FUJITSU MB86330CR-ES

FUJITSU SEMICONDUCTOR
DATA SHEET
DS07-03108-2E
Processor Digital Signal Processor
CMOS
16-bit Fixed-point DSP
MB86330
■ DESCRIPTION
The MB86330 is a 16-bit fixed-point DSP (Digital Signal Processor) that is based on Fujitsu-specific Dual-MAC
architecture, and can implement product addition operations and double transfer at a high rate and under low
power consumption.
The DSP supports a set of instructions optimum for digital signal processing in communications applications such
as handy phones.
The MB86330 consists of a core section and a peripheral section. For detailed specifications of the core section,
see MB86330DSP Core Section Specifications.
■ FEATURES
• Fixed-point operations
Multiplication: 16 bits × 16 bits → 31 bits
Addition: 40 bits + 40 bits → 40 bits
Product addition: 40 bits ±16 bits × 16 bits → 40 bits
Maximum operation speed: 100 MIPS at 3.3 V
• Memory configuration
Data RAM: Two sectors that can be accessed concurrently
An external RAM (ERAM) is supported.
Memory mapped I/O system characterized by allocation of I/O devices in the memory space
Instruction RAM: 48 Kwords × 16 bits
Table RAM: 16 Kwords × 16 bits
(Continued)
■ PACKAGE
256-pin Ceramic PGA
(PGA-256C-A03)
MB86330
(Continued)
• Addressing
Two independent address units
Eight general-purpose registers
Addressing function that can update a register
Circular addressing
Two address update registers
• Supply voltage: 3.3 V (single type of supply voltage)
• Ceramic package: PGA-256
2
MB86330
■ PIN ASSIGNMENT
(Top view)
55
54
53
52
51
50
49
48
47
46
45
44
43
42
41
40
39
38
37
56
121
120
119
118
117
116
115
114
113
112
111
110
109
108
107
106
105
36
57
122
179
178
177
176
175
174
173
172
171
170
169
168
167
166
165
104
35
58
123
180
229
228
227
226
225
224
223
222
221
220
219
218
217
164
103
34
59
124
181
230
216
163
102
33
60
125
182
231
215
162
101
32
61
126
183
232
214
161
100
31
62
127
184
233
213
160
99
30
63
128
185
234
212
159
98
29
64
129
186
235
211
158
97
28
65
130
187
236
210
157
96
27
66
131
188
237
209
156
95
26
67
132
189
238
208
155
94
25
68
133
190
239
207
154
93
24
69
134
191
240
206
153
92
23
70
135
192
193
194
195
196
197
198
199
200
201
202
203
204
205
152
91
22
71
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
90
21
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
20
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
252
251
250
249
253
248
254
247
255
246
256
245
Extra index pin
241
242
243
244
(PGA-256C-A03)
3
MB86330
Pin
no.
I/O
Pin
name
Pin
no.
I/O
Pin
name
Pin
no.
I/O
Pin
name
Pin
no.
I/O
Pin
name
1
O
PAO27
33
I/O
ED13
65
O
PAO7
97
I/O
EDI
2
I
ICAD0
34
—
N.C.
66
O
PAO10
98
I/O
ED3
3
O
PAO30
35
I/O
ICDT12
67
O
ST2
99
—
N.C.
4
O
IRO0
36
I/O
ICDT10
68
O
PAO13
100
I/O
ED10
5
O
IRO3
37
I/O
ICDT6
69
O
PAO15
101
I/O
ED12
6
I
ICAD1
38
—
N.C.
70
O
FF
102
I/O
ED15
7
O
IRO8
39
I/O
ICDT3
71
O
PAO21
103
I/O
ICDT13
8
O
IRO10
40
—
N.C.
72
O
PAO23
104
I/O
ICDT11
9
O
IRO14
41
O
AINT6
73
O
PAO24
105
I/O
ICDT9
10
I
ICAD3
42
O
AINT5
74
O
PAO28
106
I/O
ICDT5
11
O
IRO17
43
I
INT4
75
O
PAO31
107
I/O
ICDT2
12
O
IRO20
44
I
SCZC
76
O
IRO1
108
O
AINT7
13
O
IRO23
45
—
N.C.
77
O
IRO4
109
I
INT6
14
O
IRO26
46
—
N.C.
78
O
IRO6
110
I
INT5
15
O
IRO28
47
—
N.C.
79
I
ICAD2
111
I
INT3
16
O
IRO30
48
—
N.C.
80
O
IRO13
112
—
N.C.
17
—
N.C.
49
I
F0
81
O
IRO16
113
—
N.C.
18
—
N.C.
50
I
MOD0
82
O
IRO18
114
I
MCLK
19
O
PAGE1
51
O
AINT2
83
O
IRO21
115
I
BREAK
20
O
XERD
52
O
AINT1
84
—
N.C.
116
I
MOD2
21
I
WMD0
53
—
N.C.
85
O
IRO27
117
—
N.C.
22
O
EA1
54
—
N.C.
86
—
N.C.
118
I
INT1
23
O
EA4
55
I
SMCK
87
—
N.C.
119
—
N.C.
24
O
EA5
56
I
SMEN
88
—
N.C.
120
—
N.C.
25
O
EA9
57
O
PDXED
89
O
BTACT
121
—
N.C.
26
O
EA11
58
I
SYI1
90
O
XEWR
122
—
VS
27
—
N.C.
59
I
SCI1
91
I
WMD1
123
I
SY10
28
O
EA15
60
O
SDO0
92
O
EA2
124
I
SDI1
29
I/O
ED2
61
O
SDO1
93
—
N.C.
125
I
SYO0
30
I/O
ED6
62
O
PAO0
94
O
EA7
126
I
SYO1
31
I/O
ED8
63
O
PAO4
95
O
EA10
127
O
XMONI
32
—
N.C.
64
I
ICCN
96
O
EA14
128
O
PAO3
N.C.: Pin not connected
4
(Continued)
MB86330
(Continued)
Pin
no.
I/O
Pin
name
Pin
no.
I/O
Pin
name
Pin
no.
I/O
Pin
name
Pin
no.
I/O
Pin
name
129
O
PAO6
161
I/O
ED11
193
O
PAO26
225
—
N.C.
130
O
PAO8
162
I/O
ED14
194
—
VS
226
—
VS
131
O
PAO11
163
I/O
ICDT14
195
—
N.C.
227
—
N.C.
132
O
ST1
164
—
N.C.
196
—
VS
228
—
VS
133
O
PAO14
165
I/O
ICDT8
197
O
IRO7
229
I
SMDT
134
O
PAO17
166
I/O
ICDT4
198
O
IRO11
230
—
VS
135
O
PAO20
167
I/O
ICDT1
199
—
VS
231
I
SCI0
136
O
PAO22
168
I
INT7
200
—
N.C.
232
—
VS
137
O
PAO25
169
—
N.C.
201
O
IRO24
233
O
ADBRK
138
O
PAO29
170
O
AINT3
201
—
VS
234
O
PAO1
139
—
N.C.
171
—
N.C.
203
O
IRO31
235
—
VS
140
O
IRO2
172
I
PSTOP
204
—
VS
236
I
XICOPE
141
O
IRO5
173
I
PM
205
O
PAGE0
237
O
ST0
142
O
IRO9
174
I
F1
206
—
VS
238
—
VS
143
O
IRO12
175
I
MOD1
207
O
EA0
239
O
PAO18
144
O
IRO15
176
I
INT2
208
—
VS
240
—
VS
145
O
IRO19
177
I
XRST
209
O
EA8
241
—
VD
146
O
IRO22
178
—
N.C.
210
O
EA12
242
—
VD
147
O
IRO25
179
—
N.C.
211
—
VS
243
—
VD
148
O
IRO29
180
I
TCIF
212
I/O
ED5
244
—
VD
149
—
N.C.
181
I
SDI0
213
I/O
ED9
245
—
VD
150
—
N.C.
182
I
SCO0
214
—
VS
246
—
VD
151
I
BOOT
183
O
PACK
215
I/O
ICDT15
247
—
VD
152
O
XEREQ
184
I
SCO1
216
—
VS
248
—
VD
153
—
N.C.
185
O
PAO2
217
I/O
ICDT7
249
—
VD
154
O
EA3
186
O
PAO5
218
—
VS
250
—
VD
155
O
EA6
187
O
PAO9
219
I/O
ICDT0
251
—
VD
156
—
N.C.
