbacks() is called to call pasynManager->getTimeStamp() and set the pasynUser->timestamp field to this value in the callbacks.

Changed the base class readXXX() functions (e.g. readInt32(), readFloat64(), etc.) to call pasynManager->getTimeStamp() and set the pasynUser->timestamp field to this value. The readXXX() functions in derived classes should also do this, so that records with TSE=-2 will get the timestamp from the driver.

testErrorApp, iocTestErrors

Added an example of a user-supplied timestamp source.

testAsynPortDriver

Added calls to lock() and unlock() in simTask. This was previously a serious omission in this example driver. Thanks to Kay Kasemir for spotting this.

asynDriver

Restored the original versions of pasynManager->lockPort and unlockPort that were used in asyn prior to R4-14. These versions just call epicsMutexLock and epicsMutexUnlock. In R4-14 these versions were replaced with versions that queued requests to lock the port. The R4-14 versions fixed a problem with the interfaces/asynXXXSyncIO functions, but it has become clear that the original versions are useful in some circumstances. The change was done as follows:

• The lockPort and unlockPort functions in R4-20 were renamed to pasynManager->queueLockPort and queueUnlockPort.
• The interfaces/asynXXXSyncIO functions were all changed to call the queueLockPort and queueUnlockPort, so they function identically to how they have since R4-14.
• The versions of lockPort and unlockPort that existed prior to R4-14 were restored to pasynManager.

Changed the report() function so that if details<0 then asynManager does not print information for each device (address). It calls pasynCommon->report(-details) in this case so driver report functions will not be affected.

Changed the asynTrace print, printIO, vprint, vprintIO functions so they use EPICS_PRINTF_STYLE. This causes the GCC (version 3.0 and higher) and clang compilers to check the agreement of format strings with function arguments when using asynPrint() and asynPrintIO(), just as they do with printf(). This is very helpful in finding errors, and uncovered a number in asyn itself, which have been fixed.

devGpib

Changed event code readback to support both the original 'short' VAL field (EPICS 3.14 and earlier) and the new 'string' VAL field (EPICS 3.15 and later).

Added support for DSET_AIRAW and DSET_AORAW definitions.

Replace strcpy with strncpy in devCommonGpib.c to reduce possibility of errors. Use NELEMENTS macro in devSkeletonGpib.c rather than hardcoding. Thanks to Andrew Johnson for these.

devEpics

Improved the initMbboDirect function in devAsynUInt32Digital.c. If an initial value is read successfully from the driver it now sets the .Bn fields in the record. It also sets VAL rather than RVAL and returns 2 rather than 0. Thanks to Andrew Johnson for this.

Fixed a bug in devAsynUInt32Digital.c, which was missing a call to epicsMutexLock in interruptCallbackInput. Thanks to Angus Gratton for this fix.

Fixed a bug in devAsynXXXArray.h to handle case of multiple interrupt callbacks between record processing. Previously this would result in a call to the asynXXXArray->read() in the driver, which is not correct. The asynXXXArray device support does not have a ring buffer, so multiple interrupt callbacks between processing results in data being "lost", i.e. the record processes more than once with the same data. This is not really an error, but we now issue an ASYN_TRACEIO_DEVICE warning. This is analogous to ring buffer overflow for non-array data types.

Changed the "ring buffer overflow" messages from "ASYN_TRACE_ERROR" to "ASYN_TRACEIO_DEVICE", for devAsynFloat64. This was done for other device support in R4-20, but float64 was overlooked.

Interfaces

Added writeRead and writeReadOnce functions to the asynGenericPointer interface. Thanks to Florian Feldbauer for this addition.

Fixed a bug in asynCommonSyncIO that could cause a crash if the connect() function returned an error.

asynPortDriver

Added support for the asynOption interface. Added code to demonstrate and test this to the testErrors test application.

Added new method asynPortDriver::flushOctet(), which implements asynOctet::flush(). The base class implementation reproduces the behavior of asynOctetBase.c::flushIt, i.e. it calls pasynOctet->read() repeatedly with a timeout of 0.05 seconds until it gets no data back. But now drivers can implement their own version of flush() if a different behavior is desired, which was not previously possible.

Changed the code so that the length of string parameters returned in readOctet and octetCallback is now strlen(string)+1, rather than strlen(string). The length thus now includes the terminating nil. This fixes problems with clients that request long strings or subscribe to monitors with a length of 0, but don't check for a nil terminator.

Changed the meaning of the "details" argument in the asynPortDriver::report() function. The new meaning is:

• 0 = no details
• >=1: print details for parameter list (address) 0
• >=2: print details for all parameters lists (addresses)
• >=3: print interrupt callback information

Changed the connect() and disconnect() methods to return an error if the device address specified by the pasynUser is invalid (i.e. <-1 or >MAX_ADDR-1).

Fixed problem that was causing dynamic builds (e.g. SHARED_LIBRARIES=YES) to fail on Windows.

asynPortDriver/exceptions

Changes to allow compiling with the old vxWorks Tornado 2.0.2.1 compiler. Thanks to Dirk Zimoch for this fix.

Other

Many minor changes to avoid compiler warnings on Linux, vxWorks, and WIN32. Thanks to Dirk Zimoch for many of these.

asynManager

Fixed a bug that caused a deadlock if pasynManager->lockPort was called multiple times without calling pasynManager->unlockPort in between. Thanks to Sebastian Marsching from "aquenos GmbH" for this fix.

devEpics

Added support for setting the record timestamp from the driver, using a new field, pasynUser->timeStamp. The driver can set this field in read and callback operations.

Fixed a long-standing bug in devAsynXXXArray support for input waveform records with SCAN=I/O Intr. The data were being copied to the record without using dbScanLock. This meant that the data could change while the record was processing.

Changed the "ring buffer overflow" messages from "ASYN_TRACE_ERROR" to "ASYN_TRACEIO_DEVICE", so they do not appear by default. These messages are not really errors, but warnings that record processing it not keeping up with the rate of driver callbacks for records with SCAN=I/O Intr.

asynPortDriver

Fixed a bug when adding a parameter that already existed. Thanks to Hinko Kocevar for fixing this.

Documentation

Added new document "HowToDoSerial_StreamDevice.html".

Test programs

Added new test program, testConnectApp. It uses asynPortDriver, and has a polling thread that sends a user-defined string to a device and reads the response. It can be used to test the behavior when the device is disconnected and then reconnects, etc.

Build system

Added a rule in asyn/Makefile to fix a problem that could cause parallel make to fail.

