![](http://datasheet.mmic.net.cn/290000/XPC801ZP25_datasheet_16187868/XPC801ZP25_127.png)
PowerPC Architecture Compliance
7-4
MPC801 USER’S MANUAL
MOTOROLA
7
7.1.9.6 LITTLE-ENDIAN BYTE ORDERING
The load/store unit supports little-endian byte ordering as specified in
Instruction Set Architecture (Book I)
individual scalar unaligned transfer, as well as a multiple or string instruction, causes an
alignment interrupt.
PowerPC
User
. In little-endian mode, attempting to execute an
7.2 POWERPC VIRTUAL ENVIRONMENT ARCHITECTURE (BOOK II)
7.2.1 Storage Model
The MPC801 instruction and data caches are defined as follows:
Physically addressed 2K instruction cache
Physically addressed 1K data cache
Two-way set-associative managed with LRU replacement algorithm
16-byte (4 words) line size with one valid bit per line
7.2.1.1 MEMORY COHERENCE
Hardware memory coherence is not supported in the MPC801 hardware, but can be
performed in the software or by defining storage as cache inhibited as needed. In addition,
the MPC801 does not provide any data storage attributes for an external system.
7.2.1.2 ATOMIC UPDATE PRIMITIVES
Both the
lwarx
and
stwcx
architecture requirements. When the storage accessed by the
is in the cache-allowed mode, it is assumed that the system works with the single master in
this storage region. Therefore, if a data cache miss occurs, the access on the internal and
external buses does not have a reservation attribute.
instructions are implemented according to the PowerPC
lwarx
and
stwcx
instructions
The MPC801 does not cause the system data storage error handler to be invoked if the
storage accessed by the
lwarx
and
stwcx
instructions is in the writethrough required mode.
Also, the MPC801 does not provide support for snooping an external bus activity outside of
the chip. The provision is made to cancel the reservation inside the MPC801 by using the
CR and KR input pins.
7.2.2 The Effect of Operand Placement on Performance
The load/store unit hardware supports all of the PowerPC load/store instructions. An optimal
performance can be obtained for naturally aligned operands. These accesses result in
optimal performance (one bus cycle) for a maximum size of 4 bytes and good performance
(two bus cycles) for double precision floating-point operands. Unaligned operands are
supported in the hardware and are broken into a series of aligned transfers. The effect of
operand placement on performance is as stated in
Architecture Book II,
except for the case of 8-byte operands.
PowerPC
Virtual Environment