188
O
PAO12
220
—
VS
252
—
VD
157
O
EA13
189
I
XICWE
221
O
AINT4
253
—
VD
158
I/O
ED0
190
O
PAO16
222
O
SCKOUT
254
—
VD
159
I/O
ED4
191
O
PAO19
223
—
VS
255
—
VD
160
I/O
ED7
192
O
L
224
—
N.C.
256
—
VD
N.C.: Pin not connected
5
MB86330
■ EXTERNAL PIN LAYOUT
MCLK
L
PM
FF
PSTOP
BREAK
3
MOD [2 : 0]
XICWE
2
WMD [1 : 0]
XICOPE
4
F1
ICAD [3 : 0]
F0
ICCN
16
XRST
ICDT [15 : 0]
SCKOUT
XMONI
BOOT
ADBRK
32
BTACT
IRO [31 : 0]
2
32
PAGE [1 : 0]
PAO [31 : 0]
3
ST [2 : 0]
PACK
SMDT
PDXED
SMCK
XERD
SMEN
XEREQ
7
INT [7 : 1]
XEWR
7
AINT [7 : 1]
16
ED [15 : 0]
16
TCIF
EA [15 : 0]
2
SY1 [1 : 0]
SCZC
2
SCI [1 : 0]
2
SDI [1 : 0]
SYO [1 : 0]
2
2
SCO [1 : 0]
2
SDO [1 : 0]
Other pins
VD, VS, N.C.
6
MB86330
■ PIN DESCRIPTION
Pin no.
114
Pin name Bit
MCLK
1
I/O
Active
Pull up
or
pull down
I
—
—
Function
Master clock input
MCLK
SCKout
(Internal system
clock)
173
PM
1
I
—
Pull up
172
PSTOP
1
I
H
—
50,
116,
175
MOD
[2:0]
3
I
—
1 machine cycle
Internal master clock input can be selected.
0: MCLK, 1: PLL output
PLL operation setup
0: PLL operation, 1: PLL stop
Pull down Operating mode
MOD2
MOD1
MOD0
Operating
mode
0
0
0
Single chip
mode
Other than above
21,
91
WMD
[1:0]
2
I
—
Disabled
Pull down External memory WAIT mode
WMD1
SMD0
Wait
cycle
Can data be
rewritten?
0
0
0 cyc
No
0
1
5 cyc
No
1
0
15 cyc
Yes
1
1
30 cyc
Yes
174
F1
1
I
—
—
Flag input 1 (level sense)
49
F0
1
I
—
—
Flag input 0 (level sense)
177
XRST
1
I
L
—
Reset input
222
SCKOUT
1
O
—
—
Internal system clock output
44
SCZC
1
I
H
151
BOOT
1
I
H
—
Input for a BOOT mode control signal
89
BTACT
1
O
H
—
Output for a BOOT mode status indication
signal
19,
205
PAGE
[1:0]
2
O
—
—
Output for an external memory/page
selection control signal
67,
132,
237
ST
[2:0]
3
O
—
—
Internal status output
Pull down Hi-z control over SCKOOUT, POUT and EA
[15:0] (SCZC = “L”)
(Continued)
7
MB86330
Pin no.
Pin name Bit
I/O
Active
Pull up
or
pull down
Function
229
SMDT
1
I
—
—
Serial input data (16 bits) for operating mode
(SMODE) setup
55
SMCK
1
I
—
—
Serial input clock for operating mode (SMODE)
setup
56
SMEN
1
I
H
—
Pulse input for operating mode (SMODE) setup
Upon completion of setup, a positive pulse is
entered.
43, 109 to 111, INT
118, 168, 176 [7:1]
7
I
L
Pull up
41, 42, 51, 52, AINT
108, 170, 221 [7:1]
7
O
L
—
Output for INT7 to INT1 interrupt acknowl edge
signals
Input for INT7 to INT1 interrupt request
signals
180
TCIF
1
I
H
—
Used for DC setup. “0”: PCM, “1”: TCH
Used to set serial port 1.
58,
123
SYI
[1:0]
2
I
H
—
Input pins for synchronization signals for serial
input port 1/0
59,
231
SCI
[1:0]
2
I
—
—
Clock input for serial input port 1/0
124,
181
SDI
[1:0]
2
I
—
—
Data input for serial input port 1/0
125,
126
SYO
[1:0]
2
I
H
—
Synchronization signal input for serial output
port 1/0
182,
184
SCO
[1:0]
2
I
—
—
Clock input for serial output port 1/0
60,
61
SDO
[1:0]
2
O
—
—
Data output for serial output port 1/0
192
L
1
O
—
—
PLL status output
70
FF
1
O
—
—
Output for test
115
BREAK
1
I
L
Pull up
Break input for the emulator
189
XICWE
1
I
L
Pull up
Input for an emulator write signal
236
XICOPE
1
I
L
Pull up
Input for an emulator read signal
2, 6, 10, 79
ICAD
[3:0]
4
I
—
Pull down Address input for the emulator
64
ICCN
1
I
H
Pull down Input for an emulator connection signal
35 to 37,39, ICDT
103 to 107, 163, [15:0]
165 to 167,
215, 217, 219
16
I/O
—
Pull down I/O for a data bus used to access the
emulator
127
XMONI
1
O
L
—
Output for indicating emulator monitor mode
status
233
ADBRK
1
O
H
—
Output for indicating occurrence of an ADBKP
register event for the emulator
(Continued)
8
MB86330
(Continued)
Pin no.
Pin name Bit
I/O
Active
Pull up
or
pull down
Function
4, 5, 7 to 9,
IRO
11 to 16, 76 to 78, [31:0]
80 to 83, 85,
140 to 148, 197,
198, 201, 203
32
O
—
—
Instruction register output for the emulator
1, 3, 62, 63, 65, PAO
66, 68, 69,
[31:0]
71 to 75,
128 to 131,
133 to 138,
185 to 188,
190, 191, 193,
234, 239
32
O
—
—
Program address output for the emulator
183
PACK
1
O
—
—
Output for a PAO fetch clock for the emulator
57
PDXED
1
O
—
—
Output for test
20
XERD
1
O
L
—
Output for an ERAM reading signal
152
XEREQ
1
O
L
—
Output for an ERAM access request signal
90
XEWR
1
O
L
—
Output for an ERAM writing signal
29 to 31, 33,
97, 98,
100 to 102,
158 to 162,
212, 213
ED
[15:0]
16
I/O
—
Pull up
22 to 26, 28, EA
92, 94 to 96, [15:0]
154, 155, 157,
207, 209, 210
16
O
—
—
Output for the ERAM address
VD
—
—
—
—
Power supply for the digital circuit
(3.3 V, input)
122, 194, 196, VS
199, 202, 204,
206, 208, 211,
214, 216, 218,
220, 223, 226,
228, 230, 232,
235, 238, 240
—
—
—
—
GND (input) for the digital circuit
17, 18, 27, 32, N.C.
34, 38, 40,
45 to 48, 53, 54,
84, 86 to 88, 93,
99, 112, 113,
117,119 to 121,
139, 149, 150,
153, 156, 164,
169, 171, 178,
179, 195, 200,
224, 225, 227
—
—
—
—
Pins not connected
241 to 256
External data bus I/O pins
9
MB86330
■ HANDLING DEVICES
1. Take Care So that the Maximum Rated Value Is Not Exceeded. (Preventing Latchup)
Latchup may occur on CMOS ICs if voltage higher than VD or lower than VS is applied to input and output pins
other than medium- and high-voltage pins or if higher than the voltage which shows on “1. Absolute Maximum
Ratings” in section “■ Electrical Characteristics” is applied between VD and VS.
When latchup occurs, power supply current increases rapidly and might thermally damage elements. When
using, take great care not to exceed the absolute maximum ratings.
2. Treatment of Unused Input Pins
Leaving unused input pins open could cause malfunctions. They should be connected to a pull-up or pull-down
resistor.
3. Treatment of N.C. Pins
Be sure to leave (internally connected) N.C. pins open.
4. Treatment of Pins Connected to Pull-up/Pull-down Resistors
With neither a pull-up resistor nor a pull-down resistor connected, the pin state is determined depending on the
input level that reflects an internal resistor. When controlling the pin state, however, connect a pull-up or pulldown resistor.