Interfaces

Added a new interface, asynEnum. This interface is designed to allow drivers to set the strings, values, and severities for record enum fields. This can be done both at iocInit(), in init_record() with the pasynEnumSyncIO->read() function, and after iocInit via callbacks to device support.

devEpics

Added support for the new asynEnum interface for bo, bi, mbbo, and mbbi records in the asynInt32 and asynUInt32Digital device support. These records now attempt to read the initial values of the strings, values (mbbo and mbbi only) and severities for the enum fields. They also support callbacks on the asynEnum interface, so that enum values can be set dynamically at run-time.

Improved the support for setting the alarm status of records. Previously for records that were not I/O Intr scanned STAT was always to READ_ALARM or WRITE_ALARM, and SEVR was set to INVALID_ALARM. A new function, pasynEpicsUtils->asynStatusToEpicsAlarm() was added that converts asynStatus values to EPICS alarm values. This allows records to have STAT=TIMEOUT_ALARM, DISABLE_ALARM, etc. More values of STAT can be supported in the future by adding more values to the asynStatus enum.

Previously it was not possible for input records with SCAN=I/O Intr to have their alarm status set at all. This support has been added. Device support now uses the pasynUser->auxStatus field in the pasynUser passed to the callback function. If auxStatus != asynSuccess then the record alarm STAT and SEVR are set to values based on the asynStatus. asyn port drivers can now signal error status to clients in callback functions by setting pasynUser->auxStatus to asynSuccess, asynTimeout, asynError, etc. This change should be backwards compatible with all drivers because the pasynUser that is used for the callbacks is private to the callback function, and the auxStatus field is initialized to 0, which is asynSuccess.

Added new waveform record device support, asynInt32TimeSeries and asynFloat64TimeSeries. These use callbacks from the driver on those respective interfaces to collect a time series of values in a waveform record. Added new medm file asynTimeSeries.adl for this, and added an example waveform record to testEpics/Db/devInt32.db.

asynRecord

Fixed bugs that caused crashes if SCAN=I/O Intr was set at iocInit.

asynDriver and asynShellCommands

Added support for I/O redirection to the "dbior" and "asynPrint" commands.

asynPortDriver

Added support for the new asynEnum interface described above.

Added support in asynPortDriver for passing status information to clients in callbacks. Each parameter in the parameter library now has an associated asynStatus variable. New functions setParamStatus() and getParamStatus() are provided to access this variable. For example, if setParamStatus(paramIndex, asynError) is called then callParamCallbacks() will cause any input records with SCAN=I/O Intr to go into alarm state.

Moved asynPortDriver from the miscellaneous directory to its own directory. Improved the internals, but did not change the API. Thanks to John Hammonds for this.

Other

Added a new test application, testErrorsApp. This application uses a driver based on asynPortDriver to test error handling for all interfaces and all records support by the asyn standard device support (asyn/devEpics). It can be used to test error handling of records with both periodic scanning and I/O Intr scanning. It also tests the new asynEnum interface for setting enum strings, values, and severities at iocInit.

Removed the newline terminator from all messages in pasynUser->errorMessage. This formatting does not belong in the error message. Thanks to Lewis Muir for this.

drvAsynIPServerPort. Added call to epicsSocketEnableAddressReuseDuringTimeWaitState which fixes problems when the IOC is restarted and the port is still in TIME_WAIT state. Thanks to Lewis Muir for this.

Miscellaneous

Changes to avoid compiler warnings on 64-bit Darwin.

Changes to avoid compiler warnings and errors on older Solaris compiler.

Changed non-standard __VAR_ARGS__ to __VA_ARGS__ in asynPrint macro.

Added titles to EDM screens.

Added CSS BOY screens.

Fixed build problem. It was rebuilding the VXI11 code with rpcgen each time make was run, even if nothing had changed.

asynPortDriver

Cleaned up logic for callbacks on asynUInt32Digital interface. It was not doing callbacks if the value had not changed but an interrupt had occured for bits in the mask. This would happen when interrupts were only enabled on the falling or rising edges but not both. Added an additional form of setUIntDigitalParam that takes an interruptMask argument. In the previous releases drivers were calling setUInt32Interrupt for callbacks which is not correct. That function should only be called by device support to tell the driver which interrupts to recognize. These problems made the quadEM not work correctly with the ipUnidig.

drvAsynSerialPort

Bug fix on WIN32. It was waiting forever when timeout=0 when it should return whatever characters are available without waiting at all.

asynRecord

Fixed asynRecord.dbd to include promptgroup for fields that need it. Removed the SOCK field and asynSocketSetup.adl; the record no longer supports creating sockets, which can easily be done with the iocsh drvAsynIPPortConfigure command.

Added IXON, IXOFF, and IXANY fields for new XON/XOFF support on serial ports. These fields were added to the asynSerialPortSetup.adl medm screen.

devAsynXXXArray

For waveform output records the device support was always writing NELM elements, rather than NORD elements.

asynInterposeCom.c

Added missing include file, which caused asyn not to build on EPICS base versions before 3.14.10.

vxi11core_xdr.c

Removed unneeded declarations of 'register int32_t *buf' from many functions. Removed register keyword from remaining instances, because address was being taken, which is illegal in C, and was causing errors on vxWorks Pentium cross-compiler.

asynPortDriver

Added a global function, findAsynPortDriver(const char *portName), that returns a pointer to an asynPortDriver object given the asyn port name.

Added 3 new asynPortDriver methods: setUInt32DigitalInterrupt, clearUInt32Interrupt, getUInt32Interrupt. These were needed to complete the asynUInt32Digital support.

asynShellCommands

The asynOctetSetInputEos, asynOctetSetOutputEos, asynOctetShowInputEos and asynOctetShowOutputEos commands now take effect even if the port is not connected. This makes startup scripts more robust in the face of devices that are not accessible at IOC startup. Removed the unused "drvInfo" parameter from each of these functions.

drvAsynIPPort

Host name lookup is now deferred until port connection time. This makes startup scripts robust in the face of a device that is offline at IOC startup and has been offline for so long that it's DNS entry has been deleted.

Prevent reconnects during IOC shutdown. The IP Port exithandler runs before record scanning stops. In that interval, if a record is scanned then it will trigger a reconnect, and new connection can be shutdown without sending data, or without waiting for a reply. Some embedded TCP/IP stacks have problems dealing with this. Thanks to Michael Davidsaver for this fix.

drvAsynSerialPort

Added support for local serial ports on Windows, i.e.win32-x86 and windows-x64 architectures. Previously Windows local serial ports were only supported on the cygwin-x86 architecture.

Added support for serial line software handshake flags (ixon/ixoff/xany) on most architectures (e.g. Linux, Cygwin, Darwin, WIN32; see: man stty) and vxWorks (ixon only). Thanks to Dirk Zimoch for this.

