
RCPU
REFERENCE MANUAL
EXCEPTIONS
For More Information On This Product,
Go to: www.freescale.com
MOTOROLA
6-45
A tiny result is detected before rounding, when a non-zero result value computed
as though the exponent range were unbounded would be less in magnitude than
the smallest normalized number.
If the intermediate result is tiny and the underflow exception condition enable bit is
cleared (FPSCR[UE] = 0), the intermediate result is denormalized.
Loss of accuracy is detected when the delivered result value differs from what
would have been computed were both the exponent range and precision
unbounded.
When an underflow exception occurs, the action to be taken depends on the setting
of the underflow exception condition enable bit of the FPSCR.
When the underflow exception condition is enabled (FPSCR[UE] = 1) and an ex-
ponent underflow condition occurs, the following actions are taken:
Underflow exception condition bit is set: FPSCR[UX] = 1.
For double-precision arithmetic and conversion instructions, the exponent of
the normalized intermediate result is adjusted by adding 1536.
For single-precision arithmetic instructions and the floating round to single-
precision instruction, the exponent of the normalized intermediate result is ad-
justed by adding 192.
The adjusted rounded result is placed into the target FPR.
FPSCR[FPRF] is set to indicate the class and sign of the result (±normalized
number).
The FR and FI bits in the FPSCR allow the system floating-point enabled exception
error handler, when invoked because of an underflow exception condition, to sim-
ulate a trap disabled environment. That is, the FR and FI bits allow the system float-
ing-point enabled exception error handler to unround the result, thus allowing the
result to be denormalized.
When the underflow exception condition is disabled (FPSCR[UE] = 0) and an un-
derflow condition occurs, the following actions are taken:
Underflow exception condition enable bit is set: FPSCR[UX] = 1
The rounded result is placed into the target FPR
FPSCR[FPRF] is set to indicate the class and sign of the result
(±denormalized number or ±zero)
6.11.10.10 Inexact Exception Condition
The inexact exception condition occurs when one of two conditions occur during
rounding:
The rounded result differs from the intermediate result assuming the interme-
diate result exponent range and precision to be unbounded.
The rounded result overflows and overflow exception condition is disabled.
When the inexact exception condition occurs, regardless of the setting of the inex-
act exception condition enable bit of the FPSCR, the following actions are taken:
F
Freescale Semiconductor, Inc.
n
.