10
MB86330
■ BLOCK DIAGRAM
DSP core
Peripheral circuit
Controller
Clock
generator
PC
Interrupt
controller
LC0
I/O
peripheral circuit
LC1
IR1
DEC1
IR2
DEC2
RPC
DRF
MODE
RPC2
Y0
X0
SP
Y1
X1
DMAC0
X2
DMAC1
X3
DMAC2
X4
DMAC3
Bus interface
ARAM
(Data memory)
BRAM
(Data memory)
X5
X6
Adder
X7
MD
BP
Circular
addressing
TRAM
(Table data memory)
BV
Instruction
memory
Adder
Address
generator
B-bus
A-bus
Data operation
section
Input format
A1
A0
B1
B0
C2
C1
C0
D2
D1
D0
Output format
SFT
MAC#0
MAC#1
ALU
Barrel shifter
ST
SFTV
Result selection
11
MB86330
■ DESCRIPTION OF BLOCK FUNCTIONS
• Clock generator (CLOCK Gen.)
Generates a clock required for the DSP to control a system clock stop in the waste state and an entire clock
stop for sleeping.
• Interrupt controller
Controls an INT interrupt, an overflow interrupt, and a DMA interrupt.
• Controller
Generates a program address and decodes an instruction to control the entire DSP.
• Address generator
Generates an address required for memory access. It supports a circular addressing function to control the
DMA access pointer and the stack pointer.
• Data operation section
Performs data operations such as product addition, arithmetic operations (multiplication, division, addition and
subtraction), logic operations, and shift operations.
• Bus interface
Controls access to the memory space including instruction reading, data memory access and mapped I/O
access.
• ARAM/BRAM
This is a data memory for data operations. A and B use different banks, enabling double transfer without a
wait. (ARAM = 4 kwords, BRAM = 4 kwords)
• TRAM (Table data memory)
Sets table data required for applications.
• Instruction memory
Sets program data.
• Mapped I/O
Supports a macro valid for applications
Serial I/O: Two serial ports for transmitting CODEC data.
One serial input 1 system for setting operating mode (SMODE)
12
MB86330
■ MEMORY SPACE
• Configuration of the memory space
The memory space consists of a data memory space and an instruction memory space. The I/O, ARAM,
BRAM, TABLE, and ERAM (external memory) areas are allocated in the data memory space, while the
instruction memory is allocated in the program memory space. The I/O, TABLE, ERAM (external memory),
ARAM and BRAM areas in the data memory space are, however, assigned a particular data bus. The memory
for any two areas can, therefore, be accessed concurrently during one cycle. What memory is accessed is
determined automatically by an address value.
Addresses allocated for the memories, and their maximum size are determined as follows. Select a memory
size to be allocated in this range.
• Memory space mode
You can select a method for allocating the data memory space and the program memory space by operating
mode. Determine operating mode by mode pins (MOD2, MOD1 and MOD0) as follows.
MOD2
MOD1
MOD0
Operating mode
0
0
0
Single chip mode
Other than above
Disabled
• Memory map for single chip mode
The program for single chip mode (MOD2: 0 = “000”) is operated by the internal program RAM (which externally
downloads a program). Five areas for I/O, ARAM, BRAM, TABLE and ERAM are allocated on the data memory
space, with the instruction memory allocated on a program memory space completely different from the data
memory space. Although the instruction memory can have an independent address space for 64 kwords,
therefore, you cannot access data in the instruction memory using a program.
• Single chip mode (MOD2: 0 = “000”)
Data memory area
Instruction memory space
0000H
0000H
I/O area
1000H
4000H
ARAM area
Emulation work area
4100H
2000H
BRAM area
3000H
TABLE area
Internal instruction
memory for the user
7000H
8000H
ERAM area
FFFFH
FFFFH
16 bits
16 bits
: The hatched area cannot be used by the user.
13
MB86330
■ MEMORY MAP FOR BOOT STRAP
Instruction memory
ERAM
0000H
PAGE1 – 0 = 10
1000H
PAGE1 – 0 = 00
0000H
0000H
Data memory
0000H
2000H
I/O area
ARAM
BRAM
3000H
4000H
3000H
4000H
Table RAM
Table data
Program
6FFFH
7FFFH
7FFFH
PAGE1 – 0 = 11
8000H
8000H
PAGE1 – 0 = 01
8000H
Instruction
RAM
ERAM
ERAM
FFFFH
FFFFH
FFFFH
FFFFH
The ERAM area is used by the DSP to access the data memory space. Because the ERAM area ranges from 8000H
to FFFFH, however, PAGE is created in units of 32 kwords. PAGE is selected automatically.
14
MB86330
■ BASIC PIPELINE OPERATION
The DSP splits the contents of processing in one cycle to increase the number of pipeline sectors for high-speed
operation. For operations using product adders such as product addition and multiplication, and for operations
using 40-bit adders such as 40-bit addition, the processing latency is two cycles.
Pipeline phase
PC
DE1
DE2
EX1
PC
dec1
adr
dec2
ALU1
Operation (latency 2)
PC
dec1
adr
dec2
Transfer (Reg-Reg R/W)
PC
dec1
adr
dec2
R/W
Transfer (Mem Read)
PC
dec1
adr
dec2
R
Transfer (Mem Write)
PC
dec1
adr
dec2
[adr]
Operation (latency 1)
EX2
PC: Program fetch cycle
DE1: Decode 1st. cycle
DE2: Decode 2nd. cycle
EX1: Execute 1st. cycle
EX2: Execute 2nd. cycle
dec1:
dec2:
adr:
[adr]
ALU:
R:
W:
ALU2
1st. decoding
2nd. decoding
Address generation
Address maintenance
Operation
Reading
Writing
W
■ PRODUCT ADDITION
For product addition and multiplication, the latency is two cycles. Because it is provided with a dual product
adder (MAC) for alternate processing every cycle, however, the DSP can process n successive product addition
(multiplication) steps in (n + 1) cycles.
Operation latency 2
MSM (1)
MSM (3)
MSM (2)
MSM (n - 1)
MSM (4)
MSM (n)
(n + 1) cycles
15
MB86330
■ REGISTER TABLE
Register
name
Bit
length
A0
16
Data register
A1
16
Data register
AX
32
Data register
B0
16
Data register
B1
16
Data register
BX
32
Data register
C0
16
Accumulator
C1
16
Accumulator
C2
8
Accumulator
(guard register)
CX
40
Accumulator
D0
16
Accumulator
D1
16
Accumulator
D2
8
Accumulator
(guard register)
DX
40
Accumulator
X0
16
Address register
X1
16
Address register
X2
16
Address register
Y0
X0
X3
16
Address register
Y1
X1
X4
16
Address register
X2
X5
16
Address register
X3
X6
16
Address register
X7
16
Address register
Y0
16
Address register
update register
Y1
16
Address register
update register
Register type
Initial value
Undefined
Register configuration
bit 31
bit 16 bit 15
A1
bit 31
bit 0
AX
A0
bit 16 bit 15
B1
bit 0
BX
B0
Undefined
bit 39 bit 32 bit 31
C2
bit 16 bit 15
C1
bit 39 bit 32 bit 31
D2
bit 0
C0
bit 16 bit 15
D1
CX
bit 0
D0
DX
Undefined
bit 15
bit 0
bit 15
bit 0
X4
X5
X6
X7
(Continued)
16
MB86330
(Continued)
Register
name
Bit
length
BP
16
Base pointer
BV
16
Circular register
MD
16
Modulo register
RPC
16
Repeat counter 1
Register type
Initial value
Register configuration
Undefined
bit 15
bit 0
bit 15
bit 0
BP
ST
BV
bit 15
RPC2
16
Repeat counter 2
DOSTR
16
DO start address
register
DOEND
16
DO end address
register
DOSTR
LC0
16
Loop counter
DOEND
LC1
16
Loop counter
LC0
SFT
6
Shift register
LC1
SFTV
16
Shift register
ST
16
Status register
MODE
16
Mode register
DRF
16
Flag holding
register
DMAC0
16
DMA counter
DMAC1
16
DMA counter
DMAC2
16
DMA counter
DMAC3
16
DMA counter
PC
16
Program counter
SP
16
Stack pointer
bit 0
MD
MODE
RPC
DRF
RPC2
bit 15
00000000B
00000000B
bit 5
bit 0
SFT
bit 0
SFTV
Undefined
bit 15
bit 0
bit 15
PC
DMAC0
SP
DMAC1
bit 0
DMAC2
FFFD
DMAC3
Undefined
17
MB86330
■ REGISTERS
• Data registers (A0, A1, B0 and B1)
Each of the data registers consists of four words (16 bits). They can be used as four word-length registers (16
bits) and two long-word registers (32 bits) to execute various arithmetic operation instructions, logic operation
instructions, and transfer instructions.
