Features
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
22.8 SPECint95 (estimated), 17SPECfp95 at 500 MHz (estimated)
917MIPS at 500 MHz
Selectable Bus Clock (14 CPU Bus Dividers Up To 9x)
Seven Selectable Core-to-L2 Frequency Divisors
Selectable 603 Interface Voltage Below 3.3V (1.8V, 2.5V)
Selectable L2 interface of 1.8V or 2.5V
P
D
Typical 5.3W at 500 MHz, Full Operating Conditions
Nap, Doze and Sleep Modes for Power Saving
Superscalar (Four Instructions fetched per Clock Cycle)
4 GB Direct Addressing Range
Virtual Memory: 4 hexabytes (2
52
)
64-bit Data and 32-bit Address Bus Interface
32 KB Instruction and Data Cache
Eight Independent Execution Units and Three Register Files
Write-back and Write-through Operations
f
INT
Max = 450 MHz 500 MHz
f
BUS
Max = 133 MHz
Description
The PC7410 is the second microprocessor that uses the fourth (G4) full implementa-
tion of the PowerPC
™
Reduced Instruction Set Computer (RISC) architecture. It is
fully JTAG-compliant.
The PC7410 maintains some of the characteristics of G3 microprocessors:
•
•
•
The design is superscalar, capable of issuing three instructions per clock cycle
into eight independent execution units
The microprocessor provides four software controllable power-saving modes and
a thermal assist unit management
The microprocessor has separate 32-Kbyte, physically-addressed instruction and
data caches with dedicated L2 cache interface with on-chip L2 tags
PowerPC 7410
RISC
Microprocessor
Product
Specification
PC7410
In addition, the PC7410 integrates full hardware-based multiprocessing capability,
including a 5-state cache coherency protocol (4 MESI states plus a fifth state for
shared intervention) and an implementation of the new AltiVec
™
technology instruc-
tion set.
New features have been developed to make latency equal for double-precision and
single-precision floating-point operations involving multiplication. Additionally, in mem-
ory subsystem (MSS) bandwidth, the PC7410 offers an optional, high-bandwidth MPX
bus interface.
Unlike the PC7400, the PC7410 does not support the 3.3V I/O on the L2 cache
interface.
Rev. 2141D–HIREL–02/04
Screening
•
•
•
CBGA Upscreenings Based on Atmel Standards
Full Military Temperature Range (T
j
= -55°C, +125°C),
Industrial Temperature Range (T
j
= -40°C, +110°C)
CI-CGA Package Version, HiTCE Package Version
G suffix
CBGA 360
Ceramic Ball Grid Array
GS suffix
CI±CGA 360
Ceramic Ball Grid Array
with Solder Column Interposer (SCI)
GH suffix
HITCE 360
Ceramic Ball Grid Array
2
PC7410
2141D–HIREL–02/04
2141D–HIREL–02/04
Figure 1.
PC7410 Microprocessor Block Diagram
Block Diagram
128 bits
(4 instructions)
Instruction MMU
Instruction Unit
Additional features
Time Base
Counter/Decrementer
Clock Multiplier
JTAG/COP Interface
Thermal/Power Management
Performance Monitor
2 Instructions
SRs
(Shadow)
128-entry
ITLB
Tags
IBAT
Array
32-Kbyte
iCache
Fetcher
Branch Processing Unit
64-entry BTIC/512-entry BHT
LR/CTR
Instruction
Queue
6-word
Data MMU
Dispatch Unit
EA
SRs
(Original)
PA
128-entry
DTLB
DBAT
Array
Tags
32-Kbyte
DCache
64-bit
(2 Instructions)
Reservation
Station
Reservation
Station
Reservation
Station
VR File
6 Rename
Buffers
Reservation
Station
Reservation
Station
GPR File
6 Rename
Buffers
32-bit
Reservation
Station
2-entry
FPR File
6 Rename
Buffers
Reservation
Station
Load/Store
Unit
- Add -
EA Calculation
Finished Stores
Completed
Stores
Vector
Permute
Unit
Vector
ALU
VSIU VCIU VFPU
Integer
Unit 1
Add-Multiply-
divide
Integer
Unit 2
- Add -
System
Register
Unit
64-bit
Floating
64-bit
Point Unit
Add-Multiply-
divide
FPSCR
32-bit
128-bit
32-bit
VSCR
128-bit
128 bits
Completion Unit
8-entry
Reorder Buffer
L2 Controller
L2 Data
L2 Tags
Transaction
L2CR
Queue
L2PMCR
L2 Castout
19-bit L2 Address Bus
Bus Interface Unit
L2 Miss
Data
Transaction
Queue
Memory Subsystem
Data Reload Data Reload
Buffer
Table
PC7410
64- or 32-bit L2 Data Bus
32-bit Address Bus
64-bit Data Bus
Instruction
Instruction
Reload Buffer Reload Table
3
General Parameters
Table 1 provides a summary of the general parameters of the PC7410.
