Experimental Physics and
Industrial Control System

Jeff Hill <[email protected]> · Mon, 10 Feb 1997 13:03:52 -0700

Hello all,

A bug has been discovered in EPICS 3.12.2.

Conditions (all of the following are true):
-------------------------------------------
o client side program is running on a little endian machine
	with IEEE (not VAX) floating point
	=> PC
	=> dec alpha (running UNIX)
	=> others?
o double precision floating point data has been fetched
	by a CA get call back operation or a CA monitor
	operation

Symptoms:
---------
Least significant part of the double precision operand
is not byte swapped resulting in an incorrect value for
the portion of the data after the 8th digit or so (and
also, under rare circumstances, a floating point 
exception).

This problem does not occur in recent versions of EPICS
3.13 beta.

A patch has been applied to EPICS 3.12.2.

Thanks to Mark Rivers for discovering the problem and
also to Kay-Uwe Kasemir for fixing it.

Jeff

-- 
______________________________________________________________________
Jeffrey O. Hill                 Internet        [email protected]
LANL MS H820                    Voice           505 665 1831
Los Alamos, NM 87545 USA        FAX             505 665 5107

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System