• Accumulators (C0 to C1, and D0 to D2)
The accumulators can be linked as two 40-bit registers (CX and DX) to execute various arithmetic operation
instructions, logic operation instructions, and transfer instructions. The 40-bit length registers (CX and DX)
can be specified as destinations for product addition instructions. Four 16-bit length accumulators (C0, C1,
D0 and D1), and two 8-bit length accumulators (C2 and D2) are supported.
• Address registers (X0 to X7)
Eight 16-bit address registers are supported. An address register is used to specify an operand address for
transfer. Immediate values (1 to –2) or the address update registers (Y0 and Y1) can be used to update address
registers. They can also be updated automatically by transfer.
• Address update registers (Y0 and Y1)
Two 16-bit address update registers are supported. The address update registers are used to update address
registers during addressing.
• Base pointer (BP)
The base pointer consists of 16 bits. The contents of the base pointer plus a 7-bit immediate value are generated
as the address value during direct 7-bit length addressing.
• Circular register (BV)
The circular register, which consists of 16 bits, provides an offset value for circular addressing.
• Modulo register (MD)
The modulo register, which consists of 16 bits, is used to specify an addressing range for circular addressing.
• Repeat counter (RPC)
The repeat counter, which consists of 16 bits, is used to specify the number of times the REP/DO instruction
is repeated. During execution of the repeat instruction, the repeat counter is decremented by one every repeat
operation cycle.
• Repeat counter 2 (RPC2)
Repeat counter 2, which consists of 16 bits, is used to specify the number of times the REP2 instruction is
repeated. During execution of the repeat 2 instruction, repeat counter 2 is decremented by one every repeat
operation cycle.
• DO address registers (DOSTR and DOEND)
These registers maintain a loop start address (DOSTR)/end address (DOEND) for the DO instruction. They
can process only PUSH/POP.
• Loop counters (LC0 and LC1)
Each of the loop counters consists of 16 bits. They store the number of times repetition is made in a specified
address range.
(Continued)
18
MB86330
(Continued)
• Shift register (SFT)
The SFT register consists of signed 6 bits. This shift value storage register stores the number of bits shifted
during execution of the shift instruction.
• Shift register (SFTV)
The SFTV register, which consists of 16 bits, is used to store the results of CMLT and CMGT instruction.
• Status register (ST)
The status register, which consists of 16 bits, is assigned bits for storing information about results of operations
(carry and overflow) and for setting operating mode.
• Mode register (MODE)
This register is used to specify modes of operations and transfer, and interrupts.
• Flag holding register (DRF)
This register holds flags for the DO, REP and REP2 instructions. It can process only PUSH/POP. This register
is cleared by the PUSH instruction.
• DMA counters (DMAC0 to DMAC3)
When a DMA interrupt occurs, this register stores the address of the data transfer source or the data transfer
destination.
• Program counter (PC)
The program counter, which consists of 16 bits, points to the memory address that stores an instruction code
to be executed by the CPU. While it is updated automatically by instruction execution, the program counter
can be rewritten by a conditional branch, a subroutine call instruction, an interrupt, and a reset. Executing the
repeat instruction stops a program counter update.
• Stack pointer (SP)
The stack pointer, which consists of 16 bits, stores addresses for saving and transferring the contents of
registers upon execution of the PUSH/POP instruction, the subroutine call instruction, or an interrupt.
19
MB86330
■ DETAILED DESCRIPTION OF SPECIAL REGISTERS
(1) Status Register
bit 15
bit 14
bit 13
bit 12
bit 11
bit 10
IT
OV3
OV1
INT2
INT1
INT0 MDMA
Bit abbreviation
bit 8
bit 7
bit 6
bit 5
bit 4
bit 3
bit 2
bit 1
bit 0
CP
RND
ITG
V3
V2
V1
N
Z
C
Bit name
Description
C
Carry flag
Set when carry occurs as a result of operation execution.
Reset when no carry occurs.
Not changed by transfer instruction execution.
Z
Zero flag
Set when the operation result is 0.
Reset when the operation result is not 0.
Not changed by transfer instruction execution.
N
Negative flag
Set when the operation result is smaller than 0.
Reset when the operation result is equal to or greater than 0.
Not changed by transfer instruction execution.
V1
Overflow flag 1
Set when the operation result overflows.
Reset when the operation result does not overflow.
Not changed by transfer instruction execution.
V2
Overflow flag 2
Set when the operation result overflows.
Set V2 is reset only by hardware or by ST programming by the
transfer instruction.
Not changed by transfer instruction execution.
V3
Overflow flag 3
Set when the operation result of an instruction stored in CX or DX
cannot be expressed by 32 bits (but by 40 bits).
Reset when the operation result can be expressed by 32 bits.
ITG
Operating mode
specification flag
Specify this when executing multiplication in integral mode.
RND
Rounding mode setup
Used to set ON/OFF of rounding processing when data is
transferred from a register consisting of 32 or more bits to a 16-bit
register.
Clip flag
Used to specify whether the operation result is to be clipped
when overflow occurs during the operation.
DMA enable flag
Enables a DMA interrupt. (0: Disabled)
INT0
Interrupt enable flag
INT0 (SMODE) interrupt enable flag
0: Disabled, 1: Enabled
INT1
Interrupt enable flag
INT1 interrupt enable flag
0: Disabled, 1: Enabled
INT2
Interrupt enable flag
INT2 interrupt enable flag
0: Disabled, 1: Enabled
OV1
V1 interrupt enable flag
Operation overflow interrupt enable flag. An interrupt is generated
when V1 is set.
0: An interrupt is disabled. 1: An interrupt is enabled.
OV3
V3 interrupt enable flag
Operation overflow interrupt enable flag. An interrupt is generated
when V3 is set.
0: An interrupt is disabled. 1: An interrupt is enabled.
Interrupt enable flag
OV1, OV3 and INT0 to INT7 interrupt enable flag
0: An interrupt is disabled. 1: An interrupt is enabled.
CP
MDMA
IT
20
bit 9
MB86330
(2) Mode Register
bit 7
bit 6
bit 5
bit 4
bit 3
bit 2
bit 1
bit 0
INT7
INT6
INT5
INT4
INT3
—
NCT
NOG
Bit abbreviation
Bit name
Description
NOG
Operating mode
specification
0: Ordinary mode
The guard bit is used.
1: NOG mode
No guard bit is used.
NCT
Transferred data
clipping specification
0: Transferred data is clipped.
1: Transferred data is not clipped.
Indeterminate
Reserved
INT3
Interrupt enable flag
INT3 interrupt enable flag
0: Disabled, 1: Enabled
INT4
Interrupt enable flag
INT4 interrupt enable flag
0: Disabled, 1: Enabled
INT5
Interrupt enable flag
INT5 interrupt enable flag
0: Disabled, 1: Enabled
INT6
Interrupt enable flag
INT6 interrupt enable flag
0: Disabled, 1: Enabled
INT7
Interrupt enable flag
INT7 interrupt enable flag
0: Disabled, 1: Enabled
—
21
MB86330
(3) DRF Register
bit 7
bit 6
bit 5
bit 4
bit 3
bit 2
bit 1
bit 0
REPF23
REPF22
REPF21
REPF13
REPF12
REPF11
DOF1
DOF2
Bit abbreviation
22
Bit name
DOF2
DO flag 2
DOF1
DO flag 1
REPF11
REP1 flag 1
REPF12
REP1 flag 2
REPF13
REP1 flag 3
REPF21
REP2 flag 1
REPF22
REP2 flag 2
REPF23
REP2 flag 3
Description
Internal operation status holding flag.
Only PUSH and POP are available.
Cleared by the PUSH instruction.
MB86330
■ ADDRESSING
• Types of addressing
When reading/writing data from/to the memory, you can use a direct addressing method for specifying a
16-bit length address space with an immediate value, and an indirect addressing method for referencing that
space by an address register. The DSP supports eight address registers, two update registers, the base
pointer, the circular register, and the modulo register for addressing.