Miscelleaneous

Changed asyn/Makefile to fix build dependencies for rpcgen of vxi11. Thanks to Michael Davidsaver for this fix.

drvAsynIPPort

Support has been added for terminal servers which support the TELNET RFC 2217`protocol. To communicate with such devices specify "COM" as the protocol in the drvAsynIPPortConfigure command. This allows port parameters (speed, parity, etc.) to be set using an asynRecord or asynSetOption commands just as for local serial ports.

drvVxi11

Fix from Benjamin Franksen to fix problem with reconnection.

The 'timeout' argument to vxi11Configure has been changed from double to string. This allows vxi11Configure to be called directly from the vxWorks shell.

vxi11

The third argument to the vxi11Configure command is now a bit-map. The least significant bit (value 0x1) remains the 'recover with IFC' control. The next-to-least significant bit (value 0x2) when set will cause all devices to be locked when a connection is made. This allows for cooperative exclusive access to devices.

asynManager

Fixed memory leak in lockPort() when an error occured, was not calling freeAsynUser(). Thanks to Andrew Starritt for finding this.

devEpics

Bug fix for devAsynInt32 and devAsynFloat64: it was not freeing the mutex in processAiAverage if numAverage==0, i.e. there had been no callbacks from the driver since the record last processed. This would hang the next thread that tried to take the mutex, typically the driver callback thread.

Bug fix for devAsynXXXArray: it was calling drvUserCreate in the port driver even if there was no userParam in the link, which could crash the driver.

Many files

Changes to allow building dynamically on WIN32 (i.e. making DLLs).

Makefiles

Changes to allow building on Cygwin 1.7.x or 1.5.x; replaced rpc with $(CYGWIN_RPC_LIB), which allows it to link with rpc on 1.5.x and tirpc on 1.7.x. You need to add one of the following 2 lines to base/configure/os/CONFIG_SITE.cygwin-x86.cygwin-x86

    For Cygwin 1.7.x:
    CYGWIN_RPC_LIB = tirpc

    For Cygwin 1.5.x
    CYGWIN_RPC_LIB = rpc
  

asynDriver

Fixed bugs in connection management. Releases 4-11 through 4-13 had the following problems:

• If a port was multi-device and auto-connect, and was only accessed using SyncIO calls, then asynManager would never connect to the devices. If regular queued requests were used, which happens if an asyn record is connected to that port and device, then it would connect once the first queue request was done.
• For all auto-connect ports there was a race condition, such that the port might or might not be connected when the first queued request or SyncIO call was done on that port. This arose because when the port registered its asynCommon interface, asynManager queued a connection request on that port. But if that connection request callback, which executes in the portThread, had not yet occurred when a queue request or SyncIO call was made, then those operations would be rejected because the port was not yet connected.

These problems were fixed by doing the following:

• When the port registers its asynCommon interface and asynManager queues the connection request, it now waits for a short time for the connection callback to complete. The default time is 0.5 seconds, but this time can be changed with a call to the new function pasynManager->setAutoConnectTimeout(double timeout). This function can be accessed from the iocsh shell with the new asynSetAutoConnectTimeout(double timeout) command. This short timeout is designed to allow devices time to connect if they are available, but not to excessively slow down booting of the IOC by waiting, for example, for the system timeout on TCP connections. Note that this change means that it is now very likely that the pasynCommon->connect() call will occur as soon as the asynCommon interface is registered. As noted in the R4-12 release notes, this means that the driver must have already done all initialization required for the asynCommon->connect() callback before it registers the asynCommon interface!
• There is an additional new function, pasynManager->waitConnect(asynUser *pasynUser, double timeout), which will wait for the for the port to connect, up to the specified timeout. This function can be called from the iocsh with the new command asynWaitConnect(const char *portName, double timeout). This function makes it possible to wait longer for a port to connect then the current value of the global autoconnect timeout described above.

Fixed problems with the SyncIO calls, which were caused by the implementation of pasynManager->lockPort():

• The SyncIO calls (e.g. asynOctetSyncIO) are implemented by calling pasynManager->lockPort(), executing the I/O in the current thread, and then calling pasynManager->unlockPort(). These SyncIO functions are designed to be called from threads that are allowed to block, such as SNL programs, or other drivers. The problem with the previous implementation was that pasynManager->lockPort() immediately took the port mutex when it was available, rather than queueing a request to take the mutex. This could lead to one thread effectively getting exclusive access to the port, even if other threads had queued requests or tried to do SyncIO calls themselves. For example, if a device could send unsolicited input, then one might create a thread that simply called pasynOctetSyncIO->read() with a short timeout in a tight loop. The problem with this was that as soon as that thread released the port mutex when the read timed out, it would take the mutex again, blocking other threads that were trying to access the port.  Previously the only solution to this problem was to add a short epicsThreadSleep() in the read loop.
• This problem has been fixed by reimplementing pasynManager->lockPort(), which now queues a request to access the port and then blocks until the queue request callback runs in the portThread. When the queue request runs, the thread that called pasynManager->lockPort() executes, and the portThread blocks, until pasynManager->unlockPort() is called.
• Note that this change to lockPort() does change its functionality in one respect: previously it was OK to call lockPort() on a port that was not connected. This is no longer possible, because the queueRequest call in lockPort will now return an error if the port is not connected.
• The change to lockPort did not require any changes to the asynXXXSyncIO functions except asynCommonSyncIO. The asynCommonSyncIO->connectDevice and asynCommonSyncIO->connectDevice calls cannot use lockPort() any more, because as noted above it does not work with disconnected ports. Rather, these functions now directly queue a connection request with a private callback function to connect or disconnect the port.

vxi11

Fixed a bug in driver initialization. The driver had not completed all required initialization before it called pasynGpib->registerPort. Because pasynGpib->registerPort registers the asynCommon interface, that now normally triggers an immediate callback to vxiConnect, and the driver was not yet properly initialized to handle that callback.

Added an additional example driver, asynPortTest, that uses asynPortDriver. It implements the asynInt32, asynFloat64, and asynOctet interfaces to communicate with the echo server using asynOctetSyncIO calls. This tests nested SyncIO calls. Added a new startup script, database and medm screen for testing this new driver.

devGpib

Added a call to asynOctet->flush() just before the call to asynOctet->write() operation when doing write/read operations. This eliminates any stale input that may have already been sent by the device and would otherwise be incorrectly returned by the read operation.

interposeEOS

Now behaves properly even when eomReason is NULL.

testIPServerApp

Changed ipEchoServer2.c to eliminate the epicsThreadSleep(0.01) in the listener thread. This sleep is no longer necessary because of the change to lockPort described above, so the example program was changed to test and demonstrate this.

asynPortDriver

Fixed bug in getXXXParam. It was not returning error status when a parameter was undefined. This caused device support to use undefined values for output records, because the initial read from the driver during device support initialization did not return an error it should have.

asyn/Makefile

Change some dependencies to fix parallel (-j) make problem.

drvAsynIPPort

A return of 0 from a read of a TCP stream is treated as an END condition rather than as an error. This makes it easier to handle devices that close the connection at the end of a reply.

