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

changes for R3.13.5

Browse Source
This commit is contained in:
Marty Kraimer
2001-09-18 15:49:53 +00:00
parent b1725b380b
commit d56ca7f02e
1 changed files with 84 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

116
RELEASE_NOTES.html
View File

@@ -9,14 +9,52 @@
<body lang="en">
<h1>EPICS Release 3.13.5</h1>
<h4>June 2001</h4>
<h4>Sept 2001</h4>
<h3>
Changes since 3.13.4</h3>
<p>
<b>convertRelease and checkRelease</b>
<p>
New applications created with makeBaseApp.pl now have the functionality of
the makeConfigAppInclude.pl and makeIocCdCommands.pl scripts combined into
a single convertRelease.pl script.
<p>
The new script also provides consistency checking of the entries in the
config/RELEASE file, so that gnumake will stop with an error report if a
definition in any other &lt;supporttop&gt/config/RELEASE file is different to
the same definition in this application. This checking can be disabled by
editing config/Makefile.Host.
<p>
Applications created using earlier versions of EPICS will continue to work
as before and will not benefit from the consistency check functionality.
<p>
<b>devAiSoftRaw and devAoSoftRaw</b>
<p>
ai and ao records using DTYP="Raw Soft Channel" can have their EOFF
and ESLO fields set manually at design or run-time, and these values will
be used when LINR=LINEAR. The same applies to any device support that
doesn't provide a special_linconv() routine; if special_linconv() is
provided however, the original behaviour occurs (EOFF is set to EGUL and
ESLO is calculated by that routine). The ESLO and EOFF fields now have CA
monitors posted if their value is changed by the special_linconv()
routine.
<p>
<b>cac_recv_task</b>
<p>
Increased stack size from 4096 to 819l.
<p>
<b>drvTS</b>
<p>
Now uses osiSock for to handle implementation diffences
<p>
<b>Constant Links</b>
<p>
Hex and octal constants are now accepted for constant links.
recGblInitConstantLink uses %i instead of %u for unsigned types.
This is so that hex constants are accepted. %u does not accept hex.
dbPutString makes a link constant if either strtod or strtol eats entire string.
Previously only strtod was used.
<p>
<b>Breakpoint tables</b>
<p>
@@ -338,48 +376,62 @@ involved the year 00 appeared in the ascii string</p>
<p><b>Channel Access</b></p>
<p>The CA comments are from Jeff Hill.</p>
<p>
1) 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 was fixed.
This problem was discovered during load testing, and will be more likely to occur under heavy
network loading conditions.
</p>
<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>
<p>
2) 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 was fixed.
</p>
<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>
<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>
<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>
<p>
4) A bug fix was made for a 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>
<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>
<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>
<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>
<p>
6) always inform client of "no write access" for a SPC_NOMOD fields.
</p>
<p>6)  always inform client of "no write access"  for a SPC_NOMOD fields.</p>
<p>
7) calls to ca_printf removed from rsrv.
</p>
<p>7) calls to ca_printf removed from rsrv.</p>
<p>
8) prevent UDP daemon from running damaged messages with TCP protocol and
visa-versa
</p>
<p>8) prevent UDP daemon from running damaged messages with TCP protocol and
visa-versa</p>
<p>
9) use ipAddrToA() so that address print out does not come out byte swapped on the PC
</p>
<p>9) use ipAddrToA() so that address print out does not come out byte swapped
on the PC</p>
<p>
10) improved debug messages and added diagnostics
</p>
<p>10) improved debug messages and added diagnostics</p>
<p>
11) don't send an exception to client when bad request received over UDP.
</p>
<p>11)  dont send an exception to client when bad request received over udp.
<br>
 </p>