epics/epics-base
0
0
Fork 0
Code Issues Pull Requests Projects Wiki Activity

notes for R3.13.2

Browse Source
This commit is contained in:
Marty Kraimer
1999-12-16 14:58:07 +00:00
parent d6ee79d742
commit 3eaee4daca
1 changed files with 135 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

162
RELEASE_NOTES.html
View File

@@ -2,32 +2,78 @@
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<meta name="GENERATOR" content="Mozilla/4.61 [en] (X11; U; SunOS 5.6 sun4u) [Netscape]">
<meta name="GENERATOR" content="Mozilla/4.7 [en] (X11; U; SunOS 5.6 sun4u) [Netscape]">
<title>EPICS Release baseR3.13.0.beta12
</title>
</head>
<body>
<h1>
EPICS Release baseNEXT</h1>
EPICS Release 3.13.2</h1>
<h4>
July 7, 1999</h4>
Dec 16, 1999</h4>
<h3>
Changes since 3.13.1</h3>
<b>config</b>
<b>tsSubr</b>
<p>There was a bug when converting ascii string format to time stamp. The
bug involved the year 00 appeared in the ascii string
<p><b>Channel Access</b>
<p>The CA comments are from Jeff Hill.
<p>1) I fixed a severe problem where the CA client did not exit a loop
sending UDP to a list of addresses if an error condition was returned from
the system call which sends a UDP frame. This problem was discovered by
me during heavy load testing, and if
<br>it occurs in an operational system it will probably be occurring under
heavy network loading conditions.
<p>2) I fixed a problem where the client library could fail and use a deallocated
data structure if a connection dropped while it was waiting to send a message
to a client.
<p>3) The CA client now maintains a congestion avoidance threshold on UDP
search transmissions similar to the congestion avoidance algorithm employed
by modern version of TCP. There are also some small changes related to
how the search counters are maintained without allowing them to overflow.
This patch was made in R3.13.1 because I originally identified it as the
fix for (1) above, but it turned out to not be the root cause.
<p>4) A bug fix was made for Ralph Lange's problem where the CA repeater
was inheriting on HP-UX, and possibly other UNIX systems that are not Solaris,
certain open file descriptors when fork()/exec() are used to start the
CA repeater.
<p>5) Changes that allow CA to work on Microsoft Visual C++ 6.0 (which
includes a new win sock DLL and some changes in esoteric parts of the socket
API).
<p>6)&nbsp; always inform client of "no write access"&nbsp; for a SPC_NOMOD
fields.
<p>7) calls to ca_printf removed from rsrv.
<p>8) prevent UDP daemon from running damaged messages with TCP protocol
and visa-versa
<p>9) use ipAddrToA() so that address print out does not come out byte
swapped on the PC
<p>10) improved debug messages and added diagnostics
<p>11)&nbsp; dont send an exception to client when bad request received
over udp.
<br>&nbsp;
<p><b>config</b>
<ul>
<li>
Many changes.</li>
<li>
Currently gdd and cas are not built. LEFT&nbsp;FOR&nbsp;JEFF&nbsp;TO&nbsp;MAKE&nbsp;WORK</li>
Currently gdd and cas are not built.</li>
</ul>
<p><br><b>dbCommon.dbd</b>
<br><b></b>&nbsp;
<b>access security</b>
<p>A ca exception handler was added. Thus exceptions will no longer cause
the asCa task to suspend.
<p><b>dbCa</b>
<p>A ca exception handler. Thus exceptions no longer cause dbCa task to
suspend.
<p><b>dbScan.c</b>
<p>pevent_scan_list->scan_list.lock was not initialized. Amazingly it did
not cause a crash. Note that the lock is only used when the scan mechanism
of records are dynamically changed.
<p><b>dbCommon.dbd</b>
<br>&nbsp;
<ul>
<li>
several extra definitions were changed from void * to correct struct types.</li>
@@ -35,7 +81,7 @@ several extra definitions were changed from void * to correct struct types.</li>
<li>
ACKT was changed so it can be set by database configuration tools.</li>
</ul>
<b></b>
<p><br><b>dbStaticLib</b>
<ul>
<li>
@@ -52,8 +98,18 @@ CALC fields are OK if either postfix or sCalcPostfix returns success.</li>
<p>Tables names were previously forced to be alphanumeric. This restriction
is removed.
<p><b>drvTS.c</b>
<p>removed potential infinite recursion from tsForceSoftSync()
<p><b>drvSup.h</b>
<br>&nbsp;
<ul>
<li>
Removed potential infinite recursion from tsForceSoftSync().</li>
<li>
Fixed bug where time sync UDP client was getting in a state where it was
using the response from the previous request, and ending up setting the
time 10 seconds back.</li>
</ul>
<p><br><b>drvSup.h</b>
<p>removed definition for&nbsp; reboot.
<p><b>dbNotify</b>
<p>If&nbsp; dbPutNotify called dbProcess and dbProcess returned an error
@@ -62,9 +118,28 @@ the notify completion callback was called twice. This bug is fixed.
<p>This support was defining storage twice. A bug in the current gcc compiler
