MPC555
/
MPC556
BURST BUFFER
MOTOROLA
USER’S MANUAL
Rev. 15 October 2000
4-3
— Minimal performance penalty due to change of program flow execution
• Two operation modes are available: “Decompression ON” and “Decompression
OFF”. Switch between compressed and non-compressed user application soft-
ware parts is possible.
The IMPU has the following features:
• Four regions in which the base address and size can be programmed.
• Region sizes of four Kbytes up to four Gbytes (in powers of two) can be pro-
grammed. (A region must start on the specified region size boundary.)
• Overlap between regions is allowed.
• Each of the four regions supports the following attributes:
— Access protection (use r/ supervisor fetch or no access).
— Guarded attribute (causes an interrupt in case of fetch try).
— On / off option
— Compressed / non-compressed.
• Global region entry declares the default access protection and guarded attributes
for all memory areas not covered by the four regions:
• Interrupt generated upon access violation or fetch from guarded region.
• MPC555 / MPC556 MSR[IR] bit controls MPU protection.
• Programming is done using MPC555 / MPC556
mtspr/mfspr
instructions to/from
implementation specific special purpose registers.
• Designed for minimum power consumption.
• Compressed/non-compressed region with enable/disable option.
• Special reset exception vector for “Decompression ON” mode.
4.3 Instruction VocabularyBased Compression Model Main Principles
4.3.1 Compression Model Features
• Implemented for PowerPC architecture
• Up to 30% code size reduction
• No need for address translation tables
• No changes in the CPU architecture
• Compression is done off line by a special “compressor” tool, using a fixed vocab-
ulary instruction based algorithm, optimized for the PowerPC instruction set.
• Decompression is done at run-time with special hardware.
• Optimized for cache-less systems:
— Highly effective in system solutions for low cache-hit ratio environment and for
systems with fast embedded program memory
— Deterministic program execution time
— No performance penalty during sequential program flow execution
— Minimal performance penalty due to change of program flow execution
• Switch between compressed and non-compressed user application sections is
possible. (Compressed subroutine can call non-compressed one and be called
from non-compressed portion of user application)
F
re
e
sc
a
le
S
e
m
ic
o
n
d
u
c
to
r,
I
Freescale Semiconductor, Inc.
For More Information On This Product,
Go to: www.freescale.com
n
c
.
..