Support has been added for devices such as web servers that require a connect at the beginning of each transaction. To enable this behaviour, specify "http" as the protocol in the drvAsynIPPortConfigure command and ensure that each transaction ends with a read that detects the broken connection from the device. Note that the device will always appear connected. The connect/disconnect around each transaction is handled within the drvAsynIPPort driver.

drvAsynIPServerPort

Corrected the documentation to state that only the TCP/IP protocol is supported, not UDP.

asynGpib

Fix problem with NULL-pointer dereferences.

drvVxi11

Bug fix from Benjamin Franksen for devices that don't support IRQ.

devEPICS

The ring buffer code in devAsynFloat64.c, devAsynInt32.c and devAsynUInt32Digital.c has been improved. Previously when ring buffer overflow occurred during a callback the new value was simply discarded. This meant that when the record processed at the end of a rapid burst of callbacks it would not contain the most recent value. Now when overflow occurs the oldest value is removed from from the ring buffer and the new value is added, so that the record will contain the most recent callback value after a burst.

asynUInt32Digital: asynMask values implying shifts greater than 16 bits are now supported.

asynRecord

Fixed bug which caused an error when writing or reading in binary format if the driver did not implement the get(Input/Output)Eos functions. This bug was introduced when readRaw and writeRaw were removed from asynOctet in release 4-10.

Additional makeSupport.pl template

makeSupport.pl -t streamSCPI <name> creates skeleton stream protocol and database files for a SCPI (IEEE-488.2) device.

asynPortDriver

Fixed bug in readInt32Array, not taking lock where needed.

Added drvInfo strings to parameter lists, and new methods to support this: createParam(), findParam(), getParamName(). The report functions now print out the drvInfo strings, which is very useful. The base class drvUserCreate() method can now be used without reimplementing in derived classes, because the parameter names are now available to the base class. Removed asynParamString_T and drvUserCreateParam(), which are no longer needed. The testAsynPortDriver test application has been updated to use these new features.

asynManager

A problem was introduced in R4-11 by not starting the autoconnect process until iocInit(), and operations that do not use the XXXSyncIO functions thus fail before iocInit(). This means, for example, that calls to asynSetOption() to set serial port parameters fail if done in a startup script before iocInit(). R4-12 fixes this problem by decoupling autoconnect operations from iocInit(). NOTE: The first call to the pasynCommon->connect() function now happens almost immediately after pasynManager->registerInterface() is called for the asynCommon interface. This timing is different from all previous asyn releases, and it means that port drivers must initialize everything required by asynCommon->connect() before they register the asynCommon interface. This may require minor re-ordering of the initialization sequence in some drivers.

drvAsynSerialPort

Requests to set end-of-string values or serial port parameters are accepted even when the port is not connected. The request takes effect when the port is connected. This makes IOC startup more robust in the face of network or USB to serial adapters that may be unavailable on startup.

asynInt32Average, asynFloat64Average

If record is processed before new data have arrived (numAverage==0) set record to UDF/INVALID, set UDF to TRUE and leave value unchanged.

asynManager

asynReport at detail level 0 now reports only disconnected subaddresses.

The autoconnect code has undergone considerable modification. When a port is registered with autoConnect true, or whenever a port disconnect exception is raised on an autoConnect port, an attempt at connection occurs immediately followed by retry attempts at 20 second intervals. Attempts to queue requests to a disconnected port (even an autoConnect port) will be rejected. These changes have been made to reduce the occurences of 'connection flurries' and to ensure that requests do not languish in the queue when connections are broken.

Setting the trace mask or trace I/O mask for a port now also sets the trace mask or trace I/O mask for every device associated with that port.

Passing a NULL pasynUser argument to the setTraceMask and setTraceIOMask will set the asynBase (default) trace mask or trace I/O mask. To do this from the iocsh pass a zero length portName string to the iocsh commands asynSetTraceMask or asynSetTraceIOMask.

asynDriver.h

Add new version macros (ASYN_VERSION, ASYN_REVISION and ASYN_MODIFICATION). These are guaranteed to be numeric.

Add new asynStatus codes, asynDisconnected and asynDisabled. Attempts to queue a request to a disconnected or disabled port return these codes, respectively. Future changes to record support may propogate these to the record alarm status field.

asynInterposeEOS

Increased the size of the input buffer from 600 to 2048 bytes.

VXI-11

Fix from Takashi Asakawa, Yokogawa Electric Corporation, Japan to allow link identifiers with a value of zero.

devEpics (devAsynInt32, devAsynFloat64, devAsynUInt32Digital)

• Ignore callbacks when interruptAccept=0, because waiting for interruptAccept (the previous behavior) hangs iocInit with synchronous drivers. The first record process on input records will now read directly from driver, which will get the latest value.
• Added mutex around epicsRingBytes calls
• Added ASYN_TRACEIO_DEVICE when reading value from ring buffer

testIPServerApp

Add some cleanup code to eliminate memory/socket leaks.

asynOctetSyncIO

The asynOctetSyncIO openSocket method will be removed in the next release. This method is redundant since it is no different than calling the connect method with the port name from a previous drvAsynIPPortConfigure command.

asynPortDriver

A new C++ base class called asynPortDriver from which real asyn port drivers can be derived. It greatly simplifies the code required to write a new asyn port driver. It is documented here.

testAsynPortDriverApp

A new test application to demonstrate the use of the new asynPortDriver C++ class. This driver simulates a simple digital oscilloscope, and includes a C++ driver, EPICS database, medm screen, and an example iocBoot directory in ioctestAsynPortDriver. It is described in the documentation for asynPortDriver.

asynOctet

Ths asynOctet writeRaw and readRaw methods have been removed. In most cases, if your code now calls readRaw or writeRaw it should be safe to simply change these calls to their non-Raw equivalent. If you're paranoid about someone interposing the end-of-string processing layer you could add something like the following to ensure that there is no end-of-string to match:

pasynOctet->setInputEos(asynOctetPvt,pasynUser,NULL,0);

If you need to switch to 'raw' mode for a while and then back to 'eos mode', you can use code similar to that in devGpib.c:readArbitraryBlockProgramData:

char saveEosBuf[5];
int saveEosLen;
.
.
.
status = pasynOctet->getInputEos(asynOctetPvt,pasynUser,saveEosBuf,sizeof saveEosBuf,&saveEosLen);
if (status != asynSuccess) {
    epicsSnprintf(pasynUser->errorMessage,pasynUser->errorMessageSize,"Device EOS too long!");
    return -1;
}
if (saveEosLen)
pasynOctet->setInputEos(asynOctetPvt,pasynUser,NULL,0);
.
.
.
.
if (saveEosLen)
    pasynOctet->setInputEos(asynOctetPvt,pasynUser,saveEos,saveEosLen);

When compiling your code against this new version of asyn you should pay particular attention to warning messages of the form "warning: initialization from incompatible pointer type". These are a good indication that you're initializing an asynOctet structure with the old-style I/O methods.

asynManager

Add a strStatus method to convert an asynStatus code to a string.

drvAsynIPPort

Cleaned up timeout handling.