• Addressing classification
Three indirect addressing means (AD0 to AD2) by address registers can be used to transfer a register value
to a memory, data from a memory to a register, and data between memories. The available addressing means
is determined depending on the type of a register for data transfer, double transfer and transfer accompanied
by an operation. All addressing is performed in units of words.
23
MB86330
• Addressing modes
Mode
Mnemonic
Direct
addressing
AD0
Indirect addressing
AD1
AD2
Effective address
Register update
Description
(imm16)
imm16
Not updated
16-bit direct addressing
(Xk + + 1)
(Xk + + 0)
(Xk – – 1)
(Xk + + Y0)
Xk
Xk
Xk
Xk
+1
Not updated
–1
Y0
Xk can be assigned X0,
X1, X2 and X3.
(Xm + + 3)
(Xm + + 2)
(Xm + + 1)
(Xm + + 0)
(Xm – – 1)
(Xm – – 2)
(Xm – – 3)
(Xm + + Y1)
Xm
Xm
Xm
Xm
Xm
Xm
Xm
Xm
+3
+2
+1
Not updated
–1
–2
–3
Y1
Xm can be assigned X4,
X5 and X6.
[BV+X7 + + 3]
[BV+X7 + + 2]
[BV+X7 + + 1]
[BV+X7 + + 0]
[BV+X7 – – 1]
[BV+X7 – – 2]
[BV+X7 – – 3]
[BV+X7 + + Y1]
BV+X7
BV+X7
BV+X7
BV+X7
BV+X7
BV+X7
BV+X7
BV+X7
+3
+2
+1
Not updated
–1
–2
–3
Y1
For circular addressing,
only X7 can be used.
(BP+disp7)
(Xn + + 2)
(Xn + + 1)
(Xn + + 0)
(Xn – – 1)
(Xn – – 2)
(Xn – – 3)
(Xn + + Y0)
(Xn + + Y1)
[BV+Xn + + 2]
[BV+Xn + + 1]
[BV+Xn + + 0]
[BV+Xn – – 1]
[BV+Xn – – 2]
[BV+Xn – – 3]
[BV+Xn + + Y0]
[BV+Xn + + Y1]
BP+disp7
Xn
Xn
Xn
Xn
Xn
Xn
Xn
Xn
BV+Xn
BV+Xn
BV+Xn
BV+Xn
BV+Xn
BV+Xn
BV+Xn
BV+Xn
Not updated
+2
+1
Not updated
–1
–2
–3
Y0
Y1
+2
+1
Not updated
–1
–2
–3
Y0
Y1
Xn can be assigned X0,
X1, X2, X3, X4, X5, X6
and X7.
disp7 is signed 7 bits.
[ ] : Indicates circular addressing.
AD0 : For “AD0,” you cannot specify circular addressing.
AD1 : For “AD1,” circular addressing mode is set automatically when “X7” is selected as the address register. With
“0” set in the “MD” and “BV” registers, the same operation as ordinary addressing is performed even if “X7” is
selected as the address register.
AD2 : disp7 in “AD2” is signed 7 bits.
24
MB86330
■ BASIC CONFIGURATION OF THE DATA OPERATION SECTION
1. Data Format
Integral type and fixed-point data can be handled regardless of whether the data is signed or unsigned. For
signed data, the most significant bit indicates a sign. A number of 1 indicates negative data, which is expressed
by 2’s complement. The decimal point for fixed-point type data is located between the sign bit (bit 14) and its
right bit (bit 15). When the accumulator value resulting from execution of a multiplication instruction is transferred
to a 16-bit length register or memory, the decimal point is returned to the original position (between bits 14 and 15).
• Fixed-point type, signed data
15
14
13
12
11
10
9
8
7
6
4
3
2
1
0
bit
5
4
3
2
1
0
bit
5
4
3
2
1
0
bit
4
3
2
1
0
bit
5
S
Decimal part (2’s complement)
Sign
Decimal point
• Fixed-point type, unsigned data
15
14
13
12
11
10
9
8
7
6
Decimal part
Decimal point
• Integral-type, signed data
15
14
13
12
11
10
9
8
7
6
S
Integral part (2’s complement)
Sign
• Integral-type, unsigned data
15
14
13
12
11
10
9
8
7
6
5
Integral part
Since a value is handled in 2’s complement format for addition and subtraction, no distinction is made in the
result of an operation by the above four data format. Since, for multiplication, the result of an operation varies
with whether data is signed or unsigned, three combinations of “signed data x signed data,” “signed data ×
unsigned data,” and “unsigned data × unsigned data” exist for multiplication instructions.
For the fixed-point type and the integral type, the result of multiplication varies with the decimal point position.
When fixed-point type data is multiplied, the result of the operation is stored into the accumulator with one bit
shifted to the left in comparison with integral type data. The “ITG” bit in the status register is used to switch
between the fixed-point type and the integral type. With this bit set at 0, an operation is executed in the fixedpoint type format.
25
MB86330
2. Multiplication in Fixed-point Type Mode
With the “ITG” bit in the status register set at 0, fixed-point type mode is set up, and multiplication is executed
in the following format. You can use ordinary mode and NOG mode in which the guard bit is not used.
(1) Ordinary Mode
• Fixed-point type multiplication (signed data × signed data)
31 bits of the result of a signed operation are stored into bits 1 to 31 in the accumulator shifted to the left by
one bit. “0” is set to bit 0, with the same value as at bit 31 set to bits 32 to 39.
15 14 13 12
1
0
S
×
S
S
S
S
0
16 bits
S
0
40 bits
32 31 30 29 28
39 38
16 bits
Sign extension
17 16 15 14
2
1
0
31 bits of an operation result
• Fixed-point type multiplication (signed data × unsigned data)
32 bits of the result of a signed operation are stored into bits 1 to 32 in the accumulator shifted to the left by
one bit. “0” is set to bit 0, with the same value as at bit 32 set to bits 33 to 39.
15 14 13 12
1
0
S
16 bits
×
S
S
S
39
33 32 31 30 29 28
Sign extension
17 16 15 14
2
1
0
16 bits
0
40 bits
0
32 bits of an operation result
• Fixed-point type multiplication (unsigned data × unsigned data)
32 bits of the result of an unsigned operation are stored into bits 1 to 32 in the accumulator shifted to the left
by one bit. “0” is set to bit 0, and bits 33 to 39.
15 14 13 12
1
0
16 bits
×
0
39
0
33 32 31 30 29 28
0 is set.
26
17 16 15 14
32 bits of an operation result
2
1
0
16 bits
0
40 bits
0
MB86330
(2) NOG Mode
In this mode, the CX and DX registers are handled as a 32-bit accumulator in which detected overflow is clipped.
• Fixed-point type multiplication (signed data x signed data)
31 bits of the result of a signed operation are stored into bits 1 to 31 in the accumulator shifted to the left by
one bit. “0” is set at bit 0.
15 14 13 12
1
0
16 bits
S
×
S
0
16 bits
S
0
32 bits
31 30 29 28
17 16 15 14
2
1
0
31 bits of an operation result
27
MB86330
■ INTERRUPT
A software interrupt and a hardware interrupt are available. Interrupts are assigned specified types of priority.
When interrupts occur concurrently, an interrupt with higher priority is executed earlier.
Priority
Soft/hard
Interrupt type
Interrupt branch
destination
1
Hard
RST
0XFFFE
Hardware reset
External reset signal input
2
Hard
BREAK
0X0002*
Setup of emulator operation mode
External break signal input
Soft
TRAP
Soft
V1
0XFFFC
Occurrence of arithmetic operation overflow (V1)
V3
0XFFFA
Occurrence of arithmetic operation overflow (V3)
DMA0
—
DMA0 signal input (for data input)
6
DMA1
—
DMA1 signal input (for data input)
7
DMA2
—
DMA2 signal input (for data output)
8
DMA3
—
DMA3 signal input (for data output)
9
INT0
0XFFF8
SDOME interrupt signal input
10
INT1
0XFFF6
External interrupt signal input (INT1)
11
INT2
0XFFF4
External interrupt signal input (INT2)
12
INT3
0XFFF2
External interrupt signal input (INT3)
13
INT4
0XFFF0
External interrupt signal input (INT4)
14
INT5
0XFFEE
External interrupt signal input (INT5)
15
INT6
0XFFEC
External interrupt signal input (INT6)
16
INT7
0XFFEA
External interrupt signal input (INT7)
3
4
5
Hard
Cause of an interrupt
Setup of emulator operation mode
Software (TRAP instruction)
* : Memory space for a debug instruction
Notes: • Emulator operation mode is set by a hardware interrupt resulting from external break signal input and
by a software interrupt by the TRAP instruction.