Table 1.
Device Parameters
Parameter
Technology
Die size
Transistor count
Logic design
Packages
Description
0.18 µm CMOS, six-layer metal
6.32 mm × 8.26 mm (52 mm
2
)
10.5 million
Fully-static
Surface-mount 360 ceramic ball grid array (CBGA)
Surface mount 360 high coefficient of thermal expansion
ceramic ball grid array (HiTCE)
Surface mount 360-column Ci-CGA Package
1.8V ± 100 mV dc or 1.5V ± 50 mV dc (nominal; see Table 4 for
Recommended Operating Conditions)
1.8V ± 100 mV dc or
2.5V ± 100 mV
3.3V ± 165 mV (603 bus only)
(1)
(input thresholds are configuration pin selectable) or
Core power supply
I/O power supply
Note:
1. 3.3V I/O bus not supported for 1.5V core power supply processor version.
Features
This section summarizes features of the PC7410’s implementation of the PowerPC
architecture. Major features of the PC7410 are as follows:
•
Branch Processing Unit
–
–
–
–
–
•
Four instructions fetched per clock
One branch processed per cycle (plus resolving two speculations)
Up to one speculative stream in execution, one additional speculative stream
in fetch
512-entry branch history table (BHT) for dynamic prediction
64-entry, 4-way set associative branch target instruction cache (BTIC) for
eliminating branch delay slots
Full hardware detection of dependencies (resolved in the execution units)
Dispatch two instructions to eight independent units (system, branch,
load/store, fixed-point unit 1, fixed-point unit 2, floating-point, AltiVec
permute, AltiVec ALU)
Serialization control (predispatch, postdispatch, execution serialization)
Register file access
Forwarding control
Partial instruction decode
8-entry completion buffer
Instruction tracking and peak completion of two instructions per cycle
Dispatch Unit
–
–
–
•
–
–
–
•
–
–
Decode
Completion
4
PC7410
2141D–HIREL–02/04
PC7410
–
Completion of instructions in program order while supporting out-of-order
instruction execution, completion serialization and all instruction flow
changes
Fixed-point unit 1 (FXU1)—multiply, divide, shift, rotate, arithmetic, logical
Fixed-point unit 2 (FXU2)—shift, rotate, arithmetic, logical
Single-cycle arithmetic, shifts, rotates, logical
Multiply and divide support (multi-cycle)
Early out multiply
Support for IEEE-754 standard single- and double-precision floating-point
arithmetic
Three-cycle latency, one-cycle throughput (single or double precision)
Hardware support for divide
Hardware support for denormalized numbers
Time deterministic non-IEEE mode
Executes CR logical instructions and miscellaneous system instructions
Special register transfer instructions
Full 128-bit data paths
Two dispatchable units: vector permute unit and vector ALU unit
Contains its own 32-entry 128-bit vector register file (VRF) with six renames
The vector ALU unit is further sub-divided into the vector simple integer unit
(VSIU), the vector complex integer unit (VCIU) and the vector floating-point
unit (VFPU).
Fully pipelined
One-cycle load or store cache access (byte, half-word, word, double-word)
Two-cycle load latency with one-cycle throughput
Effective address generation
Hits under misses (multiple outstanding misses)
Single-cycle unaligned access within double-word boundary
Alignment, zero padding, sign extend for integer register file
Floating-point internal format conversion (alignment, normalization)
Sequencing for load/store multiples and string operations
Store gathering
Executes the cache and TLB instructions
Big- and little-endian byte addressing supported
Misaligned little-endian supported
Supports FXU, FPU, and AltiVec load/store traffic
Complete support for all four architecture AltiVec DST streams
32K 32-byte line, 8-way set associative instruction cache (iL1)
•
Fixed-point Units (FXUs) that Share 32 GPRs for Integer Operands
–
–
–
–
–
•
Three-stage Floating-point Unit and a 32-entry FPR File
–
–
–
–
–
•
System Unit
–
–
•
AltiVec Unit
–
–
–
–
–
•
–
–
–
–
–
–
–
–
–
–
–
–
–
–
•
–
Load/Store Unit
Level 1 (L1) Cache Structure
5
2141D–HIREL–02/04