Fix memory leak (one epicsAtExit entry was allocated for every connect).

asynReport no longer reports as connected a port which has successfully disconnected.

Improved diagnostic messages.

drvAsynIPServerPort

Fixed bugs that caused the thread that listened for new connections to exit when errors occurred. Such errors included too many simultaneous connections.

interfaces

Added three new array interfaces, asynInt8Array, asynInt16Array, and asynFloat32Array. asynInt8Array and asynInt16Array are the same as asynInt32Array except that the data types are for epicsInt8 and epicsInt16 respectively. asynFloat32Array is the same as asynFloat64Array except that the data type is for epicsFloat32.

Added asyn(XXX)ArraySyncIO for synchronous I/O to all array interfaces.

Added new asynGenericPointer interface. The datatype for this interface is void*, so it can be used to pass a pointer to anything. Includes asynGenericPointerSyncIO for synchronous I/O.

Added asynStandardInterfaces.h and asynStandardInterfacesBase.c to simplify driver initialization when using the standard asyn interfaces defined in asyn/interfaces (common, octet, int32, etc.)

devEpics

Undid the change that was done in R4-9 with direct calls to dbScanLock and process in the interrupt callback functions. This could lead to deadlocks in some circumstances. The original reason for changing from scanIoRequest to (dbScanLock, process) was because callback values would be lost if the callbacks came so close together in time that the single callback value stored in device support was overwritten before scanIoRequest could process the record. This problem has been fixed by adding a FIFO (ring buffer) to the device support for the scalar interfaces asynInt32, asynUInt32Digital, and asynFloat64. The ring buffer is only created when the record is put into I/O Intr scan, so the storage is not allocated for records that are not I/O Intr scanned. The ring buffer default size is 10 values, but this can be changed on a per-record basis using the dbInfo string "FIFO" with a value such as "100".

Added support for bi and bo records for asynInt32 interface. Previously these records were only supported for the asynUInt32Digital interface.

Added waveform record device support for the asynInt8Array, asynInt16Array, and asynFloat32Array interfaces. These are the same as devAsynInt32Array and devAsynFloat64Array with changes to the data types.

asynShellCommands

Remove duplicate windows function decorations.

asynInterposeEos

Improved diagnostic messages.

testIPServerApp/ipSNCServer.st

Fixed timeout bug introduced in R4-6 when timeouts of 0.0 and -1.0 were defined.

devEpics

Replaced scanIoRequest with direct call to rset->process in interrupt callback routines in all device support. Without this fix if another interrupt occurred before the first scanIoRequest was complete bad things could happen. The data from the first interrupt would be lost, and the read function in the driver would be called when it should not have been.

drvAsynSerialPort

Added support for 28,800 baud for those architectures which support this unusual speed.

Added stub routines for WIN32 so that a separate DBD file is no longer needed.

drvAsynIpPort

Added short delay in cleanup routine so sockets would close cleanly.

asynInterposeEos.c

Fixed bug which caused asynOctetRead() to return prematurely with eomReason=ASYN_EOM_CNT if the port driver returned 600 bytes in readRaw().

RTEMS

Build driver for Greensprings IP-488 if IPAC is defined in configure/RELEASE.

VXI11

Added stub routines for WIN32 so that a separate DBD file is no longer needed.

VXI-11

Avoid duplicate clnt_destroy operations.

devEpics/devAsynInt32

Added support for link specification of

@asynMask(portName,addr,nbits,timeout)drvParams

in addition to the previous support for

@asyn(portName,addr,timeout)drvParams

This allows device support to work with drivers that cannot return meaningful values in pasynInt32->getBounds because they do not know the range of the device. This is true, for example of Modbus ADCs. The nbits parameter is defined as follows:

  nbits > 0  Device is unipolar with a range from 0 to 2^nbits-1
  nbits < 0  Device is bipolar with a range from -2^(abs(nbits)-1) to 2^((abs(nbits)-1)-1
           Values returned on the asynInt32 interface will be sign extended
           using the sign bit (e.g. bit abs(nbits)-1 starting at bit 0).
  

devEpics/devAsynInt32Array, devAsynFloat64Array

Added support for callbacks from driver to device support. This allows waveform records to have I/O Intr scanning, as already supported for other records in devEpics.

devEpics/asynEpicsUtils

Changed parser so it requires an exact match to "asyn(" or "asynMask(". Previously it tolerated other characters before the "(", and in particular it accepted "asynMask(" when "asyn(" was expected. Note that this change could cause problems with database files if they did not follow the documented syntax, which has no white space between "asyn" or "asynMask" and the "(" character.

asynManager

Fix errors in format strings for asynPrint.

Temporary fix to asynReport thread for Cygwin. If the amount of output is small the thread exists for a very short time, and this causes a crash. The fix is a short wait, but it should really be fixed in base/src/libCom/osi/os/posix/osdThread.c.

testIPServerApp

Fix problem with Makefile.

devGpib

Add more SCPI commands to devGpib template.

drvAsynIPPort

Close sockets on application exit. This is very important for vxWorks, otherwise sockets are not closed cleanly which often leads to problems when the IOC reboots.

drvGsIP488

Hold off SRQ callbacks until iocInit.

drvAsynSerialPort

Clean up operation on POSIX/termios systems (everything but vxWorks). The old mechanism was prone to polling during read operations rather than using the termios read timeout mechanism.

devGpib

asynRecord sets line-buffering on trace file.

Peter Mueller provided code to remove/restore a device from/to the SRQ polling list.

drvGsIP488.c

Clean up dangling 'default' statement.

devGpib

Fixed error in GPIBACMD operations.

linuxGpib

Patches from Gasper Jansa to improve option handling.

devEpics

Fixed null pointer dereference for all device support when SCAN=I/O Intr and asyn port could not be found.

asynRecord

Fixed buffer overflow error when NRRD>40 and IFMT=ASCII.

asynGpib

Read method now sets return status and *eomReason properly.

drvAsynIPPort/drvAsynSerialPort

*eomReason now set to ASYN_EOM_CNT when read count has been satisfied.