• Any interrupts other than a reset are disabled during downloading.
28
MB86330
■ INSTRUCTIONS
ST
–
–
–
–
–
–
↓
↓
↓
↓
↓
↓
Conditional relative branch
–
–
–
–
–
–
CALL
Subroutine jump
–
–
–
–
–
–
CMGT
Transfer with (major)
comparison conditions
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
CMLT
Transfer with (minor)
comparison conditions
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
CMP
Comparison
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
Block repetition
RPC/imm 10
Division support
AX, BX, CX, DX/A0, A1,
B0, B1, C0, C1, D0, D1
↔
↔
–
–
–
–
–
↔
↔
↔
–
–
–
–
–
–
JPC1
Conditional absolute branch
–
–
–
–
–
–
JUMP
–
–
–
–
–
–
LSL
Absolute branch
....
C←
←0
↔
↔
↔
Conditional absolute branch
–
–
–
LSR
0→
↔
↔
JPC0
–
–
–
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
↔
→C
↔
....
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
↔
–
DSTP
↔
↔
↔
↔
↔
↔
↔
↔
–
↔
↔
↔
↑
↔
↑
↔
↔
↑
↔
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
↔
reg ↔ (AD1)
↔
CX, DX
↔
DO
....
↔
↔
↔
↔
BRCC
MSB
C
↔
↔
↑
ASL
C
↔
↑
ASR
←0
Shift with transfer
....
MSB ←
→C
ASL
↔
↑
↔
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
–
–
↔
←0
–
↔
Logical AND
....
MSB
–
↔
A0, A1, B0, B1, C0, C1,
D0, D1
–
↔
Logical AND
↑
↔
AND
↔
Xn ← Xn + Immediate
value
↔
Address register addition
↔
ADX
↔
A0, A1, B0, B1
↔
Addition and subtraction
↔
ADSB
↑
↔
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
↔
Addition with carry
↔
ADDC
↑
↔
Acc ← S1 + A1
reg ↔ (AD1)
↑
↔
Addition with transfer
V3
↑
↔
ADD
V2
↔
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
V1
↔
Addition
–
↔
ADD
0
↔
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
C
↔
Absolute value calculation
Z
↔
ABS
N
↔
Flag change
Operation overview
↔
Mnemonic
(Continued)
29
MB86330
↔
MSM
Signed product addition
A0, A1, B0, B1/CX, DX
0
MSM
Signed product addition with
duplicate transfer
acc ← acc + A0 × A1
A0 ← (AD0), A1 ← (AD1)
acc ← acc + B0 × B1
B0 ← (AD0), B1 ← (AD1)
0
MSM
Signed product addition with
transfer
acc ← acc + (A0 or A1)
× A1
reg ↔ (AD1)
0
MSMS
Signed and unsigned product
addition
A0, A1, B0, B1/CX, DX
0
MSMS
Signed and unsigned product
addition with duplicate
transfer
acc ← acc + A0 × A1
A0 ← (AD0), A1 ← (AD1)
acc ← acc + B0 × B1
B0 ← (AD0), B1 ← (AD1)
0
MSMU
Unsigned and unsigned
product addition
A0, A1, B0, B1/CX, DX
0
MSMU
Unsigned and unsigned
product addition with
duplicate transfer
acc ← acc + A0 × A1
A0 ← (AD0), A1 ← (AD1)
acc ← acc + B0 × B1
B0 ← (AD0), B1 ← (AD1)
0
↑
↑
↑
↑
↑
–
↑
–
↑
–
↑
–
↔
0
↔
A0, A1, B0, B1/CX, DX
↑
↔
Unsigned and unsigned
product addition
↔
MRDU
↔
0
↔
A0, A1, B0, B1/CX, DX
↔
Signed and unsigned product
addition
↔
MRDS
↑
↔
0
↔
acc ← acc – A0 × A1
A0 ← (AD0), A1 ← (AD1)
acc ← acc – B0 × B1
B0 ← (AD0), B1 ← (AD1)
↔
Signed product addition with
duplicate transfer
↔
MRD
↑
↔
0
↔
A0, A1,B0, B1/CX, DX
↔
–
Signed product addition
↔
–
MRD
↔
–
↔
–
↔
–
↔
–
CX, DX ← (AD1)
↔
–
Duplicate transfer to the
accumulator
↔
–
MOVT
↔
–
↔
–
↔
–
↔
–
↔
–
Inter-register transfer
REG ↔ REG
Transfer between a
memory and a register
MEM ↔ REG
Duplicate transfer
B0 ← (AD0), A1 ← (AD1)
B0 ← (AD0), B1 ← (AD1)
Immediate value transfer
to a register
REG ← Immediate value
Inter-memory transfer
(AD0) ← (AD1)
↔
–
Data transfer or duplicate
transfer
↔
V3
↔
V2
↔
V1
↔
C
↔
Z
↔
N
↔
Flag change
↔
MOV
Operation overview
↔
Mnemonic
(Continued)
30
MB86330
V2
–
–
–
–
↔
–
–
–
–
–
–
–
–
–
–
–
–
–
–
MULS
Signed and unsigned
multiplication with duplicate
transfer
acc ← A0 × A1
A0 ← (AD0), A1 ← (AD1)
acc ← B0 × B1
B0 ← (AD0), B1 ← (AD1)
MULU
Signed and unsigned
multiplication
A0, A1, B0, B1, C0, C1,
D0, D1/CX, DX
MULU
Signed and unsigned
multiplication with duplicate
transfer
acc ← A0 × A1
A0 ← (AD0), A1 ← (AD1)
acc ← B0 × B1
B0 ← (AD0), B1 ← (AD1)
MVCC
Conditional transfer
Inter-register transfer
REG ↔ REG
NOT
Logical NOT
A0, A1, B0, B1, C0, C1,
D0, D1
NEG
2’s complement operation
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
NOP
None executed
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
–
↔
↑
↔
↔
–
↔
–
↔
–
↔
–
↔
A0, A1, B0, B1, C0, C1,
D0, D1/CX, DX
↔
Signed and unsigned
multiplication
↔
MULS
↔
acc ← (A0 or A1) × A1
reg ↔ (AD1)
↔
Signed multiplication with
transfer
↔
MUL
–
↔
acc ← A0 × A1
A0 ← (AD0), A1 ← (AD1)
acc ← B0 × B1
B0 ← (AD0), B1 ← (AD1)
↔
Signed multiplication with
duplicate transfer
↔
MUL
–
↔
A0, A1, B0, B1, C0, C1,
D0, D1/CX, DX
↔
Signed multiplication
↔
MUL
V3
↔
V1
↔
C
↔
Z
↔
N
↔
Flag change
↔
Operation overview
↔
Mnemonic
–
↑
↑
↑
↑
↑
↑
Return of a register from the
stack
–
–
–
–
–
–
PUSH
Saving of a register to the
stack
–
–
–
–
–
–
REGU
Auxiliary normalization
operation
CX, DX
–
↑
–
–
–
–
REP
Repeated execution of the
subsequent instruction
RPC/imm 10
–
–
–
–
–
–
REP2
Repeated execution of the
subsequent instruction
RPC2/imm 10
–
–
–
–
–
–
RET
Return from a subroutine
–
–
–
–
–
–
RET1
Return from an interrupt
routine
–
–
–
–
–
–
OR
Logical OR
A0, A1, B0, B1, C0, C1,
D0, D1
ST
POP
(Continued)
31
MB86330
(Continued)
XOR
Exclusive logical OR
A0, A1, B0, B1, C0, C1,
D0, D1
↔
[Flag indications]
: Set or reset by an operation
↓ : Not changed or reset by an operation
– : Not changed by an operation
32
–
–
–
–
–
–
–
↑
↑
↑
↑
–
–
–
↑ : Not changed or reset by an operation
0 : Reset by an operation
↔
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
–
↔
Subtraction with carry
–
↔
SUBC
–
↔
acc ← acc – A1
reg ← (AD1)
–
↔
Subtraction with transfer
–
↔
SUB
–
↔
A0, A1, B0, B1, C0, C1,
D0, D1, AX, BX, CX, DX
↔
Subtraction
–
↔
SUB
–
–
↔
A0, A1, B0, B1
–
↔
Subtraction and addition
–
↔
SBAD
–
↔
(AD0) ← A0, A0 ← (AD1)
(AD0) ← A1, A1 ← (AD1)
(AD0) ← CX, CX ← (AD1)
(AD0) ← DX, DX ← (AD1)
–
↔
Store and load
↑
↔
STLD
–
↔
Standby status
V3
↔
SLP
V2
↔
A0, A1, B0, B1, C0, C1,
D0, D1
V1
↔
Reverse shift
C
↔
RVL
Z
↔
CX, DX
N
↔
Auxiliary search for the
minimum normalized value
Flag change
↔
RGLT
Operation overview
↔
Mnemonic
–
MB86330
■ BOOT
1. BOOT Mode
BOOT mode supports the following functions.