supplied with tornado does not properl;y compile such code. The extra definitions
are removed.
<p><b>aiRecord and aoRecord</b>
<p>eoff was not being reinitialized when conversion related fields were
dynamically modified.
<p><b>aiRecord</b>
<ul>
<li>
eoff was not being reinitialized when conversion related fields were dynamically
modified.</li>
<li>
hyst and lalm are changed from float to double</li>
</ul>
<b>aoRecord</b>
<ul>
<li>
eoff was not being reinitialized when conversion related fields were dynamically
modified.</li>
<li>
hyst, lalm, hihi, high, low, lolo all changed from float to double</li>
</ul>
<b>calcRecord&nbsp; calcoutRecord selRecord subRecord</b>
<p>hyst and lalm changed from float to double
<p><b>dfanoutRecord</b>
<p>hyst, lalm, hihi, high, low, lolo all changed from float to double
<p><b>compressRecord</b>
<p>compressALG_N_to_1_Median (compress keeping median value) added.
<p><b>mbbiDirectRecord</b>
@@ -73,7 +148,18 @@ were posted, no monitors on the single bit fields.
<p><b>devApsEr.c</b>
<p>ErGetTicks now locks while updating ticks
<p><b>devCommonGpib</b>
<p>sscanf response now checked to be equal to 1 rather than just not zero.
<br>&nbsp;
<ul>
<li>
sscanf response now checked to be equal to 1 rather than just not zero.</li>
<li>
all mallocs changed to callocs. This is needed for the new unbundled bitbus
support.</li>
</ul>
<b>drvGpib.c</b>
<p>all mallocs changed to callocs. This is needed for the new unbundled
bitbus support.
<p><b>bitbus support</b>
<p>This support is now unbundled. DO NOT use the version in base. This
includes the following
@@ -117,11 +203,23 @@ drvMsg.h</li>
<li>
drvRs232.h</li>
</ul>
NOTE: drvBitBusInterface.h and drvBitBusErr.h appear in both base and unbundled
because drvGib uses it.&nbsp; Must be fixed when gpib is unbundled.
NOTE: drvBitBus.h, drvBitBus.c, drvBitBusInterface.h, and drvBitBusErr.h
appear in both base and unbundled because drvGib uses it.&nbsp; Must be
fixed when gpib is unbundled.
<p><b>drvBBMsg.c drvBBMsg.c drvRs232.c drvTy232.h</b>
<p>Removedfrom base. Seemed to be old duplicate version of drvBB232.c or
not used at all.
<p><b>devXxDig500Msg232.c</b>
<p>removed, was replaced by the digitel record type &amp; device support
<p><b>CAN and IPAC</b>
<p>This support has been available for some time as an unbundled product.
It is not longer included with base.
<p><b>Allen Bradley</b>
<p>The allen bradley support has been available for some time as an unbundled
product. It is not longer included with base.
<p><b>devOpt support</b>
<p>The following gpib support modules have been changed to fix compiler
bug fot vxWorks 5.3:
bug for vxWorks 5.3:
<ul>
<li>
devAnalytekGpib.c</li>
@@ -156,6 +254,10 @@ devXxSr620Gpib.c</li>
<p><b>makeBaseApp</b>
<ul>
<li>
makeBaseApp.pl</li>
<br>&nbsp;copy any non-Apps under $top; allow links for apptype directories
<li>
dbExample2.template</li>
<br>fixed mistake
@@ -163,9 +265,17 @@ dbExample2.template</li>
makeIocCdCommands.pl</li>
<br>This now looks for all external products in config/RELEASE and generates
commands
<br>to cd to product top and product bin.</ul>
command to cd to product top and product bin.</ul>
<b>calcoutRecord</b>
<p>Under certain conditions (an input link referencing a PV on another
ioc and an unusually long ioc initialization), the calcout record would
cause an&nbsp; error during iocInit by attempting to dbScanLock a record
prematurely. This anomaly has been resolved with this version. No changes
to the functionality of the calcout record were made.
<p><b>dbLoadTemplate</b>
<p>Allow pattern with no substitutions.
<p><b>devAaoCamac.c</b>
<p>This had an error that prevented it from compiling.
<h3>
Changes between beta12 and 3.13.1</h3>
<b><tt>iocLogClient</tt></b>
@@ -414,9 +524,8 @@ in 2036. However it was modified so that no overflows should occur while
converting between NTP, UNIX, and EPICS epochs. In addition the conversion
of fractions of a second to nanaoseconds was changed and formatting was
changed so that interesting code does not run off right side of page. Hopefully
EPICS
base is now fine for Y2K. In fact it should be fine (as far as time is
converned) until the Unix epoch overflows a 32 unsigned integer in the
EPICS base is now fine for Y2K. In fact it should be fine (as far as time
is converned) until the Unix epoch overflows a 32 unsigned integer in the
year 2106.
<br>&nbsp;
<h3>
@@ -1427,9 +1536,8 @@ devLoSoft, devLoSymb, devMbboDirectSoft, devMbboSoft, devPtSoft, devSoSymb
<li>
devHistogramSoft, devHistogramTestAsyn - In init_record the udf field is
set
to false only if svl is a constant link and recGblInitConstantLink returns
success.</li>
set to false only if svl is a constant link and recGblInitConstantLink
returns success.</li>
</ul>
</ul>