Fix timeout settings on RTEMS.

Add support for UDP broadcasts. Specify "UDP*" and the network broadcast address in the port configuration command:
drvAsynIPPortConfigure("L0", "192.168.1.255:1234 UDP*", 0, 0, 0)

drvAsynSerialPort

Full support for new timeout semantics (timeout<0 means "wait forever for characters to arrive", timeout=0 means "return characters immediately available", timeout>0 means "return a timeout status if no characters are received within the specified number of seconds").

drvAsynIPPort/drvAsynSerialPort

Fixed NULL pointer dereference.

drvAsynIPPort

Previous versions of drvAsynIPPort.c (1.29 and earlier, asyn R4-5 and earlier) attempted to allow 2 things:

1. Use an EPICS timer to time-out an I/O operation, such as send(), recv() and connect().
2. Periodically check (every 5 seconds) during a long I/O operation to see if the operation should be cancelled.

Item 1) above was not really implemented because there is no portable robust way to abort a pending I/O operation. So the timer set a flag which was checked after the poll() was complete to see if the timeout had occured. This was not robust, because there were competing timers (timeout timer and poll) which could fire in the wrong order.

Item 2) was not implemented, because asyn has no mechanism to issue a cancel request to a driver which is blocked on an I/O operation.

Since neither of these mechanisms was working as designed, the driver has been re-written to simplify it. If one or both of these are to be implemented in the future the code as of version 1.29 should be used as the starting point.

If pasynUser->timeout < 0 an infinite timeout is now used.

Fixed bug so that ports connected with a file descriptor in pasynUser->reason execute code to set timeouts.

Fixed bug to return error if pasynCommon->connect is called when port already connected.

asynTrace

Added two new functions which are related to pasynTrace->print and pasynTrace->printIO the way vprintf is related to printf.

• pasynTrace->vprint Same as pasynTrace->print except that instead of a variable of arguments it takes a va_list argument as its last parameter.
• pasynTrace->vprintIO Same as pasynTrace->printIO except that instead of a variable of arguments it takes a va_list argument as its last parameter.

asynManager

Changed pasynManager->connectDevice for ports which have the properties autoConnect=1 and isConnected=0. In this case a request is queued to call asynCommon->connect for that port. This ensures that ports that have a pasynUser connected to them will report being connected even if no I/O has yet been done. Previously such ports reported a disconnected state until the first I/O or operation such as setTraceMask. This was confusing.

Clarify documentation on meaning of pasynUser->timeout. Previously there was no documented method of specifying an "infinite" timeout to a driver, and the meaning of timeout=0.0 was not defined. The new definitions are:

> 0.0 Wait for up to timeout seconds for the I/O to complete

= 0.0 Peform any I/O that can be done without blocking. Return timeout error if no I/O can be done without blocking.

< 0.0 Infinite timeout. Wait forever for I/O to complete.

devEpics

Fixed bugs with asynFloat64Average device support. The wrong interrupt function was being called, and UDF was not being cleared.

memMalloc/memFree

memMalloc was allocating the amount of memory the caller requested rather than the amount required for the freeList. If memFree was called and the memory reallocated to a user requesting a larger size, memory corruption occured. This is fixed.

SyncIO routines

If the connect call fails the asynUser is no longer freed. Instead a message is put into asynUser.errorMessage. The caller must call disconnect in order to free the storage for the asynUser.

The SyncIO routines no longer call asynPrint if there is an error and there is a valid asynUser available. Rather they return an error message in pasynUser->errorMessage. The SyncIO*Once functions still call asynPrint for errors, because they do not have a way of returning an error message.

Serial, TCP/UDP/IP

Handle 0-length write requests.

TCP/UDP/IP

Added drvAsynIPServerPort to support TCP and UDP socket servers.

Added iocBoot/testIPServer to test TCP server support.

drvAsynIPPort now closes TCP sockets when remote system closes connection.

drvAsynIPPort connect function now uses pasynUser->reason as a file descriptor if it is > 0. This allows drvAsynIPServerPort to re-use asyn ports it creates.

Made drvAsynIPPort add null byte at end of input if there is room.

Made drvAsynIPPort:readRaw set eomReason to 0. It was not setting eomReason at all previously.

Serial

Made drvAsynSerialPort add null byte at end of input if there is room.

Made drvAsynSerialPort:readRaw set eomReason to 0. It was not setting eomReason at all previously.

Interfaces

Added asynCommonSyncIO for synchronous support of the asynCommon interface.

VME GPIB

Add delay loops to get these boards to work with faster VME CPU modules.

VXI11

Better support was provided for VXI-11.3 controllers, i.e. talking directly to an ethernet port on an instrument. In particular a TDS3054B was tested.

WARNING: The VXI-11.1 ansd VXI-11.3 standards do NOT allow access to GPIB lines, i.e. conmmands like Untalk/Unlisten are not possible. The previous support issued these commands after each read or write. Some really old GPIB devices may fail. If so the device specific code must be modified to sent these commands separately.

win32

Changes were made to allow asyn to build on native Windows (win32-x86) architecture.

There are two asyn components that do not yet work on win32-x86.

1. Local serial ports. The asyn uses the "termios" API for serial ports, and termios is not available for native Windows.
2. VXI-11. The asyn VXI-11 support uses the Sun XDR API, which is not available for native Windows.

Users who want to use local serial ports or VXI-11 on Windows can use the Cygwin EPICS build (cygwin-x86).

devGpib: devGpibConvertExample

An example of how to implement convert routines for devGpib support modules is available in asyn/devGpib/devGpibConvertExample.c

devAsynOctet.c

The UDF field is now set FALSE when the VAL field is updated.

asynManager

• lock/unlock renamed to blockProcessCallback/unblockProcessCallback

  The names have been changed and now these methods only work for asynchronous ports. An error is returned if blockProcessCallback is called for a synchronous port.

• lockPort/unlockPort

  These are new asynManager methods. They can be used in place of queueRequest if the caller can block. They have been added to make it easier to implement a driver with one addressing scheme that is a asynUser of a driver with a different addressing scheme. For example a multi-drop serial driver can be implemented that calls a standard serial driver.