• Ordinary BOOT function (Booting the instruction RAM (48 kwords)/the table RAM (16 kwords))
• Simplified BOOT function (Booting some instruction RAMs)
2. Setting BOOT Mode
Only single chip mode is available.
(1) Command Setup
Load the ERAM area (FFE0H to FFE3H: PAGE1-0 = 11) with the following contents during booting.
• Address FFE0: Command contents
0001H to 000BH : Reserved
000CH
: Simplified booting
000DH to 0010H : Reserved
Others
: Ordinary booting
• Address FFE1: Start address
Specify the start address of a program to be booted during simplified booting.
• Address FFE2: Size
Specify the program size of a program to be booted during simplified booting.
• Address FFE3: Execution start address
Specify an execution start address after completion of simplified booting.
(2) Ordinary BOOT function (Command = other than 0001H to 0010H)
The ordinary BOOT function loads a full-word program (48 kwords) and table data (16 kwords), then moves
execution to the user program (jump to address FFFE) or ICE (jump to address 0002).
(3) Simplified BOOT function (Command = 000CH)
The Simplified BOOT function loads and executes a program of 48 k or fewer words.
It also sets loaded start address (FFE1H), the number of program words (FFE2H), and execution start address
(FFE3H).
33
MB86330
3. BOOT Timing
Performing BOOT processing requires satisfaction of the following operation.
(1) Take setup of two or more MCLK clocks from a fall rising edge.
(2) Fetch information about the BOOT pin at a fall rising edge, and the BTACT pin will be set at “H”.
(3) Reset the BOOT pin at least two MCLK clocks after a fall edge observed at the BTACT pin.
(4) When the BTACT pin is changed from “H” to “L”, BOOT operation is terminated.
• PM = 0 (When PLL is not used)
XRST
MCLK
[RESET]
SCKOUT
BOOT
BTACT
(3)
(1)
(2)
• PM = 1 (When PLL is used)
XRST
MCLK
[PLL-clk]
[RESET]
SCKOUT
BOOT
BTACT
(1)
(3)
(2)
34
MB86330
■ PLL
1. PLL operation
Performing this DSP operation using PLL requires satisfaction of the following operation.
When using PLL, set the PSTOP pin at “H” for 1 µsec or more for a reset, then at “L”, and wait for lockup time
or more time.
(1) Take enough time for MCLK input and for lockup at the PLL operation state with PSTOP equal to "L".
(2) Hold the DSP reset state until PLL is locked. (XRST = “L”)
(3) When PLL is locked, the “L” pin goes “H”.
(4) After PLL has been locked, change XRST from “L” to “H” to start DSP operation.
VDD
MCLK
PSTOP
lock up time
L
Initial time (
XRST
µs)
2. PLL standards
Input clock (MHz)
Look up time (µs)
20 to 25
200
Remarks
When PLL is used
35
MB86330
■ ELECTRICAL CHARACTERISTICS
1. Absolute Maximum Ratings
(VSS = 0 V)
Parameter
Symbol
Condition
Value
Unit
Min.
Max.
VSS – 0.5
4.0
V
Remarks
Power supply voltage
VDD
Input voltage
VI
—
VSS – 0.5
VDD + 0.5
V
Input pin
Output voltage
VO
—
VSS – 0.5
VDD + 0.5
V
Output pin
BUS pin
Maximum output current
Storage temperature
VDD – VSS
IO
VO = VDD
—
14
mA
IO
VO = 0 V
—
–14
mA
Output drive pin
(IOL) 4 mA
–65
+150
°C
Ceramic package
Tstg
—
WARNING: Semiconductor devices can be permanently damaged by application of stress (voltage, current,
temperature, etc.) in excess of absolute maximum ratings. Do not exceed these ratings.
2. Recommended Operating Conditions
(VSS = 0 V)
Parameter
Symbol
Value
Min.
Typ.
Max.
Unit
Power supply voltage
VDD
3.0
3.3
3.6
V
“H” level input voltage
VIH
0.65 VDD
—
VDD + 0.3
V
“L” level output voltage
VIL
VSS
—
0.25 VDD
V
Ambient temperature
TA
0
—
+40
°C
Remarks
VDD
WARNING: The recommended operating conditions are required in order to ensure the normal operation of the
semiconductor device. All of the device's electrical characteristics are warranted when the device is
operated within these ranges.
Always use semiconductor devices within their recommended operating condition ranges. Operation
outside these ranges may adversely affect reliability and could result in device failure.
No warranty is made with respect to uses, operating conditions, or combinations not represented on
the data sheet. Users considering application outside the listed conditions are advised to contact their
FUJITSU representatives beforehand.
36
MB86330
3. Operating Frequency
Operating Frequency
TA = +40°C
Frequency [MHz]
110
105
100
95
90
3.0
3.3
3.6
Power supply voltage [V]
4. DC Characteristics
(VDD = 3.0 to 3.6 V, VSS = 0 V, TA = 0°C to +40°C)
Parameter
Symbol
Condition
Value
Min.
Typ.
Max.
Unit
IDDS
Standby mode*1
—
50
—
µA
IDD
Ordinary operation
mode
—
62
—
mA
“H” level input voltage
VIH
—
0.65 VDD
—
VDD
V
“L” level input voltage
VIL
—
VSS
—
0.25 VSS
V
“H” level output voltage
VOH
IOH = –4mA
VDD – 0.5
—
VDD
V
“L” level output voltage
VOL
IOL = 4mA
VSS
—
0.4
V
Input leakage current*2
(Tri-state pin)
ILI
–5
—
5
µA
–5
—
5
µA
Pull-up/pull-down resistor
RP
25
50
200
kΩ
Output current
(Shorting circuit)
IO*3
VO = VDD
—
VO = 0 V
mA
+60
—
–60
mA
Power supply voltage
ILZ
VI = 0 – VDD
—
State
Normal/IOL = 4 mA
*1: The memory is set at the standby state with VIH = V DD and VIL = VSS.
*2: With the input pin provided with a pull-up or pull-down resistor, the standard value may be exceeded.
*3: Maximum supply current at the output section, and the VDD or VSS circuit
37
MB86330
5. AC Characteristics
(1) ERAM Interface
SCKOUT
tDEA
tHEA
EA [15 : 0]
tFREQ
tRREQ
XEREQ
tDRD
tWRD
XERD
tDWR
tWWR
XEWR
ED [15 : 0]
in
out
valid
valid
tSDRD
tDDWR
tHDRD
tHDWR
Note: During a wait, the state indicated by double lines above is held for a wait cycle.
(VDD = 3.0 V to 3.6 V, TA = 0°C to +40°C, output pin load = 50 pF)
Parameter
38
Symbol
Pin name
Value
Min.
Typ.
Max.
Unit
EA output delay
tDEA
EA
—
—
6.2
ns
EA hold time
tHEA
EA
2.6
—
—
ns
XEREQ falling delay
tFREQ
XEREQ
—
—
5.5
ns
XEREQ rising delay
tRREQ
XEREQ
—
—
4.2
ns
XERD rising delay
tDRD
XERD
—
—
1.0
ns
XERD “L” pulse width
tWRD
XERD
5.8
—
—
ns
XEWR rising delay
tDWR
XEWR
—
—
2.5
ns
XEWR “L” pulse width
tWWR
XEWR
4.2
—
—
ns
ED setup time for XERD
tSDRD (in)
ED
10.3
—
—
ns
ED hold time for XERD
tHDRD (in)
ED
1.5
—
—
ns
ED delay time for XEWR
tDDWR (out)
ED
—
—
8.5
ns
ED hold time for XEWR
tHDWR (out)
ED
0.3
—
—
ns
MB86330
(2) Serial I/O Interface
• Serial input
tCYCSCI
SCI0/1
SYI0/1
tHSYI
tSSYI
SDI0/1
Valid
tSSYI
Valid
Valid
tHSDI
tSSDI
• Serial output
tCYCSCO
SCO0/1
SYO0/1
tHSYO
tSSYO
SDO0/1
Valid
tSSYO
Valid
Valid
tDSDO
39
MB86330
(VDD = 3.0 V to 3.6 V, TA = 0°C to +40°C, output pin load = 50 pF)
Parameter
40
Symbol
Pin name
Value
Min.
