
RCPU
REFERENCE MANUAL
EXCEPTIONS
For More Information On This Product,
Go to: www.freescale.com
MOTOROLA
6-39
FPSCR are indicated in parentheses.
Invalid floating-point operation exception condition (VX)
— SNaN condition (VXSNAN)
— Infinity–infinity condition (VXISI)
— Infinity/infinity condition (VXIDI)
— Zero/zero condition (VXZDZ)
— Infinity*zero condition (VXIMZ)
— Illegal compare condition (VXVC)
These exception conditions are described in
6.11.10.6 Invalid Operation Ex-
ception Conditions
.
Software request condition (VXSOFT). These exception conditions are de-
scribed in
6.11.10.6 Invalid Operation Exception Conditions
.
Illegal integer convert condition (VXCVI). These exception conditions are de-
scribed in
6.11.10.6 Invalid Operation Exception Conditions
.
Zero divide exception condition (ZX). These exception conditions are de-
scribed in
6.11.10.7 Zero Divide Exception Condition
.
Overflow Exception Condition (OX). These exception conditions are described
in
6.11.10.8 Overflow Exception Condition
.
Underflow Exception Condition (UX). These exception conditions are de-
scribed in
6.11.10.9 Underflow Exception Condition
.
Inexact Exception Condition (XX). These exception conditions are described
in
6.11.10.10 Inexact Exception Condition
.
Each floating-point exception condition and each category of illegal floating-point
operation exception condition have a corresponding exception bit in the FPSCR. In
addition, each floating-point exception has a corresponding enable bit in the
FPSCR. The exception bit indicates the occurrence of the corresponding condition.
If a floating-point exception occurs, the corresponding enable bit governs the result
produced by the instruction and, in conjunction with bits FE0 and FE1, whether and
how the system floating-point enabled exception error handler is invoked. (The “en-
abling” specified by the enable bit is of invoking the system error handler, not of
permitting the exception condition to occur. The occurrence of an exception condi-
tion depends only on the instruction and its inputs, not on the setting of any control
bits.)
The floating-point exception summary bit (FX) in the FPSCR is set when any of the
exception condition bits transitions from a zero to a one or when explicitly set by
software. The floating-point enabled exception summary bit (FEX) in the FPSCR is
set when any of the exception condition bits is set and the exception is enabled (en-
able bit is one).
A single instruction may set more than one exception condition bit in the following
cases:
The inexact exception condition bit may be set with overflow exception condi-
tion.
The inexact exception condition bit may be set with underflow exception con-
dition.
The illegal floating-point operation exception condition bit (SNaN) may be set
F
Freescale Semiconductor, Inc.
n
.