• asynLockPortNotify

  This is a new interface for driver's that call other drivers.

cancelInterruptUser

The cancelInterruptUser methods of all interfaces has been changed from

    asynStatus (*cancelInterruptUser)(void *registrarPvt, asynUser *pasynUser);

to

    asynStatus (*cancelInterruptUser)(void *drvPvt, asynUser *pasynUser,
                                      void *registrarPvt);

asynTrace

The length and size arguments now have type size_t.

devGpib

Several improvements were made to devSupportGpib.c. All changes should be transparent to code that uses devGpib.

asynOctetBase

The maxchars argument to callInterruptUsers has been removed.

asynReport

The filename argument has been removed.

devAsynFloat64

For asynAoFloat64 it now uses oval instead of val.

asynFloat64Array, asynInt32Array, asynOctet, asynOctetSyncIO

All length and size arguments now have type size_t.

asynFloat64SyncIO, asynInt32SyncIO, asynOctetSyncIO, and asynUInt32DigitalSyncIO

These all use lockPort/unlockPort instead of queueRequest.

devAsynFloat64

Device support was not returning 2 (do not convert) for ai records when it should. This meant that the VAL field was being set back to 0 by the record after device support wrote to it. This bug is fixed.

asynRecord

The record sometimes did not read the current input and output EOS values from the driver when it connected. This bug is fixed.

Acknowledgement

Yevgeny A. Gusev has again reported some hard to recognize bugs. He must have spent many hours looking at the code. His extra set of very good eyes are much appreciated!!. He also thought of the way to handle support that uses one addressing scheme but wants to use support that has a different addressing scheme. For example support for mult-drop serial that wants to use the standard serial support

asynInterposeEos

If read reads maxchars, it forced the last character to be 0 and returned asynOverflow if it wasn't. This is fixed.

drvAsynSerialPort,drvAsynIPPort - Error reporting

These did not properly set an error message in asynUser.errorMessage when they returned asynError. This is fixed.

drvAsynSerialPort - serial port options

Changes were made to the way serial port options are handled.

• initial values

  Previosly defaults were assigned for all options. Now the initial values are fetched from either the termios (POSIX) or sioLib (vxWorks).

• vxWorks clocal and crtscts

  The vxWorks sioLib uses clocal for what POSIX calls crtscts. The new serial support for vxWorks accepts both clocal and crtscts to specify RTSCTS (Request to send, Clear to send).

asynRecord - Serial Port Options

This has a new option to set Modem Control.

The only code change was to fix the drvAsynIPPort and drvAsynSerialPort segmentation faults on cygwin-x86.

Incompatible Changes.

APOLOGY: Many interfaces have changed since release 3-3. This is the reason this release is called 4-0.

• asynManager - Only report has changed. Many new methods have been added. asynManager:report takes an additional parameter, portName.
• asynOctet - The read/write methods are similar to before. The new version has sepearate end of string methods for input and output.
• register based interfaces - Extensive changes have occured.
• devGpib - Unless special conversion routines call low level drivers, debGpib support should work without any changes.
• devEpics -
  • Extensive changes have occured. The support naming now follows the names of asyn interfaces. For example devAsynInt32 contains device support for interface asynInt32.
  • device definitions have changed.

    Blanks no longer appear in the menu choices. For example.

        device(ai,INST_IO,devAiAsynInt32,"asyn Int32")
           

    is now

        device(ai,INST_IO,asynAiInt32,"asynInt32")
           

• asynOctetSyncIO
  • All read and write methods now return asynStatus and have additional args to return the number of bytes sent/received.
• asynRecord
  • The IEOS and OEOS fields are set to the current values for the port when the record connects to the port. If they are modified after the record connects to the port, then the EOS strings will be changed using asynOctet->setOutputEos or asynOctet->setIbputEos. IMPORTANT: The values of IEOS and OEOS in the database file are never used, because they are modified when the record connects to the port.

New Features

• interrupt support -A major new feature is support for interrupts. See asynDriver.html for details.
• linuxGpib - This is support for the linuxGpib open source project. It contains support for many linux gpib kernel drivers. The asyn support was provided by Rok Sabjan (cosyLab).
• cancelRequest - If cancelRequest is called while either the process or timeout callback is active, it now waits until the callback completes.
• asynRecord
  • Added PCNCT field to connect/disconnect from port, and to indicate if port is currently connected.
  • Added DRVINFO and REASON fields to provide control for the drvUser interface.
  • Added support for register interfaces (asynInt32, asynUInt32Digital, and asynFloat64). New I/O fields for this support are I32INP, I32OUT, UI32INP, UI32OUT, UI32MASK, F64INP, and F64OUT. The new IFACE field is used to select the currently active interface.
  • Added new fields to indicate if a particular interface is supported by the driver. These fields are OCTETIV, I32IV, UI32IV, F64IV, OPTIONIV, and GPIBIV.
  • Added support for I/O Intr scanning for any driver/interface that supports callbacks. asynOctet does not yet support callbacks.

Incompatible Changes.

• MAY BE MORE - This release has major new features. Implementing the new features may have caused some imcompatibilities, This list is likely to grow as existing users report problems.
• queueRequest - If the portDriver does not block then the queue callback is called by queueRequest rather than by a separate thread. User code can call canBlock to find out how the callback is called.
• registerPort - The argument multiDevice has been replaced by attributes. Attributes currently has two bits ASYN_MULTIDEVICE and ASYN_CANBLOCK. The port driver is responsible for setting both bits correctly.
• setOption/getOption have been moved from asynCommon to a new interface asynOption.

Major New Features

• Support for synchronous drivers.
• Support for register based drivers.
  • Generic register based device support for EPICS records.
  • Additional fields have been added to asynUser.
  • Added pasynManager->memMalloc() and pasynManager->memFree() for allocating and freeing memory with a freelist. This is primarily meant to be used with pasynManager->duplicateAsynUser() and the new pasynUser->userData field.

asynDriver.h

The following changes have been made

• userData - this is a new field in asynUser.
• registerPort - Field multiDevice is replaced by attributes. Currently two attributes are defined: ASYN_MULTIDEVICE and ASYN_CANBLOCK.
• duplicateAsynUser - This is a new method that creates a new asynUser and initializes it with values from an existing asynUser. freeAsynUser now puts the asynUser of a free list rather than calling free.
• memMalloc/memFree - New methods that manage storage. It uses a set of free lists of different sizes.
• asynCommon no longer has methods setOption/getOption.
• asynOctet is now defined in a separate header file.

asynRecord

Removed the GOPT field. This is no longer necessary because the port options are automatically read whenever connecting to a port. "special" requests are now queued without changing the state of the record, using the new duplicateAsynUser, memMalloc(), and memFree() methods. This means that there is no longer a possibility of a special request being rejected because the record is busy. It is no longer possible to cancel a special request.

asynGpib

• asynGpibPort:srqStatus returns asynStatus
• asynGpibPort:serialPoll returns asynStatus. It now only calls a registered callback only if statusByte&0x40 is non zero.

devSupportGpib

• setEos is now a method.
• completeProcess is a new method. This was added to support synchronous drivers.
• Failures for GPIBREADW|GPIBEFASTIW were not handled properly. This could cause and assert failure. This is fixed.

drvAsynSerialPortFlush

Flushes input only.

asynInterposeEOS

The EOS read method now calls the low-level read method only once and returns as many characters as the low-level method supplies. This makes the EOS read semantics match those of the low-level serial and IP drivers.

drvVxi11

vxi11SetRpcTimeout - now handles fractions of a second properly

asynRecord

A new field has beem added, AQR (Abort Queue Request)

The semantics have been changed as follows: process is responsible for all and only for I/O operations. Only I/O operations cause the alarm status and severity to change. Special is responsible for all other operations performed by asynRecord.