Typ.
Max.
Unit
Serial input clock cycle
tCYCSCI
SCI [1 : 0]
SYI signal setup time
tSSTI
SYI [1 : 0]
2.1
—
SYI signal hold time
tHSYI
SYI [1 : 0]
1.4
—
SDI signal setup time
tSSDI
SDI [1 : 0]
1.5
—
—
ns
SDI signal hold time
tHSDI
SDI [1 : 0]
2.0
—
—
ns
Serial output clock cycle
tCYCSCO
SCO [1 : 0]
SYO signal setup time
tSSYO
SYO [1 : 0]
2.2
—
—
ns
SYO signal hold time
tHSYO
SYO [1 : 0]
1.4
—
—
ns
SDO signal output delay
tDSDO
SDO[1 : 0]
—
—
5.6
ns
ns
500
—
ns
ns
ns
500
MB86330
(3) SMODE
SMDT
MSB
14
13
2
1
LSB
SMCK
SMEN
tCYCSMCK
SMCK
tSSMENlh
tHSMENlh
tSSMENhl
tHSMENhl
SMEN
Valid
Valid
SMDT
Valid
tSSMDT
tHSMDT
(VDD = 3.0 V to 3.6 V, TA = 0°C to +40°C, output pin load = 50 pF)
Parameter
Symbol
Pin name
Value
Min.
Typ.
Max.
Unit
Serial input clock cycle
tCYCSMCK
SMCK
SMDT SMCK setup time
tSSMDT
SMDT
0.8
—
—
ns
SMDT SMCK hold time
tHSMDT
SMDT
1.8
—
—
ns
SMEN SMCK setup time
tSSMENhl
SMEN
0.6
—
—
ns
SMEN SMCK hold time
tHSMENhl
SMEN
0.6
—
—
ns
SMEN SMCK setup time
tSSMENlh
SMEN
0.6
—
—
ns
SMEN SMCK hold time
tHSMENlh
SMEN
0.8
—
—
ns
ns
500
41
MB86330
(4) PLL and Others
SCKOUT
tDST
tHST
Valid
ST [2 : 0]
tDPAGE
tHPAGE
Valid
PAGE [1 : 0]
SCZC
tDEASC
tHEASC
tDSCKSC
tHSCKSC
EA [15 : 0]
SCKOUT
(VDD = 3.0 V to 3.6 V, TA = 0°C to +40°C)
Parameter
42
Symbol
Pin name
Value
Min.
Typ.
Max.
Unit
ST output delay
tDST
ST [2 : 0]
—
—
3.3
ns
ST hold time
tHST
ST [2 : 0]
0.8
—
—
ns
PAGE output delay
tDPAGE
PAGE [1 : 0]
—
—
5.3
ns
PAGE hold time
tHPAGE
PAGE [1 : 0]
1.3
—
—
ns
EA output delay for SCZC
tDEASC
EA [15 : 0]
—
—
5.4
ns
EA hold time for SCZC
tHEASC
EA [15 : 0]
2.2
—
—
ns
EA output delay for SCZC
tDSCKSC
SCKOUT
—
—
3.2
ns
EA hold time for SCZC
tHSCKSC
SCKOUT
1.1
—
—
ns
MB86330
(5) MCLK, XRST
MCLK
tCYC
XRST
tPWRST
(VDD = 3.0 V to 3.6 V, TA = 0°C to +40°C)
Parameter
Symbol
Pin name
Value
Min.
Typ.
Max.
Unit
Remarks
MCLK cycle
(when PLL is used)
fCYC
MCLK
20
—
25
MHz
*
MCLK cycle
(when PLL is not used)
fCYC
MCLK
—
—
160
MHz
*
* : Input the MCLK cycle value so that the MCLK duty-cycle becomes 50% ±5%.
Parameter
XRST “L” pulse width
Symbol
tPWRST
Pin name
XRST
Value
Min.
Typ.
Max.
10tCYC*
—
—
Unit
ns
* : tCYC = 1/fCYC
43
MB86330
■ ORDERING INFORMATION
Part number
MB86330CR-ES
44
Package
256-pin Ceramic PGA
(PGA-256C-A03)
Remarks
MB86330
■ PACKAGE DIMENSIONS
256-pin Ceramic PGA
(PGA-256C-A03)
0.13
0.46 +– 0.05
DIA
.005
(.018 +– .002
)
2.54 (.100)
MAX
C1.02 (.040) TYP
(4PLCS)
1.27 (.050) DIA TYP
(4PLCS)
45.72 (1.800)
REF
INDEX AREA
50.04 ± 0.51 SQ
(1.970 ± .020)
 1994 FUJITSU LIMITED R256003SC-3-2
1.27
2.54 ± 0.25
+ 0.25
– 0.76
(.100 ± .010)
+ .010
– .030
(.050
)
6.35 (.250)
MAX
3.30
+ 0.51
– 0.25
(.130
+ .020
– .010
EXTRA
INDEX PIN
)
Dimensions in mm (inches)
45
MB86330
FUJITSU LIMITED
For further information please contact:
Japan
FUJITSU LIMITED
Corporate Global Business Support Division
Electronic Devices
KAWASAKI PLANT, 4-1-1, Kamikodanaka
Nakahara-ku, Kawasaki-shi
Kanagawa 211-8588, Japan
Tel: 81(44) 754-3763
Fax: 81(44) 754-3329
http://www.fujitsu.co.jp/
North and South America
FUJITSU MICROELECTRONICS, INC.
Semiconductor Division
3545 North First Street
San Jose, CA 95134-1804, USA
Tel: (408) 922-9000
Fax: (408) 922-9179
Customer Response Center
Mon. - Fri.: 7 am - 5 pm (PST)
Tel: (800) 866-8608
Fax: (408) 922-9179
http://www.fujitsumicro.com/
Europe
FUJITSU MIKROELEKTRONIK GmbH
Am Siebenstein 6-10
D-63303 Dreieich-Buchschlag
Germany
Tel: (06103) 690-0
Fax: (06103) 690-122
http://www.fujitsu-ede.com/
Asia Pacific
FUJITSU MICROELECTRONICS ASIA PTE LTD
#05-08, 151 Lorong Chuan
New Tech Park
Singapore 556741
Tel: (65) 281-0770
Fax: (65) 281-0220
http://www.fmap.com.sg/
F9810
 FUJITSU LIMITED Printed in Japan
46
All Rights Reserved.
The contents of this document are subject to change without
notice. Customers are advised to consult with FUJITSU sales
representatives before ordering.
The information and circuit diagrams in this document are
presented as examples of semiconductor device applications,
and are not intended to be incorporated in devices for actual use.
Also, FUJITSU is unable to assume responsibility for
infringement of any patent rights or other rights of third parties
arising from the use of this information or circuit diagrams.
FUJITSU semiconductor devices are intended for use in
standard applications (computers, office automation and other
office equipment, industrial, communications, and measurement
equipment, personal or household devices, etc.).
CAUTION:
Customers considering the use of our products in special
applications where failure or abnormal operation may directly
affect human lives or cause physical injury or property damage,
or where extremely high levels of reliability are demanded (such
as aerospace systems, atomic energy controls, sea floor
repeaters, vehicle operating controls, medical devices for life
support, etc.) are requested to consult with FUJITSU sales
representatives before such use. The company will not be
responsible for damages arising from such use without prior
approval.
Any semiconductor devices have an inherent chance of
failure. You must protect against injury, damage or loss from
such failures by incorporating safety design measures into your
facility and equipment such as redundancy, fire protection, and
prevention of over-current levels and other abnormal operating
conditions.
If any products described in this document represent goods or
technologies subject to certain restrictions on export under the
Foreign Exchange and Foreign Trade Law of Japan, the prior
authorization by Japanese government will be required for
export of those products from Japan.