Changed and obsolete features

• INCOMPATIBLE CHANGE -- The units of the respond2Writes field, if greater than 0, are now seconds rather than milliseconds. This change was made so that all time values set in the instrument support initialization routine are specified uniformly in units of seconds. Very few instrument support files are likely to be affected by this change
• The contents of asynRecDevDrv.dbd have been placed in asyn.dbd and asynRecDevDrv.dbd has been removed. This allows applications to get correct dbior reports and access to asynRecords by including any low-level driver .dbd file.
• The drvAsynTCPPort driver has been renamed drvAsynIPPort since it now supports both UDP and TCP protocols. The protocol is selected by adding a "UDP" or "TCP" after the "hostname:port" in the drvAsynIPPortConfigure command. A missing protocol is taken to be "TCP".
• Work around 'missing SPD' bug in HP E2050 GPIB/LAN adapter. SRQ handling is much more robust on all supported hardware.

Major New Features

• National Instruments NI1014 VME GPIB interface is now supported.
• GPIB - All low level GPIB support (vxi11, gsip488, and ni1014) now fully support the GPIB specific features defined by asynGpibDriver.h
• Implementation of asynTrace is more consistent across the low level drivers.
• Added makeSupport script and template instrument support. Updated tutorials to reflect these additions.

Changed and obsolete features

• The drvGenericSerial driver has been split into drvAsynSerialPort and drvAsynTCPPort drivers for local and remote serial ports, respectively. End-of-string processing has been moved to an interposed interface.
• The diagnostic facilities previously provided by asynTrace.db have been replaced with the much more general asynRecord.db
• All asynManager,asynCommon, and asynOctet methods except report now return asynStatus. Methods that previously returned a different value now have an additional argument for this value.
• Low-level driver read and write methods now return asynStatus and are passed an additional argument through which they store the actual number of characters read or written.
• The createSocket method in the asynSyncIO interface has been replaced by openSocket. openSocket does not call asynSyncIO->connect(), that must now be done by the caller.
• Removed code for "flush" from gpib drivers. The implementation caused infinite loops on devices that speak when not spoken to.
• asynRecord
  • asynOctetRecord has been renamed to asynRecord
  • TIOM, TMSK, TSIZ, TFIL, AUCT, ENBL, CNCT, ERRS, TINP, NAWT fields have been added
  • Default values of OMAX and IMAX decreased from 512 to 80
  • Options that are unknown for a device (e.g. baud on a Moxa terminal server) are shown as Unknown.
  • Many bug fixes and improvements in logic and functionality

Major New Features

• Connection Management - A framework for connection management is provided. It provides the ability to connect/disconnect to/from a port or port,addr. It also provides enable/disable and autoConnect/noAutoConnect. See the asynDriver for details.
• devAsyn - Generic device support for connect management for a specific device.
• devAsynGeneric - Generic support for connection management and traceing. This support dynamically attaches to a port,addr. Thus one set of records and one medm display can be used for all devices connected to an ioc.
• asynOctetRecord - A generic record and medm display that allows interactive access to many asynDriver features.
• asynSyncIO - A synchronous interface to asynDriver. This can be used by code, e.g. SNC programs, that are willing to wait instead of following an asynchronous model.

Changed and obsolete features

• devAsynTrace is replaced by devAsyn and devAsynGeneric.
• asynManager.h
  • disconnectDevice has been renamed to disconnect.
  • The interface to low level drivers has been drastically modified in order to support the new connection management features. See the asynDriver documentation for details.
• asynGpib
  • registerPort has additional arguments multiDevice and autoConnect.
  • setPortOption,getPortOption are setOption,getOption

• Promote VXI-11 RPC definition files to vxi11 directory. Use rpcgen to build RPC support files for targets for which this is possible.
• Run rpcgen on Solaris with 'multi-thread' flag.
• Use separate GPIB message/response buffer for each port instance.
• Use sscanf to convert GPIB stringin response.
• Fix race condition in GPIB finish routines.

This release includes support for the following:

• asynTrace - A trace facility is now implemented.
• gsIP488 - The Greensprings Industry Pack IP488 is now supported

Modifications include:

1. Added asynSetPortOption and asynGetPortOption to manipulate port options.
2. Changed serial support to use asynSetPortOption/asynGetPortOption.
3. Added devGPIB GPIBCVTIO commmand type to allow custom conversion routine to perform all I/O operations.
4. Changed rules for return value from devGPIB custom conversion routines.
5. Added dbior support.
6. Changed devGPIB to no longer cache EOS.

Support Provided in addition to asynDriver and asynGpib

• devGpib - The Winans/Franksen gpib device support.
• vxi11 - Support for instruments that support the VXI-11 standard.
• drvGenericSerial - Support for devices connected to serial ports or to Ethernet/Serial converter.

Future Support

• Other device support methods especially streams, devSerial, and mpfSerial.
• NI1014 VME Gpib driver.
• Industry Pack IP488 Gpib driver.
• Successor to GI (GPIB Interact).

Testing

The vxi11 support has been tested on the following platforms: Solaris, Linux (redhat 9), Darwin, Windows XP (Cygwin), and vxWorks. It has been tested with the following vxi11 controllers:

• Agilent E2050A LAN/GPIB Gateway.

  It's vxiName must start with "hpib" rather than "gpib".

• Agilent E5810A LAN/GPIB Gateway.
• Tektronix TDS3014B Scope.

  When communicating with the Ethernet port it acts like a VXI-11.2 rather than a VXI-11.3 device. It seems to just accept any GPIB address. SRQs did not work when connecting via the ethernet port but did when communicating via a LAN/GPIB gateway.

The generic serial support has been tested with the following:

• xvWorks with a GreenSprings Octal UART Industry-Pack module on a VME carrier.
• Linux and Windows XP (Cygwin) with PC hardware serial port (/dev/ttyS0).
• Solaris hardware serial port (/dev/cua/a).
• Linux, Solaris, Darwin, vxWorks, and Windows XP (Cygwin) with a Moxa NPort Ethernet/Serial converter.

Two Device Support modules have been converted from the 3.13 gpib support: DG535 and TDS3014B Scope.